For companies that are looking to hire a CRO for the first time?

1. What are the core skills to look for?

2. Is it advisable to hire an experienced CRO -- or grow their own?

3. Where do you *find* talented CRO individuals?

@dharmesh 

1. CORE SKILLS

1.  Great optimizers are polymaths

Optimizers have to be multi-disciplinary, and very good at at least 2-3 disciplines (and good enough at others). Optimizers don’t have the luxury of just being copywriting wizards or analytics ninjas – have to be both and then some. They need to be curious and always learning more stuff.

2. They have critical thinking, a knack for asking the right questions

The world is not simple, yet it’s natural for people to oversimplify everything. Optimizers have to be better than that. There is no “people always prefer” or “who would ever”. A good optimizer knows that the world is not the way she sees the world, and that .

3. They (try to) understand human behavior

Doesn’t matter if it’s B2B or B2C, ecommerce or SaaS, we’re deadling with (irrational, emotional) human beings.

Humans don’t come with an operating manual, but we do have some scientific research and frameworks we can rely on to help us in our optimization work.

4.  They’re a friend of numbers

A lot of people think that numbers are not sexy, and are even afraid of analytics. But you don’t have to be a total math nerd in order to be a friend of numbers. A good optimizer is not afraid of numbers, data or analytics.

5. They have persuasive copywriting skills

Are you we-we’ing all over your copy? Or are you able to write copy that articulates value in a user-centric way while making the text interesting to read?

Good copywriting is very important for high conversions, and an optimizer needs to know her way around. 

6. They are good with people

While optimizers sound like nerds (which they kind of are), I’ve never actually met a great optimizer who was shy, timid and full-on introvert, quietly chipping away at the corner of an office.

#7 They know good design and user experience

A critical skill an optimizer has is the ability to tell good design apart from bad design, and be able to articulate the difference. Anyone can say “this sucks”. What specifically? What would be better and why? What does the data say? How are users currently using it? What kind of problems do they run into? Optimizers need to be able to dissect, analyze and test designs.

#8 They have (at least basic) coding skills

Optimization involves running tests. Setting up A/B tests that are involve more than just changing button copy, need basic coding skills – javascript/jquery. Optimizers need to have a level of command of html/css, and know their way around browser consoles to troubleshoot stuff.

2. HIRE OR GROW?

CRO is no different from anything else. Growing one takes YEARS - if they have the right mentor. If you want immediate impact, you hire somebody who know what they're doing.

3. WHERE TO FIND?

Social media, blogs :) A lot of good CRO people come from analytics or UX. 

@peeplaja question owned!

@oligardner 10 points for @peeplaja.

@peeplaja I recognize this answer as a blog from http://conversionxl.com. <3 Solid write-up.

@peeplaja Nailed it!

@dharmesh

1. Curiosity. Willingness to be proven wrong, Full-stack analytics knowledge.
2. I’d hire an experienced CRO to hire and train others. Maybe even a consultant if you can’t find one. The most important parts of CRO can probably be taught in a day (rules not to break). It’s all about how to create a valuable test hypothesis, how to interpret data and how to know when you should and shouldn’t be stopping tests. How you go about doing that is difficult. But knowing the *true* fundamentals and where 99% of people go wrong, is invaluable. Otherwise you’ll be 12 months in before you suddenly *get* testing.
   
   
   The number of CROs available however, is so small that it’s very difficult to find talent, but it’s also harder to try and start from scratch. If you were going to hire someone less experienced, I’d start with someone with an analytical mindset.
3. A talented CRO is a bit of a unicorn. Primarily because most of the people who would be good at it don’t self-identify with it yet, don’t put it as a skill or title in LinkedIn, making it impossible to find them. I believe we need more exploration of the skill set needed to be good at it, so we can start inserting CRO as a disciplinary choice in a career path.

@dharmesh Really great question, Dharmesh!

1. I'm speaking from the perspective of writing copy for CRO, but I think these skills apply to anyone in the field:

+ Ability to create and follow processes. You need to be systematic in how you think about, assemble and test the elements of CRO, or you'll wind up chasing rabbits down worthless holes.

+ Research skills. A huge part of setting up tests that win or baseline pages that convert is the ability to step into the shoes of the people buying from you. To do that, you need to be able to research and analyze.

+ Analytical thinking. Like Oli said, you need to be able to deal with numbers and interpret them. I'm not necessarily "full stack" in this regard, but I know enough to know what's happening and ascertain why that is.

Beyond that, I think creativity and a lack of ego are huge. There can be no egos in CRO - anything can change, and you can't get married to one element or angle, because it could be the wrong one.

2. I'd definitely hire an experienced CRO - otherwise, it's the blind leading the blind. An experienced CRO can teach you, train you, show you things you'd never thought of and most importantly, challenge your thinking and assumptions. Learning by trial and error and mistakes is huge and ongoing even for pros, but I think you need an experienced pro to help you start moving the needle.

3. Quietly doing the work in the back of digital agencies. Your analytics folks are there, crunching numbers - rarely the loud, boisterous marketing types at the front of things. From a copywriting standpoint, there are plenty of groups you can check out where they congregate - from the Cult of Copy on Facebook to The Pit here on Inbound. Beyond that, Oli is totally right - not many people self-identify with this yet, so you want to look for skills over titles.

@JoelKlettke lack of ego << nice one. It can be really hard to let something go when you've sweated over it (like the explainer video story I shared further down the page). If it loses, you have to have the courage to throw it away. The best way to do that is to celebrate the fact that you learned something from the failure.

@dharmesh Here are the top 10 traits to look for in a CRO specialist:

1. Be Empathetic

2. Have Insatiable Curiosity

3. Have An Understanding Of Buying Psychology 

4. Understand Usability

5. Have A Different Sense of Design

6. Have An Understanding of Good Classic Copy Writing

7. Be A Critical Thinker

8. Be Humble

9. You Gotta Love of Data

10. Have An Open Mind

You can read more on my blog:

http://creativethirst.com/blog/the-top-10-traits-a-great-conversion-rate-optimizer%E2%80%A8-must-have/ 

@dharmesh Problem solvers, hard working and willing to think out the box.

Go to conference focused conference and meet people who know their stuff.

Hi guys - thank you all for doing this, and thanks to the folks at Unbounce for organising CRO day!

I've a question for you all: do you remember your first *epic* win? What was it, what was the improvement and how did you learn to do it - assuming it wasn't just an accident! ;) ?

@edfryed Perhaps a little tangential, but my first and most epic win was bumping into a Grizzly Bear in 1996 in Glacier National Park. If that hadn’t happened I wouldn’t have moved to Canada to be a wildlife photographer, wouldn’t have survived the 2001 dot com crash with 3 of the 6 Unbounce co-founders, and wouldn’t be part of setting up and celebrating the biggest event in the history of Conversion Optimization. So, pretty much a complete accident in how I *optimized* my life.

@edfryed It came after the fabled time in my life known as "The Year of No Wins." I was leading product for an e-commerce company and was tasked with improving conversions through testing. That year, we spun up different tests in a row--one each month--on different areas of the site. Everything from small tweaks, to homepage redesign, to cart changes... and every. one. lost. It was incredibly disheartening for the team and frustrating, but we kept at it. And then, the 10th test (larger text in the top navigation) boosted conversion rate AND AOV by a significant amount. It paid multiples on the team's time, but it was no accident. It was absolutely a testament to our commitment to continuous testing and our understanding that if were disciplined, smart, and consistent--we could keep getting gains. It wasn't the most epic in terms of lift, but it certainly felt the most epic after the preceding months, and really informed my philosophy on testing since. And of course--we kept testing!

@edfryed Following in @Oligardner's footsteps my first epic win was jumping out of a plane almost 6 years ago and deciding to keep doing so for another 600 jumps. Through these years I've managed to optimize every jump I make to become better and more successful skydiver :)

In terms of tests, I think my most epic test was about 3 years ago when I was asked to optimizes sales for a company that sells car insurance online. Our mission was to optimize their landing page that required people to sign up and make enough impact so that they would purchase insurance at the end of the funnel. The biggest issue was that once we dived a little more into our research we realised that their main traffic source was worth nothing. Basically they had 300,000 visitors per month to their landing pages from redirected traffic. This meant that these people were looking for something completely different and was being redirected to their LP.

To make the story short, we created two landing pages, each completely different than the other in structure, strategy, colors and messaging and the result was 79.3% increase in signups and 107% increase in paying customers. This was epic in my books. They saw a $8,700 increase in revenue per month over 12 months of testing and more importantly those customers had higher return rate and were spending more time on the site than before.

@edfryed Hi Ed, my first "epic win" was turning a music blog into an online community into an in-person community that participates in organizing music festivals and also has an annual meetup. <3 It was a long-term win made up of incremental wins.

@edfryed The most impactful win for me wasn't the first big win we had at WiderFunnel. It was a surprising one that reinforced the importance of testing "left field" variations. 

@edfryed My first "epic win" came from the world of sales.

When I first started in sales (a very long time ago), I was given a script to follow. And I followed it. Sometimes I won the day. Most of the time I lost.

I noticed that the best performing salespeople around me were keeping notes that included much more data than I was. I quizzed them. I asked what they recorded. I started doing the same.

While not a web or email A/B split test, I did split test what I was saying, and noted the results. The biggest early win was in cold calling to get to the decision maker.

Instead of saying "who is the best person to speak to about...", I changed it to "who, other than you, is the best person to speak to about..." - a subtle change.

The result? I remember it being in the order of several thousand percent improvement. That little psychological trick - letting people know that I value them, but still want the referral - got me to the decision maker almost every time.

This is one of the reasons I like to see tools like ClearSlide around. Being able to A/B split test what salespeople say when giving presentations is huge for managers and business owners. Turns out, CRO isn't just about websites, emails and other 'digital marketing' channels. :)

@edfryed when I was in corporate, working for a popular diet and fitness dot com. We sold the atkins diet. We optimized the heck out of the custom checkout out page. This begin 2005, it took us 3 months to set up this split test and it ended with a 89% conversion improvement, then the real Dr. Atkins split his head on the ice and our conversion when though the roof with epic days for weeks. So timing and conversion testing go hand in hand. Its those experiences of epic impact you remember, not the daily grind. Great question. 

What are a few best practices (popular wins) for checkout funnel pages on ecommerce websites?

i.e. product page -> cart -> checkout

@oritzvielli "popular wins" is a difficult thing to quantify, because everything you see published is a result of survivor bias. 

If I say something like "add trust logos" or "reinforce free shipping policy" it's arbitrary and meaningless because it is completely out of context for your buying experience. 

The actual best practice you could execute is to get comfortable paying attention to your data & drop off rates in the cart, use heat maps and session tracking to identify behavior, watch for patterns, form hypothesis and test changes. 

If you want "best practices" you need actually practice something, not just sample tactics.

@tommyismyname What are your thoughts on heat maps in general? 

@NikkiElizDemere I love heat maps! I think they can provide a lot of insight, and especially show you where people just don't care about. 

@tommyismyname Good to know, I've read a lot of criticism about them. Have you written any counter-arguments?

@NikkiElizDemere Heat maps are helpful for finding the places on the page no one interacts. The two main problems I see with heat maps are:

1) They tend to show the same thing over and over again: visitors follow the F pattern. They scan the top of the page and then down the left hand side of the page.

2) They don't differentiate the behavior of people who buy and people who don't buy. They ought to connect the sessions of people who subsequently convert (or not) and allow you to see the difference between the two groups (or any other segments you care about like channel, device, etc).

@oritzvielli best practices are just a starting point, not an optimization strategy. The typical funnel I build and optimize includes 2 upsells and 1 downs after the initial purchase. Typically you want to offer an upsell that speeds up or makes it easier to achieve the results of the initial product on the up front sale.

@Bobbyhewitt Thanks Bobby for the recommendation. 

@oritzvielli The first thing you should keep in mind is that a "best practice" should be a starting point. There are so many factors that will cause variance in what you should most likely do - so you need to gather data to inform your decision. That being said - read this article for some guidance: http://conversionxl.com/how-to-design-an-ecommerce-checkout-flow-that-converts/

@oritzvielli As others have said, it's hard to say anything is truly a best practice.

A few big things I would think about:

- Is your purchase flow setup to accomodate towards your majority visitor type? (Ex: if your site sees 80% new visitors, are you setting up your shopping cart/login flow/check out for new users or return users?).

- How do you set cart items to be stored when a user is logged in or not? Have you tested these durations or rules?

@oritzvielli Make it very easy for the user to checkout and feel comfortable about using their credit card.

You should very easy for the user to checkout. The buttons and headlines should tell people what to do next. Never make assumptions that you know what the customer should do. Increase security and trust logos along with a BBB logo to make the user feel.

Offer guest checkout out and paypal. Use icons that people can easily understand. Don't use new age edsign/flat icons. Use generic 90-day and money back guarantee seals. Also show your phone number.

Use a cart abandonment app to automate the emails to people that do not checkout.

Add qualaroo to find out why people are converting and why people are not converting.

@oritzvielli 

As others have already pointed out, any best practices or likely wins are still something you need to test.

But send good cart abandonment emails (and A/B test them), minimize distractions during the checkout process, make it clear to the customer what's happening in the process and when, try to avoid anything that makes it look like you're springing surprise fees or clever accounting on the customer, and reinforce why they're buying from you (painless pre-paid returns process, best in class quality, social proof of satisfied customers, etc. etc. – test what works best for your customers).

Well, this is only going to take a few days to read...

@EdLeake It will be worth it. :) 

@NikkiElizDemere absolutely - just never enough hours in the day. My desk is full of great looking books, all page turners I'm sure, and yet not one read as yet. :-<

@EdLeake Hahaha, yes, I have so many books to read too!

@EdLeake As comedian Michael McIntyre said:

"A good book is called a page turner. Surely that is the minimum you expect from any book." :)

@TheRealSJR Stewart - I have an addiction I'm not proud of; a book case full of great looking books but all of them virgin soil... and yet, I keep buying more books!

I need help.

Super excited for this AMA, can't think of a better group to answer these questions.

Correlation does not imply causation. We have all seen hundreds of CRO examples and sales increase by 500% in 1 month of A/B Testing.

In addition to forming a hypothesis, what are some steps to take to ensure your results are statistically significant and not seasonal or attributed to outlying factors? 

@LuizCent if your question is about when to call a test so you can be confident about the results, then you need to account for the following factors.

First of all: Don’t stop the test just when you reach 95% confidence (or higher) . This is very important. 

1. Sample size. By running tests you include a sample of visitors in an experiment. You need to make sure that the sample is representative of your overall, regular traffic. So that the sample would behave just as your real buyers behave.

You need to pre-determine the needed sample size. There are many great tools out there for that, like this one. Or here’s how you would do it with Evan Miller’s tool: http://www.evanmiller.org/ab-testing/sample-size.html

2. Test duration

Your test should run for 1 or better yet 2 business cycles, so it includes everything that going on:

• every day of the week (and tested one week at a time as your daily traffic can vary a lot), 
• various different traffic sources (unless you want to personalize the experience for a dedicated source), 
• your blog post and newsletter publishing schedule, 
• people who visited your site, thought about it, and then came back 10 days later to buy it, 
• any external event that might affect purchasing (e.g. pay day) 
  

and so on.

Typically your test should run 2, 3 or 4 full weeks. I myself run most tests for 4 full weeks as I see too often that a winner a 2 week mark will end up "no difference" by 4 weeks (imaginary lift).

3. Only once your first two criteria are met, you look at the statistical confidence.

Detailed explanation here: http://conversionxl.com/stopping-ab-tests-how-many-conversions-do-i-need/

@peeplaja @LuizCent For transactional sites (e-commerce, marketplaces) you also need to understand the distribution of your order values and understand if they are normally distributed or follow a power law distribution, so you know if the test you're using to calculate statistical significance on revenue per visitor is valid. For sites whose orders follow a power law distribution, you typically need a larger sample size--this may be difficult to get for smaller sites, but e-commerce sites of significant volume should definitely understand and track this. 

There's a good write up on quora: http://www.quora.com/In-an-A-B-test-how-do-I-calculate-statistical-confidence-for-revenue-per-visitor

I agree with @peeplaja. Multiple tests I've run show a big lift in the first couple of days and even might be considered statistically relevant by a software like VWO only to disappoint me by dropping down to earth a week later. It's definitely worth testing for at least two weeks and ideally as many as four at a minimum to make sure you take different variances in the business cycle into account and to ensure you're not implementing an imaginary lift.

@LuizCent You want to strike a balance between obtaining valid results and quickly iterating. What we do at WiderFunnel is ensure we reach 90% statistical significance in any experiments before drawing any conclusions.

Then, after 6-12 months or testing, we re-test our latest winner against the original control to re-confirm that the results have stuck and weren't just due to random chance. We compare the result of that follow-up A/B test against the calculated cumulative conversion rate lift. In addition, we reconfirm using 2 duplicates of both the control and latest winner to measure margin of error as well. 

That's the gold standard validation for both validity and permanence of the result.

When crafting copy for buttons, what generally is best practice? Does 'Click Here to ... ' hold any relevance to the user to respond to it?

@alisonmf I just listened to a Social Media Examiner podcast and Joanna Wiebe (who is on this panel) says “Take the phrase “I want to” and whatever follows that becomes your button copy".  It was a great podcast on conversion copy and split testing, here's the link- http://www.socialmediaexaminer.com/split-testing-with-joanna-wiebe/

@MariaSallis I love this approach because it helps you understand what it is that people want at that stage. Do they want to "Buy Now," "Learn More," or "Download My Free Guide." Putting "I want to" in front helps to determine what it is the visitor wants and how to turn that into button-copy text.

@JosephPutnam yeah, i'd push that just a bit further to get to what they really want. So nobody wants to "learn more" or "download my free guide". They want the promise of doing that action; they want the value or the outcome on the other side of it. I don't take a negotiations course because I want to take a course (action); I take a negotiations course because I want to make more money for the same amount of work (value / outcome).

I've been breaking CTAs down into calls to action vs calls to value. Calls to action speak to the exact thing you're doing: Download free CRO glossary. Calls to value speak to the reason you're doing it: Speak confidently about CRO. Test to see which works better at each point in the funnel. 

The deeper into the funnel you get - and certainly in-app - it's often wise to pull back on the call to value and focus more on the call to action. For example, when you're confirming your order in a clothing retailer's checkout process, "Place Order" (call to action) is likely a much better button than "Dress for success" (call to value). But early in your prospect's experience with you, calls to value tend to perform best. 

(And thanks @mariasallis!!)

@copyhackers Very helpful, as usual. Thanks, Joanna.

I totally agree with @copyhackers. 

I personally hate "click here" prefixes, and so do search engines. (It hurts SEO.) It begs the question, does your CTA not already look like a clickable button??

For both headlines and CTAs, I use a variation of the fore mentioned formula:

"I'd like to..." [WHAT - Specific Action]

"Because I want to..." [WHY - Specific Value] 

It's important to pair WHAT and WHY together. Sometimes this can be accomplished in one line. Two lines (headline+subhead, 2-line CTA, CTA+booster) are more often needed though. This shouldn't be feared if it provides more clarity and value. 

Example:

"Get the Replacement Window Buying Guide"

"Avoid Common Window Buying Mistakes"

I write CTA ideas like this first for every campaign. I flip around what/why to why/what as needed and create mobile-friendly CTA versions that are ideally 4 words or less. I take the longer versions and formulate attention getter headlines.  

This formula helps keep the page focused on one purpose by closely aligning headline and CTA as well.

Test Ideas for "click here":

• Is it a dominance factor? Test against larger button or more words in the CTA.
• Does it provide a "directional cue" effect? (e.g. Look left -->) Test against icon arrow substitutes.  

@alisonmf Hey Alison!

So - a lot of this has to do with context. Best practices are dangerous because as someone like @peeplaja will be quick to point out, most of the time people aren't even sure why they do what they do. 

But some things I definitely keep in mind:

1. No "mystery doors"

It has to be abundantly clear what's going to happen when someone clicks that button. What are they going to get? Are they scheduling a demo, or signing up for that demo right then and there? You can't afford to leave people wondering, or they won't click out of nervousness.

2. Written as an answer to "I want to..." is a great place to begin.

As I learned from @copyhackers, many strong call to action buttons use text that can finish the customer saying, "I want to..."  - You want your button text to be an action a customer already wants to take (not "submit"); a call to value they want to receive. 

3. Commitment level and context count.

This is where empathy comes in. For example, when dealing with a high-rate, elite consultant, "Schedule my appointment" works because there's a sense of exclusivity and prestige implied. But for something lower-key, asking someone to schedule their appointment right then and there might actually turn them off more than "contact us", even if they're in a mindset of wanting to schedule. It's weird, but true. The offer has to match the commitment level the customer is ready to make.

I'm sure others will have more tips for you :)

@alisonmf A simple formula to follow for button CTA's is "Action Verb" + "Benefit"

@alisonmf I've seen case studies where including the word click increased... clicks. But like every case study, it isn't a panacea and should be taken with a grain of salt. You can't apply case study learnings, only use them to serve as inspiration and to be used to generate your own related hypothesis.

@oligardner ^^^ Totally agree with that. The problem with sharing CRO wins is that the best they are is inspiration to institute in your own tests. You can't simply expect that every audience will react the same way. It comes down to:

• Being curious
• Empathizing with people's behavior
• Narrowing down your hypothesis into testable variables
• Testing them and choosing the winner

That being said some inspiration I would take away on call to actions are:

• Try to make your buttons benefit-oriented (what are they going to get by clicking? "subscribe to get the ebook" etc.)
• Try to make the call to actions sound actionable and like they are achieving something ("click to subscribe" = better than "click here") 

But again, use this as inspiration, you need to test everything at the end of the day.

@alisonmf Lately, we’ve been taking a different approach to button copy testing at WiderFunnel.

It’s great to start with thinking about what people want, whether it’s the specific offer or the value. But, what if the button test could tell you about what drives that customer segment to act? That’s where button testing can start to tell you about your prospect triggers. 

@alisonmf I have always tested well with "Continue"

Like "Start Free - Continue"

What are some lesser known tools or resources that aid CRO that you absolutely love (or would love to try out) and why? (i.e. platforms, SaaS tools, browser plugins, testing frameworks, process/layout templates, etc. Anything CRO-related -- analytics, UX, usability, design, user testing, etc.

@aschottmuller The single most impactful tool for CRO is curiosity.

@oligardner It's more than just curiosity, it's empathy + curiosity + psychology + data + .... so the single most important tool is "you"

@Bobbyhewitt I consider my answer optimized.

@aschottmuller I love Heap Analytics - great tool to build ANY funnel on the fly based on user behavior. Auto-event tracking without any setup.

@peeplaja Cool. Hadn't heard of Heap Analytics. I'll check it out. Thx, Peep!

@aschottmuller inspectlet, hot jar, quillengage, shopperapproved, yotpo, cyfe, unbounce, leadpages, adespresso, socialleadfreak,optimizepress,vwo

@alexdesigns quillengage, shopperapproved, cyfe, and yotpo are new to me. Thx, Alex!

@aschottmuller I want to understand MixPanel better, right down to the API endpoints.

@NikkiElizDemere Hi Nikki! I only have one client on MixPanel, so I'm not that savvy at it. I recall the integrated email lacking from an analytics perspective. I do like the funnel visualization though. What is your specific data reporting or visualization goal?

@aschottmuller Google Tag Manager, HotJar, UserTesting.com, GrowthCanvas (http://beta.canvas.growthhackers.com)

My question is about the word Amazing- my husband (who is not in marketing) says it is overused. What do you think about using it in copy? 

@MaryGreenIM amazing question!

@MaryGreenIM Superlatives tend to lose against specifics ("amazing pizza" vs "stone-oven baked pizza by an Italian master chef", "fastest pizza delivery" vs "delivery in 15 minutes") 9 times out of 10

Instead of superlatives, offer lots of detail and specifics. 

@MaryGreenIM I think opinions are irrelevant. You should A/B(n) split test the word amazing against awesome, fascinating, incredible, marvelous, prodigious, shocking, stunning, surprising, unbelievable, and wonderful, and then let the data decide...

@TheRealSJR it's rather silly advice to waste precious testing time to test a single word against all these variations

@TheRealSJR opinions aren't irrelevant, everything we test or try is based on an opinion of what might relate to our audience. Opinions without testing are risky, that's a better way to phrase it. I would test your call to actions or headlines that have "amazing" in them against other headlines or call to actions that choose less-used words that might stand out. That being said the very fact that so many people say "amazing" might be a reason it would work really well. The proof is in the pudding, not in the hypothesis. Test all the things!

@MaryGreenIM Like everything #CRO, You Should Test That! :)

@MaryGreenIM Saying words are overused, is overused. 

@MaryGreenIM 

It's all about context, to be honest - and unlike what others have suggested, I'd rarely run a full A/B test over a single word. That seems a bit extreme to me, unless that word is in a primary headline or something.

The thing is, we all want to turn our noses up at some words, because they're misused and abused. "Revolutionary", for example, is a word that has lost much of its meaning because everything from pens to ankle socks have been called "revolutionary",

But my tip is simply this: Instead of obsessing over individual words, think about your context and slash hyperbole wherever it stands. If the claims you are making are believable, hit on customer pain points and directly explain a benefit, then the verbiage you use to describe that benefit can be flexible, so long as it fits the context.

Because words like "amazing" still pique our interest. We all want "amazing results" when we're in a moment where we feel inadequate. We all want something "incredible" when we feel mediocre. Those words still carry power in context for the right customer.

Maybe a result really is AMAZING. Maybe your technology really IS revolutionary.

But you can't just TELL people that, you also have to lead them to believe it on their own. You have to write in such a way that people draw the conclusion that what you're offering is "amazing" on their own instead of beating them over the head with it.

@MaryGreenIM "Amazing" is an uninstantiated claim. If a guy walked up to you'r younger self at a bar and said "Hi, I'm Amazing." would you believe him? Any claim needs to be supported with proof. 

@MaryGreenIM Slightly less snarky answer...

I once got a press release for a video game that used the word 'epic' in the 350-word release over 40 times. One of my guys wrote an article about how this game was going to suck beyond belief, based on the fact that if they were trying to convince us it was this epic, it probably wasn't. Turns out he was right on the money - it sucked.

In this PR example at least, it is true that the more they try to tell you it is going to be amazing, the less likely it will be. :)

@MaryGreenIM 

It's probably overused, and I would be surprised to see it win against customer-specific language.

But that's something you can test!

Also, what other buzzwords you believe are currently overused?

@jdquey i personally loathe "rockstar". i've used it. i'm embarrassed about it. but, omg, when i see it on a page today, i instantly get that feeling that an old person is trying to sound young. 

@jdquey Congratulations!

If you want your business Facebook posts to squeeze through the filters and rise to the top of your fan’s feeds, including the word Congratulations! can make Facebook’s algorithm think it’s connected to a life event such as a birth, birthday, wedding or anniversary, humanizing the event and giving it prominence.

Congratulations!

Funny? Absolutely. Does it work? Yup. For now. Should you do it? No! It’s a bs marketing tactic rooted in deception.  But very funny.

@oligardner Yeah, I heard that Mark Zuckerberg missed hearing the birth of his nephew (or something like that), which triggered the word "congratulations" to make its way through the filters.

A legitimate route is to find ways for your audience to say congrats, but again, just making sure not to abuse it.

@jdquey 

"Quality"

"Revolutionary"

"Epic"

"Unique"

"Seamless"

@jdquey Awesome. Horribly over-used!

@TheDaveCollins This is a good TED talk about the overuse of the word awesome.

@TheDaveCollins i KNEW i'd see you comment on this, dave!! :D

@TheDaveCollins 

But what if everything really IS awesome?! 😉

https://m.youtube.com/watch?v=StTqXEQ2l-Y

@aschottmuller Haha! I don't even know how to respond, other than awesome ;).

Does price of a product or service matter for conversion? Obviously different prices will appeal to a variety of people, but what about stuff like $29/month vs $30/month. I see different companies use different amounts; some will embrace the whole 'only 29.99' thing, while others are just plain and simple saying: 'it's 30 bucks'. Is there data supporting one over the other?

@alisonmf hi Alison, Great question. There are many articles on the subject of pricing as pricing has many psychological sides to it. In general there are many cognitive biases that have an impact on our behavior. Cognitive biases are tendencies of our brains to think in certain ways, sort of like “unconscious” triggers that make different connections in our brain to help us make decisions. The $29/month vs $30/month is a way to capitalize on a specific cognitive bias. This specific tctc has been researched, tested and proven to work, in fact both researchers at MIT and the Quantitative Marketing and Economics journal found the number 9 worked better than "sales" tactic (when you show the before and after price). There are a few good articles you can follow on the rt of pricing that I recommend:

1. Start by watching Dan Ariely's video about the decoy effect (it's another really cool example of cognitive biases and how they work)
2. 10 Principles of Effective Pricing Pages by @peeplaja
3. 8 Psychological triggers to optimize your pricing page
4. 26 Pricing Page Examples and Best Practices

@TaliaGw's response is awesome. 

There are case studies out there, but it's most important for YOU to test your audience. Lots of my clients lock down their pricing, and won't budge on testing it.  Sometimes their pricing model (dealer/retailer/regional networks) are too complex already. Still, pricing is ABSOLUTELY something that you should test and learn for your audience. It can make a gigantic impact on revenue when you understand what does and doesn't work, and the insights can be scaled across the organization.

@alisonmf So @taliagw gave you a killer answer here, but I want to add one more thing:

Different markets are... well, different. For example, if you're marketing an energy supplement to college kids, you need to keep their economic status in mind. No matter how great your product, a price point about $19.95 probably won't fly (hat tip to @explorionary for that one).

Yes, there are some pricing heuristics and cognitive biases you can exploit, but ain't none of 'em gonna work if you've priced yourself poorly for the market.

@JoelKlettke Completely agree. Best practices are great but they won't necessarily work for you. Test everything. That's what CRO's for :)

@alisonmf There was also the case where JCPenny tried doing their "no nonsense, everyday" pricing, and found that when they stop artificially inflating their prices, just to have things go on sale, their sales and stock took a plummet. 

As with most of this stuff, it completely depends on the market. 

@alisonmf Marks & Spencer have done the rounded number thing for decades. No pennies allowed. Did they test it? I can't find any evidence, and am more inclined to lean towards it being a branding thing. It is interesting though, especially considering virtually no one else in retail does it.

Canada has removed pennies entirely from the country's currency, so once that ripples through commerce, there won't be any .99 allowed anyway, making the psychological impact null and void (on the penny level).

@alisonmf What other people have tested has NO IMPACT on how 29.99 vs 30 would play out on your site. Audiences and markets are different. While there are numerous studies out there about the magic 9, that doesn't mean it's gonna work best for your particular context. 

Only testing will give you the answer. 

Does button design matter?

Do buttons with slight bevels/embossing/shadows perform better than buttons with a flat, square look? I've also read that size matters although in some A/B tests I've seen if the button is too large, it doesn't perform as well either. Then there is the whole red vs green vs blue argument, however, for color deficient users, what is the best button color for converting (do we even have this data)?

Is all this true and could you comment about designing buttons for converting?

@alisonmf It's fairly unlikely that one little change to button design will have an impact. However, if you apply many principles at once you'll have a better chance. I see this as more of a legitimate use of a best practice as opposed to an optimization tactic.

Here are some of the characteristics of a usable button:

• size
• shape - circles don't look like buttons
• colour - not as important as contrast. Persuasion through colour is only really possible when considering the entire page.
• contrast - does it stand out from the dominant hue of the page?
• depth - flat or 3D
• position
• location - if it's part of a form we know it's a button (more often than not), if it's a standalone click-through button it has a bigger job to do
• proximity to other elements - what you write close to your button can impact its use
• surrounding element design - directional cues pointing at a button can make a button appear more buttony
• clickability - ghost buttons are a massive design trend, and the buttons simply don't look clickable in the truest sense. Will we train ourselves so this doesn't matter? Perhaps.
• interaction: rollover state - is there a strong contrast when you mouseover?
• interaction: rollover cursor change - does the cursor change to provide feedback based on link conventions?

@oligardner In other words, the button needs to look like a button ;-)

@tommyismyname if a button lives on a page, but nobody clicks it, does it even exist?

@tommyismyname 

Clear affordances & good relative visibilty. Button design is a UX question, more often than not.

@alisonmf

@oligardner Great list!

Ghost buttons drive me crazy. It goes against usability. The concept is a designer's fantasy trend that should die. The only time I find this tactic useful is when a client insists in having two CTAs on the page, and I basically want one to disappear. Ghosted buttons have ghost conversions. 

I've seen button design become increasingly more important these days. Since too many sites don't reserve a complementary color for their primary CTA, users have growing uncertainty for what's actually clickable. This is especially problematic on mobile devices, where less white space real estate width is available and hover effects aren't visible. (I was recently comparing click heatmaps for a very similar yearly campaign, and I was very surprised by the click behavior changes.) 

Tips: (in addition to Oli's list)

1. Apply the Clockwork Conversion Color Model/Test to your site to ensure the primary CTA stands out.

2. Take effort to make your button look like a 3D button.  (3-5px rounded edge, gradient background,etc.) << Do this with CSS3, NOT images. VIP for mobile performance.

@oligardner Awesome, awesome, awesome! To expand on the 'proximity' thing Oli shared, which I think is particularly important - the content RIGHT ABOVE a button is so critical.

Usually, I use that space to do two things:

1. Eliminate any final friction (if someone is worried about risk, I use the copy nearby to quash that fear)

2. Direct people to take action (plainly tell people what you want them to do next).

If your button is buried next to a section of content that doesn't naturally lead someone to think, "hey, I should do something now!", then it sort of comes as a speed bump instead of a natural progression.

@alisonmf At WiderFunnel, we've had experiments where a button color increased conversion rates by over 150%. (Hint: because it created better Relevance and continued the scent trail.)

I’ve also seen experiments where size mattered.

Copy absolutely matters.

It all matters.It really depends on what the real conversion barrier is. 

But, are there are bigger fish to fry? For example, what’s your value proposition doing for you? 

Rather than seeking tactical tips, seek frameworks to help you discover your prospects' conversion barriers and solve them with relevant hypotheses. Ultimately, using framework thinking is the most important tip. 

Some of the frameworks we've developed at WiderFunnel answer questions like: 

• Where should I test? Answer it with the PIE Framework
• What should I test? Find out using the LIFT Model
• What Value Proposition should I test? Use the PODs, POPs and POIs framework.

@alisonmf Everything you've read about button design is true, and false, and somewhere in between.

If you truly believe that the best hypothesis and test you can come up with - the one that will deliver a 200% increase on conversions - is to change the button, then you should run A/B or multivariate tests against all of those options to see what works for your audience.

The fact is, different audiences relate to different designs, language, reading levels, colors, and more. Averages across industries won't help you here.

@alisonmf So, I want to breach this question carefully in saying that I believe YES, button design matters, but NO, there's not one design or layout or format that works best for everyone.

I will confidently say two things:

1. People should know, without question, it's a button.

2. People should be able to see it without squinting. It should neither be too small so that it's ignored, or so large it looks like part of the page design. And it can't blend in - it's not so much about 'which color is best' (there isn't one) and more about contrast (does it stand out against your other color pallet?)

Beyond that, size, embossing, all that nonsense... test it. There's no hard and fast rule here.

@alisonmf Button tests are the worst kind of test possible. There are always BIGGER gains you can get by testing more meaningful chnages. Here are some more thoughts on this: http://creativethirst.com/blog/please-dont-test-this/ 

@alisonmf Button design itself matters little, it's all about visual hierarchy - does the button stand out in relation to other design elements? is it big enough? Does it have unique, contrasting color?

Whether its round, rectangle etc does not fundamentally change user behavior, will have marginal impact on anything - so don't waste too much time thinking about this. Fix real problems. 

@alisonmf I have tested rounded buttons on banners & they seen to get more clicks than square buttons. The color is only important to show that is contrasts the complementary colors.

What are the best practices for Local Businesses landing pages?..Let's say for professional services like Real Estate brokers, lawyers and for clinics & spas

@saxestmedia Awesome question!

I'll answer this in a couple ways.

1. The copy on this page actually needs to have value (I wrote about this here: http://businesscasualcopywriting.com/make-local-landing-pages-100-less-awful/) - a big hindrance on conversion rates and SEO alike is content that reads like generic fluff for the sake of targeting phrases.

2. From a conversion standpoint, a local landing page isn't that much different from your typical landing page. You still need to convince a person to take action - but the context might be different depending on your service.

For some businesses with urgency baked in (restaurants, plumbers, etc.), you want to make sure that your phone number is front and center, because so much of local traffic is done on mobile where the person doesn't want to read a sermon and just wants to pick up and call.

The other thing I'd say is that it's just as much about your scheduling/sign up process (whether it's a form submission or calendar-based system) - if I encounter a single annoying bump in trying to contact you, I'm definitely leaving on a mobile device and probably leaving on a desktop. You need to make converting totally frictionless and avoid bouncing me from domain to domain or from trusted-looking page to random-looking scheduling software.

3. It's important everywhere, but if you're a professional service whose clientele care about a personal touch, a human element to the page (REAL photos of you/your local team) can play a role in building trust and credibility. Worth testing, but not a hard and fast rule.

Hope that helps.

@JoelKlettke I understand your point. Thanks for your response. As part of the work I am doing for a local (Dubai) real estate advisory firm (I HATE the word broker) I am helping them get more enquires from overseas as one of the target markets. For that I am doing a lot... a LOT ...of landing pages that are not typical listings as people overseas expect more info as they're not even familiar with the address/location value. 

And I am doing other pages to help their target audience learn more about the product type, returns, views, legal framework, guarantees, paperwork, etc. etc...with out any sales attached to that. Just to get them closer to get in touch. 

What I am struggling a bit with are the irresistible offers that each landing page must have. Free Consultation doesn't cut it for international buyers :)

Blasko

@saxestmedia Well, in your context, think of the customer.

What don't they know?

What are they nervous about? 

What can you show them to make them understand?

Free consultation can't be done in person, but a CTA to talk to a local expert who can answer their questions and give them clear information might be attractive. It's more about your wording, perhaps.

If there are questions you're repeatedly asked or information people repeatedly seek, you might consider doing a downloadable asset (NOT GATED) and then remarketing to those people who download with a follow-up to talk to someone for help.

@JoelKlettke ....mmmhhmmmm...speak to a local expert sounds really good.....damn, I didn't think of it before. Thanks

@saxestmedia Words matter :) 

That's why I love what I do so much.

@JoelKlettke you are rights about words. Just yesterday I heard Dean Jackson say that the best way to do CRO (he doesn't quire call it that way) is to ask ourselves: "what would be their dream come true?"....I must admit that once I heard it said that way it made a looooot of sense to me.

@saxestmedia

You need to find the emotional hook. Travelers are naturally curious, so feed them  

1. GEO IP Content. The different locations sound like a great opportunity to leverage geoIP data to dynamically personalize the content. How does the overseas customer's weather/climate and regional dialect differ? How might that impact what they would say, wear (clothing), drive, spend in Dubai? Highlight similarities and differences. << Test what factors the customer desires to see change in or not. 

2. Interactive Quizzes. Try an "Is this destination right for me?" survey... that ends with a lead form to get a complete report. This can be especially successful with tablet and mobile users that are drawn to interaction. The gamified approach works great for younger demographics and anyone high in the decision-making funnel as well.

@aschottmuller 

I might need to go back to Uni to get all you have shared :)

Seriously though, I think the GEO IP is a real great idea. It will save me time from doing many different pages - as I am targeting the entire Sub Sahara Africa - and yet be 100% relevant to them and their reasons for investing in Dubai.

The Quiz idea is beyond brilliant. I am afraid I'll have to use the word Genius. It fits perfectly with what I am doing. And it can be pushed to even the type of investment: "Is Villas the right type of investment for me?"

You gave me two solid suggestions I'll surely use.

Thanks again

@saxestmedia Happy to help! Let me know how it goes. It's nice to see these things come to fruition. =)

@saxestmedia 

Make it clear that it's local, where it's local to, what services you do or don't provide, what specifically is involved in giving you money to get those services, and what I can expect to happen if I decide to give you money.

In general, know who your customer is and what role their visit to your site plays in helping you do business with them. The easiest way to do this is to talk to 5-10 customers and simply ask them. Talking to customers about how specific aspects of your business are or aren't helping them achieve success with whatever it is that convinces them to give you money is a bit of a panacea.

It's also not that hard to get your core customers to open up, as long as you approach it in a non-salesy way and are genuinely interested in listening to them instead of only pushing marketing on them. People love talking about themselves & what affects them (and hopefully the reason they pay you money affects them significantly, or they won't be doing it for long).

@grayj_ 

Talking to clients is fun

You are right about that. I'll do it this week itself.

Thanks

@saxestmedia Best practices are just a starting point, not an optimization strategy. But in general for a local business you want to establish trust and there's no better way that to provide proof. Include a picture, show an address and a map of the real-estate or local business location, include testimonials, etc. I'm covering 30 point of proof later today. www.croday.com

@Bobbyhewitt I am looking forward to that. Thanks for the answer

@saxestmedia making the website and the about page personal. Tell your story to showing your empathy towards how to make the customers life better. People like to work with people they like so you want to make people know your authentic as quckly as possible.

@alexdesigns 

Back at work with my About page :)

Thanks

First some background for context:

@BryantJaquez 

You're talking about customer lifetime value (CLTV) optimization, instead of single-conversion optimization. 

Email marketing is an amazing tool if you use it right, so it makes a good default action to push for. These leads tend to start valuable and they are relatively easy to enrich over time. Of course, many companies do email very badly, so there's still the quality question – but, powerful tool.

You say that "buy-in to the lifestyle" is a major factor. For the identifiable customer segment which has collectively contributed the most profit to the company within a reasonable choice of time window to study, what was their journey in becoming a customer and becoming a high-value customer? Ideally you want to start by identifying these people & studying their history with you, then pick a few at random from the customer segment and actually interview them to get their side of the story.

Is this more a problem of acquiring more of the right leads (i.e. some leads will just never be high-CLTV customers), or are there initial actions or products which you can push which predictably pull people deeper into your ecosystem? It's probably a mix of both, but studying it will tell you more about which segments to market to and via what approaches.

A good conversion expert is an expert on your customers, and the only way to become an expert about your specific customers is by studying them. This involves a mix of quantitative data about their behaviors and qualitative data obtained by interviews and observation and (carefully) tools like surveys. The most powerful of these and the easiest to execute on well is interviews, as long as you focus on drawing them out and listening instead of you talking a lot and inadvertently pushing a viewpoint you want to hear on them.

The better you understand your customer, the better you can structure out a marketing flow which improves the rate at which you capture and convert leads, and the degree to which you capture the right leads to improve your company's outlook. Free users mostly just don't convert to high-tier users, you have to capture people who have high-tier need for your company & lower price sensitivity.

@BryantJaquez For this type of test you want to do a cohort analysis and look at different groups of visitors over time, as well as some sentiment analysis like more traditional branding studies with surveys. It's a big topic but I hope this gives you a starting point at lease.

I'm pretty new to conversion rate optimization, but I'd love to know more about their processes.

• When not to test?
• What data to share with clients? How often? How?
• How do you work with other marketing teams that are working towards the common goal? For example, you are focused on conversion optimization, but there is an inbound and outbound team as well making edits. Shared calendar?
• Tracking amount of conversions vs conversion rate?

Thanks!

Maria

@MariaSallis 

Don't test if the result won't matter.

Don't test if you know you can't get good data.

If you need to change many things this month because reasons, and they all affect the same thing, you won't be able to isolate it and get good data. If you don't get enough traffic through it to get data before months have passed, you won't be able to get good data. If you can't get good data, go with whatever you have reason to believe will be a better experience for your customers, and consider coming back to it when you can get good data.

I prefer not to share intermediate data on tests that are currently running, and I generally won't look at it myself either.

From there, I would commonly share a list of things I'm planning to test and what rationale there is for doing so (often including an estimate of value through conversion point), a list of any definitive results, and then a summary of tests which didn't obtain a significant result in either direction.

Sharing upcoming tests can lead to push-back and second-guessing. But you really need to let the client know what you're planning on doing in case there are any blocking issues which you weren't aware of. And clients will tend to trust you more as you demonstrate results, particularly if you're helping to teach them as you go.

Other teams...calendar, Trello. Really, whatever works for you both. Communicate. Preserving scent throughout the customer journey does wonders for conversion rates, so keeping that consistent requires coordination between different marketing assets.

I hate seeing just percentages. If you get a 50% conversion rate...what's that based on, two leads? For that matter, I don't much like seeing absolute numbers without context for how many leads you had to shovel into the funnel to get that result. Give me enough data to know what actually happened. Ideally, this includes the type of leads that came in (sources & segments), their volume, the conversion volume, and then the conversion rate.

@MariaSallis 

1. Always be testing. Sort of. You should be very careful in testing situations where you have low, low numbers that don't lend well to statistical significance. 

2. If you're an agency handling CRO testing for a client, you want to be sharing two types of data - outcome data (conversions) and testing data (any data that supports or rejects your hypothesis for testing, whether it's heatmap data, scroll data, bounce rates, etc.) - how often? Set a schedule to check in and discuss tests. You definitely want to do round-ups/analysis at the end of a defined testing period/when your numbers are significant enough to make a call.

3. You can't have too many cooks in the kitchen, and everyone needs to communicate. Processes have to be put in place to discuss and evaluate changes before they're blindly implemented. SEO/inbound/whoever all need to be connected and aware of what the others are doing. I've found from a copy standpoint, check-in meetings where everyone can plan and discuss can be useful, as long as there's a moderator who can weigh up the feedback and inevitable conflicts and make a decisions as to what's next.

4. Conversion rate. Number of conversions is irrelevant - you always want to earn that next %age. But even more than this, you want to track the  TYPE of conversion, especially when you sell multiple products. You want to know the kinds of people who convert and go on to become higher lifetime value customers, the customers who convert and then ask for refunds, the customers who convert and never show up again. Not all conversions are created equal, and if your relationship with testing ends after initial conversion, you're missing a ton.

Thanks for doing this. I was wondering how you guys track conversions through a customer's lifecycle. Meaning, once they give you an email or make their initial purchase, do you continue to track those customers to see how profitable they are in the long run?  If so, how do you do measure it? 

It seems like there are situations where a marketer might get a little too eager pursuing initial conversions and use tactics that drive less than ideal (or not as profitable) customers. Conversely, I could see how a tactic may drive fewer initial conversions  but the conversions result in more loyal (or profitable) customers in the long run. 

Thanks!

@ryanevans We have ended up coding connections between our billing software (Recurly) our app (Unbounce) and KISSmetrics, so that we can do cohort analysis based on activation.

We consider a customer activated when they've done 3 of 5 things. Publish x pages, ran x traffic, paid us twice etc. 

This lets us see the impact of testing on a more important metric than simple acquisition.

@ryanevans Personally, I don't think enough companies do this. I hate when I get emails from my bank, for example, that tells me about a product I already own. 

The best tools for the job will greatly depend on what kind of business you're in too. Custora for example, does a great job of lifecycle tracking for ecommerce solutions, and can automatically identify certain behaviors. 

But a good deal of it comes down to the questions you ask, and the segments you create in your own data. If you're in ecommerce, have you segmented out your top 1% of buyers in all of your programs (GA, KISSmetrics, Email Lists, etc) If you have, than you should use what you know about them to: 

A.) Push the behavior a little more (get them buying more)
B.) Learn from them how you might encourage the segment right below them (top 10% of buyers) to become a top 1% customer. 

There's research that shows the top 1% spends 30x more than your "average" customer, so how can you nudge more people into that category? 

@ryanevans 

Record significant customer events, minimally including those where they decide to give you money, or where they otherwise opt in to receiving value from you (e.g. newsletter subscriptions). This should also include support events.

If you generate a marketing contact with a customer, and it shows ignorance of facts which you know-in-the-database about them, or should have known, I count that as a marketing failure. It's very common, but it annoys more or less everyone who it happens to, and it absolutely excels at creating missed opportunities and customer dissatisfaction.

It's not that hard to do just a little bit better with personalized marketing, and it often ends up generating meaningfully increased value for both your customers and your company. And increased customer success tends to snowball into increased customer lifetime value, so it's not just the one event which your company benefits from.

"High conversions here, wrong customers later" is a common pitfall. Say I start everyone on a free product. Are the people who are happy to shell out $500/mo for a simple but effective solution and the people who want a free product then complain because it doesn't also do their laundry even the same people? No, probably not. Optimize for your actual target customer, not just single-point conversions.

@ryanevans Enterprise CRM platforms have really improved in this area over the years. The technology exists, it's just largely untapped. I know for certain that Salesforce and SilverPop have the ability to track all engagement and events if the web pages are tagged correctly. (From a landing page development perspective, it's key to collaborate with your CRM expert to set up forms, etc. correctly to leverage these benefits.) You'd be stunned at what can be measured these days.

If you drive traffic via email, and that user shares on Facebook or watches a video, you can report on it... PLUS, you can segment that behavior into a unique list.  You can combine and split lists based on all sorts of behavior that you choose to tag over time. The user's email is a primary key to which lots of data and behavior can be tied -- including long-term purchases.

Depending upon your business size, investment in a quality CRM can yield huge returns and optimization insights.

Even within relatively large companies, I don't encounter people with the CRO title very often.

At what size do you find that most companies have a person dedicated for this role?

@dharmesh Speaking internally, Unbounce just hit 100 employees, and we are only just at the point of having full time dedicated CRO. Prior to that it was just me doing it part time, with scattered instances from a few others in the company.

We might also want to band together on CROday and change the acronym to something that doesn't represent Croatia ;)

@oligardner HubSpot is at 800+ people and one could argue a pretty sophisticated marketing team.  We don't have a dedicated CRO person (yet).

@dharmesh Is it part of the hiring plan?

@dharmesh The question "how many of you have CRO as your primary role" was asked in the room at Conversion Conference in London last year, a significant number of people raised their hands. I didn't have time to count, but let's say that Mr. Pareto applies and it was 80%. When I looked at their job titles on the business cards I collected, very few actually had CRO in the title. Anecdotal, I know, but it appears (right now) that having CRO as your primary role doesn't require having it in your job title.

@TheRealSJR Thanks for sharing that anecdote.  It's actually not surprising.  

One of the challenges with the CRO title is that it's a CxO title -- so can be easily confused with it's most common incarnation:  Chief Revenue Officer.  I could be wrong here, but that potential for confusion may actually keep organizations from adopting that title.  

What would be interesting is for someone to do a study of CRO folks and ask what title is on their business card.  If VentureBeat ever wants to partner up with inbound.org and run such a quick study/survey, let me know.

@dharmesh That last part might work. Some may classify CRO as a part of their growthhacker role. CRO is one of my responsibilities, but my business card says Inbound Marketing Engineer.

@HooijerErik that's the thing, it's too often "one of my responsibilities", and not sole responsibility

@peeplaja Valid point, but would you say that is really necessary?

@HooijerErik depends on the size of the business - starting from a certain point - yes absolutely (and likely more than 1)

@peeplaja agreed, and then all of those can focus on a certain aspect of CRO, much as we do now at our 8-person agency.

@HooijerErik I agree with @peeplaja... Yes, a dedicated role is absolutely necessary. @dharmesh, I'm shocked that HubSpot doesn't have someone dedicated for this role. Perhaps it's just called something else? (Who's responsible for testing strategy roadmaps and customer/persona intelligence tracking?)

Imagine if companies had just as many resources focusing on converting traffic as they do driving it...  

In fact, I'd propose a spin on @oligardner's Attention Ratio metric. Let's measure a "Traffic Attention Ratio" to assess the investment balance for driving vs. converting traffic ratio. THAT would be an insightful metric! =D

@dharmesh We tried very hard to find this out--there are a little under 10k people that talk about CRO in their linkedin profiles, and less than 1/3 of them are at a director level or above. Pull out everyone that is within a consultancy or agency and the numbers dwindle. Embedded CRO teams and leadership are pretty rare. We typically see it owned by CMO or VP of Product and executed on by a product manager (who has other responsibilities.) 

@dharmesh Dell-sized enterprises tend to have one or more :) For all others it's pretty difficult to establish a trend.

Some aggressive startups like Zalando have 80 full-time optimizers on staff (hence out of this world growth), some focus on traffic, content etc (like perhaps Hubspot)

To me, one of the most frustrating things about the CRO community today is that there's a very strong focus on anecdotes ("How I increased my conversions by 527% and you can too!") and not enough around process and methodology.

Teardowns and success stories and "foolproof tactics" are fun to talk about, but I think it hurts more than it helps in many cases because it gives the illusion that things are apples-to-apples, which just isn't the case.

What are your thoughts on this?

@rsobers Amen brother. CRO is becoming the next SEO where any high school kid thinks they can do it, and CRO is just a bunch of checklists and tactics. 

People stop split tests when they have 8 vs 13 conversions, and write a blog post how they proved smth wrong or right.

Bfffffff....

The good thing is that actually CRO people (who know what they're doing) can't be replaced by a high school kid. This shit is hard.

Well everybody deserves their CRO guy. 

@rsobers Anecdotal wins are great, but I agree--it's building a scalable and repeatable process for CRO that's more exciting. While I can't reasonably predict what variations or changes are "foolproof" for any given site, I can estimate for you given your traffic & conversion rate what testing velocity you need to get to to predictably drive a positive ROI for testing. Once you know this, it's all about building a process and team for generating hypothesis, designing variations, doing analysis, and cycling back in a streamlined way. The strategy & operations of CRO are what get me up every day, not the one-off tactics. 

@clairevo Agree 100% Thanks Claire! Hopefully we'll start to see more people writing and talking about their methodology and processes.

@rsobers I see that as well - along with all the other clickbait tactics. I actually stopped reading all of the drivel and signed up for Peep Laja's coaching class. The material in the class is all I need. I thought I had my process down before but - well shit - I didn't. I think the truth is doing CRO takes a lot of grit... and knowledge in a lot of different skill areas. So "How I increased my conversions by 527% and you can too!" is going to get a crap ton more clicks than "Know almost all of the things about Google Analytics" or "101 starting points for gathering a list of questions during heuristic analysis that you may need - but remember - it's just a list of questions so gather data and validate/invalidate your questions". Of course I'm just spewing hyperbole. I'm really just here to research my latest blog post on 3,123 ways to increase CR by 5000% - missing like 2 more ways then I'm good.

@jperezish Totally. In fact, I read a fantastic series on Google Analytics on @peeplaja's site by a guest poster who demonstrated a fascinating depth of knowledge (like, suuuper deep) and now I'm contracting with her on a project. You just don't see much depth these days. I suppose it's a good way to sort people out. :)

@rsobers Let's face it - the stories are the things that sell people and get them excited about CRO, and they won't be going away. If nothing else, though, readers need to learn that one case studies outcome will not be theirs, even if they implement VERBATIM.

Process and methodology are hella important - they're the core of your success. But they're not sexy to talk about, so they don't get as much airtime. They're crucial if you're trying to sell yourself to a client (here's how we do it!) and to substantiate claims.

I think one of the biggest challenges in CRO is that it's such a jungle; I think in many ways, processes can even differ and outcomes can still be really positive.

But like you, I'd like to see more emphasis on this.

@rsobers best practices, LP audits, teardowns and all other approaches are great, but won't necessarily work for you. The goal in CRO is not to improve one KPI, it's to understand your results and implement them across the board - learn from them, understand them and utilize them on every aspect of your business. Testing correctly will help you get to know your customers better, understand them and help change entire process within your business - from customer service, to sales and even development. To do all this, you need a CRO strategy, you need an hypophysis and you need to spend a lot of time planning, before testing. To summarize: I agree. Spend more time planning and strategizing before actually testing so you can actually do something with those results. 

@rsobers Agreed. It's really the difference between customer speak vs. practitioner speak. So sadly "Ho I increased my X by XXXXX%" will never go away.

@rsobers I agree with you. And I don't agree with you.

I agree that these anecdotes are strong on 'look at me' and weak on process.

But they also 'stir the pot' for CRO, which helps the industry gain word of mouth and new fans.

As @peeplaja says, not everyone can do this. Not everyone that uses an analytics tool is a data scientist (or can be). Not everyone that bought Ventura Publisher DTP in the 1990s was a desktop publisher.

But if people are talking about CRO - even in a mildly narcissistic way - it helps those that can do this stuff, and it opens up the possibility of a new career for those that never heard of it before.

Hello guys! - First of all, thanks for making this awesome AMA.

Here is my questions:

#1 Based in your experience and knowledge on all SaaS you have known/worked with, which registration method and user activation seem most optimized? Which one do you really think is a role model?

#2 And do you think there are any "must follow" directives, which always work, even if you don't measure conversions?

Greetings from the Tractionboard team!

@davidrodriguezc Hi David! 

For a SaaS, ensure that you:

• Attract your ideal customer (IMO this is the most important thing you can do)
• Manage their expectations
• Remove any barriers for registration 

Check out this article that I wrote based on @lincolnmurphy's work, 40 Articles on SaaS Free Trials Distilled Down - I have a feeling it might answer your questions. <3

@davidrodriguezc 

There isn't really anything that always works.

Most consistent rules are "don't do this, it almost always hurts, or at least you can do much better." Don't use "submit" as button copy. Don't distract users with twenty competing CTAs. Don't mislead users. Don't break scent between how users arrive and what they see on successive pages if you can at all help it.

Always measure conversions. If nothing else, how will you know when you improve (or cause damage)?

Specific advice about tractionboard.io - 

"Get early access" isn't terribly compelling, and the people likely to put their email in aren't likely your core customers. What I would do instead, after having done some customer development interviews to be sure I had a clear picture of my target customer & what value I'm offering to them, would be to set up a drip course which helps teach them about the topic you're planning to help them with.

An "early access" mailing list is a collection of low-relevance cold leads. Ugh, right? But a drip list is warm, and made warmer over time as you provide value. Honestly, as long as you keep providing regular value, you can keep it warm almost indefinitely. Lead quality doesn't fall off because you're continuously delivering value.

Then, when it comes time to sell, you have a list of thousands of people who trust you & your expertise around this topic. It's also not oriented around a launch event like an early access list – as long as you keep delivering value you can keep marketing again and again.

I'm very interested to get an idea of some good guidelines around having stopping rules for experiments. Particularly for ecommerce sites, ~25k/month sessions.

So far I'm thinking of introducing the rules around minimum time period, and minimum number of transactions. Thoughts?

@StevieHamilton I answered a similar question above, pasting the answer here.

First of all: Don’t stop the test just when you reach 95% confidence (or higher) . This is very important.

Before stopping a test, you need to look at the following:

1. Sample size. By running tests you include a sample of visitors in an experiment. You need to make sure that the sample is representative of your overall, regular traffic. So that the sample would behave just as your real buyers behave. 

You need to pre-determine the needed sample size. There are many great tools out there for that, like this one. Or here’s how you would do it with Evan Miller’s tool: http://www.evanmiller.org/ab-testing/sample-size.html 

2. Test duration 

Your test should run for 1 or better yet 2 business cycles, so it includes everything that going on: 

• every day of the week (and tested one week at a time as your daily traffic can vary a lot),  
• various different traffic sources (unless you want to personalize the experience for a dedicated source),
• your blog post and newsletter publishing schedule,
• people who visited your site, thought about it, and then came back 10 days later to buy it,
• any external event that might affect purchasing (e.g. pay day)

and so on.

Typically your test should run 2, 3 or 4 full weeks. I myself run most tests for 4 full weeks as I see too often that a winner a 2 week mark will end up "no difference" by 4 weeks (imaginary lift). 

3. Only once your first two criteria are met, you look at the statistical confidence. 

Detailed explanation here: http://conversionxl.com/stopping-ab-tests-how-many-conversions-do-i-need/

@peeplaja Thanks again. Valuable advice. One question - was the another sample size tool you meant to include a link to? I just read that like you may have forgotten to add a link?

Thanks for taking the time to participate here!

And a follow up question: With this modest amount of traffic, would you recommend prioritising conversion testing, or prioritise getting more people to the website (with the idea that more people to experiment with will give better results to the testing)?

@StevieHamilton these things are not mutually exclusive, you need to be working on both at the same time.

You don't want to temporarily increase traffic (buy traffic) just to increase a sample size of a test since you want the traffic in the test be representative of your typical traffic. If you don't have regular paid traffic on the site, don't buy it for testing purposes.

@peeplaja Good advice, thanks again. Knowing that focusing on one at the expense of the other isn't helpful is, well, helpful to me. :)

Do you recommend to study a behavioral psychology course to every marketer who is directly related to CRO?

@MUmar_Khan I personally believe that emotional marketing, neuromarketing, psychology, and related skills will become part of the armory of the top CRO hires as we move forward. Right now, these 'unicorns' are about as hard to find (and expensive to hire) as marketing technologists are, but that will change as more people take those skills on board.

@TheRealSJR You're absolutely right Stewart. There are some people who possess this skill-set but getting them on-board can be little expensive for the companies. 

@MUmar_Khan 

Having studied both, you might get more out of a human-computer interaction design course.

CRO is effectively UX applied to a conversion goal with the assistance of statistical tests. Studying UX makes you better at CRO. Also it makes you better at spotting places which, while they don't necessarily have a direct effect on conversions, but you could stand to improve the way that you're treating users.

@grayj_ Thanks Jim for the great piece of advice. I'm planning to get this book "Smashing UX Design: Foundations for Designing Online User Experiences" by Smashing Magazine's Jesmond Allen. Hope it will have something for me! 

@MUmar_Khan From a copywriting standpoint, absolutely. The more I write, the more interest I'm taking in things like heuristics and the ways we process information/buy. So little is understood, but there are still many things that have been measured that are useful for someone in the persuasion business to pick up. I'd argue that you don't even need a formal course - there's so much free information written about persuasion and persuasiveness in copy that you can tap into. I spend a good amount of time these days digging into the work of others and analyzing how they've successfully gotten people to take action.

Knowing this stuff is hugely helpful.

@JoelKlettke You're absolutely right Joel. It's better to learn from others and apply the outcome in your work and keep experimenting with it.

@MUmar_Khan The more I learn, the more fascinated I become about behavioral psych and behavioral economics. There is such a wealth of data that comes from understanding what makes people tick, and it only helps you become stronger as a marketer.  

@tommyismyname I agree.. Who's your favorite writer on this subject? and which blogs do you read?

@MUmar_Khan 

ME! Lol :-P

Seriously though, I'm partial to @peeplaja & ConversionXL.com and I like a lot of the stuff @gregory_ciotti comes out with. 

I write a lot, so most of my reading comes in the form of researching for other articles, and I pick up bits and pieces here and there. 

I came across http://www.psychologyofgames.com/ researching a piece a little while ago, and even though the focus is on video games, a TON of the research applies to what we're talking about here. 

To me, what's more important than subscribing to a single blog and worldview, is to ask as many questions as possible. 

Why do people abandon checkouts? Why do people seem to prefer simple websites? What is it about detailed product photos that make people want to buy more? 

These questions have lead me to some really facinating research, case studies and white-papers that deepened my knowledge of buying behavior. 

@tommyismyname Yes you definitely is the one!! Personally, I'm a huge fan of @NathalieNahai :)

Thanks for explaining your way of perceiving things and I'd say any marketer can follow it if he like to learn & implement new dimensions in his work.

How do you determine what stage where a visitor is in the experience process when writing copy for a particular page? (i.e. first time visitor, just coming on board, visited 50th time, or long time veteran)

What core drives and desires change between each of those?

Thanks everyone for putting this together!

@jdquey Landing pages drive targeted traffic, so first... Who are you targeting? (i.e. Don't try to guess. Target more strategically, so you know what traffic you're driving.)

@aschottmuller Very intriguing. I've done things similar to a customer journey map before but definitely will be exploring this in more detail.

Found a great article by Peep on customer journeys that I'll be delving into more today!

@jdquey To some extent, this hugely depends on the type of landing page you're putting together. For example, sometimes I'm writing pages that are purely for remarketing, when the customer has familiarity, or a campaign where they come in off an email list where the relationship is established. I talk about the intent of the page and the audience with the business beforehand as it uncovers the obvious use-cases. I partner up with the analytics folks who are dropping cookies and serving up unique experiences.

As for a more general landing page - sometimes, you can't know. 

A home page, for example, will have all kinds of users - so you need your UX and copy to combine to serve them both. 

A download page for an asset like an ebook could serve multiple markets as well - people who know and love the company, and people who are new to it.

I think there's a ton of crossover, no matter if someone is brand new or knows the brand like a charm - so often, I look for the overlap (what are my must-haves for building trust, no matter how well the lead knows us?) - for example, it's never, ever detrimental to list the credibility of the authors on a download page for an ebook, nor will the stage of the funnel really change the CTA in that scenario.

When in doubt, I cater to the less-experienced user and trust  that the more familiar user will scroll to the information they need - keeping my focus on both providing all the info and making it easy to comb over if necessary.

As for core drives and desires - usually, the more a client knows you and trusts you, the more they want an immediate purchase or frictionless conversion. When I know I'm dealing with a recurring visitor, I spend less time on the basics and more time hammering home the purchase-driving differences and USPs; less on the "here's why you should trust us" and more on "here's why this offer is going to give you the outcome you really want."

I'd love to chat more on this - I feel like I rambled a bit there :)

@JoelKlettke To add what to Joel is saying here too, I think it's important to know what persuasion triggers are more important to your market or not. (background reading on Cialdini's persuasion triggers). 

For some markets, "social proof" might be a big one that influences the sale. How it influences the sale matters too, are they talking about quality of the product, or quality of customer service, or knowledge of the staff... If I were looking for affordable suits that make me look rich, personally, I'd like to know that the staff knew how to dress me, and that's what I'd be looking for out of a testimonial. 

On the other hand, "social proof" could actually hurt in a highly visible place. The example that comes to mind is a high end luxury handbag. In this sense, the visitor wants to feel like they're the only person buying the product. They don't want to be like everyone else, and knowing that somebody else owns this thing might turn them off. 

When I try to answer this question for "everyone" type pages, I look at the persuasion principles and what I know about the market, and determine what the "volume" should be like on each one. 

What are your favorite frameworks for rapidly prototyping and deploying a landing page(s)

@patrickcoombe 

When I'm not tied to an existing stack, I tend to just code the thing up in Sass, with the help of a library like Bourbon Neat. Simple is good. I keep a library of screenshots for reference (although crayon.co is increasingly helping there).

I'll use React, or really whatever fits JavaScript-wise, if it's merited. Calculators and mini-apps can be effective drivers of targeted traffic. Traffic-attracting pages in turn act as conversion-oriented landing pages, if you design them that way.

It's hard to go wrong with an Unbounce subscription.

@patrickcoombe can you clarify what you mean by frameworks in this context? tools?

@peeplaja He's talking about the front end. 

I don't see any developers in the list. I'll take a stab at this.

Lately, I've been digging Brent Jackson's stuff (as he's writing in CSS4). Basscss is a decent prototyping option. If you are more familiar with the command line, Tachoyns, by Adam Morse, can potentially be really fast. 

@patrickcoombe 

• unbounce.com (not clean code, customizable layout, low difficulty)
• leadpages.com (clean code, not customizable layout, low difficulty)
• webflow.com (very clean code, customizable layout, med/high difficulty)

What are the most important factors to convert users for a SaaS offering?

@dinwal 

If you're offering a free trial:

You can’t optimize what you don’t measure, so make sure every decision in your Free Trial process is metrics-driven. Only measure metrics that are actionable, and be sure to identify things most customers did during the Free Trial that ultimately lead to sales. Hint: Number of Free Trial users (i.e. those who have only created an account but haven’t engaged with it) is a vanity metric.

@dinwal SaaS is no different from anything else, its mainly about value and relevancy - does what the SaaS does solve the users problem (relevancy) and does it add more perceived value than the cost of the product. To communicate the value you need copywriting and examples (screenshots, videos etc) to do the job. 

@dinwal This really depends on the SaaS model you are using and price point.

Freemium and Trial are quite different than Instant Subscription. A low priced stand-alone tool is also quite different than and Enterprise Software Solution.

Given that, the first place I would start is ensuring you have clear business model metrics.

• What is the life time value of a customer?
• What is the profitable retention point in the customer lifecycle?
• Often the hardest question for SaaS companies to answer is “what is a “conversion?” -- an email address? sign-up for trial? 6 months of successful subscription?

Having clear measures to assess your efforts is critical for SaaS CRO success because ROI often takes longer that pure e-commerce businesses. This longer time to ROI often results in SaaS businesses abandoning CRO too early (often just before they are about to get a big payoff)

@dinwal 

Whether the person looking is actually a member of the customer segments you're targeting.

Whether the service you're offering is relevant to this particular customer, and whether you're presenting it in a way which speaks to their needs in words they'll recognize (best found by taking these words directly from your target customers).

Whether you've already adequately demonstrated to the customer that they should trust you, and that throwing money at you will make their problem stop creating enough of a hassle for them that they're considering onboarding with a new tool, which is a huge hassle, particularly once we factor in onboarding a whole team (or company).

I recently had to filter through, I don't know, about twenty different tools relative to relieving a specific pain for members of the Inbound.org team. Maybe more of them than I came away with realizing would have worked. But I wasn't really price shopping, you know? I was looking through tools until I found one that fit what we needed, then I stopped looking.

Make sure that when your target customer comes across you, they know that they can stop looking. And then make it as easy as possible for them to find you, preferably by making them aware you are trustworthy well in advance of the day when they suddenly decide to go looking.

Also, email is an incredible vector for building trust in advance. Decide now to become great at it.

@dinwal 

1. How does it give your customer base what they want?

2. Why is it worth what you;'re asking them to pay?

3. How does it work (what's the process), and how is this process an improvement?

@peeplaja really answered this well - it's like anything else. You need to know what makes your product valuable to the customer, what problem they're "hiring" your SaaS to solve and then clearly and plainly show them the obvious, wonderful ways you can get them where they want to go.

Biggest pet peeve?

Mine is multiple page checkouts.

@Martin_HarrisPR half-assed value props. I hate when writers rely on old, tired crap like:

"We do X so that you can focus on what matters!" (...so.. what matters?)

"We get to know our customers" (everyone does)

"We're the highest quality" (what does that even MEAN? nobody wants high quality!)

Amateurs just don't carefully consider their USPs or offers, so they come off sounding incredibly generic and boring.

@Martin_HarrisPR People that get one win and assume their page is THE BEST now so stop testing. Always be testing!

@Martin_HarrisPR When others pretend like the data doesn't exist. 

@tommyismyname or worse, when others attempt to manipulate math for statistical significance to claim that the data qualifies as a valid test. Statistical significance is not the same as validity. 

@Martin_HarrisPR Multiple pop-ups with copy that reads like "No, I don't want to improve my website and get more customers" if you're not interested 

@Everette ...but it works SO often. (Worth testing, naturally.)

The idea is not to hurt people's feelings when they opt out, which a lot of those aggressive opt-out buttons do, unfortunately. Rather, the idea is to give people a consequence for opting out, and to make them complete the act of opting out. They have to choose the unpleasant thing. 

Most pop-ups let you passively opt-out by not requiring you to click a No, Thanks button. "Consequence pop ups" remove all the passivity and confront your visitor with a consequence that they have to say yes to in order to move on. Potentially powerful when done right.

(Most teachers are well-versed in choice vs consequence. It's been working in the classroom [and in parenting] for, like, ever.)

@copyhackers 

Not everything that works is something I want to do.

My feeling around aggressive opt-out copy is that you can probably get much the same results with other non-passive copy choices. Either way, that's something which can be tested, and you'll have to decide if the results are worth implementing or if other factors are more important. Is this the kind of thing we just don't do, even if it boosts conversions by 5%? Maybe yes, maybe no.

Also, if your opt-out button copy is a make-or-break decision, you have bigger problems.

@Martin_HarrisPR Multiple page checkouts often get a bad rep, and it is a common "best practice" to reduce the number of stages of your checkout funnel. WiderFunnel has tested this many times, most recently for Telestream. In this case, a 4-page checkout funnel actually converted higher than a 3-page funnel. In the end, adding a checkout page increased conversions by 6.8% for Telestream. Don't assume until you've tested it. 

@chrisgoward Interesting, What do you think of single page checkouts? We use this https://www.youtube.com/watch?v=AyY2a2iRQLI Which is all above the fold. It's massively helped conversion for all our clients.

Let's say you have a new company that decided to work with you. 
Which are the first 3 things you are looking at, in order to build hypothesis for improvement?

@ValentinRadu 

1. The validity of their current and past data. This is important because you need accurate data to make determinations.

2. Their business model/competitors 

3. Their customers

@ValentinRadu 

Target customer, existing lead acquisition flow, then ideally attach that to customer lifetime value data or some sort of lead valuation. Estimate that last if you have to. You can't really do much unless you know who you're marketing to, how you're acquiring them, and how that connects to generating profits for the business.

@ValentinRadu 

1. Their customers.

2. Their USP.

3. Their CTAs.

If any one of those things is out of whack, ain't nobody doing nothin'. I will rarely come in guns blazing with ideas to change this that and the other thing - I need to do some research first to find out why their current performance is at where it's at, and whether or not what they're doing now makes an ounce of sense.

Tx, @JoelKlettke 
What do you mean by  "their customers" - things form analytics, or some kind of profiling, buyer persona?

@ValentinRadu 

i mean who the customers are, what they want to accomplish by "hiring" your product or service, what their pain points are.. It's like a persona I guess but more focused on tangible things you can find out by talking to your sales team , reading online conversations and so on. GA won't tell you what a customer values or what objections you need to overcome.

Great, tx @JoelKlettke & have a great week-end!

@ValentinRadu 

1. Their business model (you need to know how to win the game)

2. Their "best" conversion funnel (What path are people taking as related to "best" meaning the most profitable as related to highest trafficked)

3. Their customer profile (Who are they selling to?)

@ValentinRadu 

1. Proper measurement and analytics configuration.
2. Business model metrics (lead value, lifetime value)
3. Customer intel (buyer personas, research, past test results, etc.)
4. Business differentiator (UVP)

@aschottmuller short & sweet, tx!

What are your favorite tools to measure CRO? 

@JeffreyWisard 

Server-side tracking, if I can get it, since JavaScript trackers are implicitly lossy.

Per-user analytics of any sort from there.

@JeffreyWisard measure CRO (conversion rate optimization)? It's impossible to really measure optimization with a tool.

Or did you mean "measure conversion rate"? Any web analytics tool (like Google Analytics) will do the job just fine. 

@JeffreyWisard Good ol' Google Analytics, when configured right, can give you some really rich data.

As for measuring other elements of CRO, like engagement or UX, I like Rob Flaherty's Scroll depth and riveted, and it's hard to argue against something like Crazy Egg for heatmaps.

@JoelKlettke 

• Google Analytics - Event Tracking:  Here's a JavaScript bundle that covers view time, scrolling, jump-to links, and more: http://j.mp/gaeventtrackingforcro 
• AttentionWizard or Eyequant - Attention heatmaps
• HotJar - Click and scroll heatmaps, recorded sessions, funnels, polls
• MixPanel - Funnels
• UserTesting.com 
• VWO.com 
• Monetate.com - personalized, site experience testing

(That's a simple starter list. It goes on and on...)

@JoelKlettke @PeepLaja - Thanks for the help!

@JoelKlettke its actually easy to argue against CrazyEgg - overpriced and feature poor compared to alternatives (Hotjar, Inspectlet)

@peeplaja hah! That's why I prefaced with "something like".

Admittedly, not as well versed in the tools area of things as someone like you would be, and quite OK to admit that.

@JoelKlettke too many tools, too little time

Do you have any specific strategies for CRO on mobile?

@KateGwozdz Great question! 

First question is should you go responsive or dedicated design? I'm a huge advocate for designing different user experiences for mobile, according to your customer's needs, goals and behavior. The fact remains that we use devices differently - Condensing your desktop screen to a mobile one is forcing users to browse/shop/search in a way that doesn't fit them.  So I'd definitely test dedicated landing pages and funnels for movile.

A few general tips:

• Understand your customer's goals - we've covered this - see what mobile users really want to do on your site and provide that experience.
• Define clear goals - What do you want your visitors to do? is this relevant for mobile?
• Remove external links 
• Use Localization
• Enhance search
• Clean out the clutter 
• Utilize Click to call
• Carry user's identity - Allow people to log in or save a link for later (that way they can save stuff for later and continue on desktop if needed)
• Build for limited data – Mobile data connections can be slower than home broadband connections - so make sure stuff loads quickly.
• Less is more

These are just a few :) You can get some more tips here.

@KateGwozdz some great info here: http://www.slideshare.net/SearchMarketingExpo/converting-the-mobile-sem-visitor

@KateGwozdz From a copy standpoint..

1. Make it easy to read (font size)

2. If it's a local biz with any sense of urgency, make the phone number HUGE and obvious and near the beginning

3. Just like any device, avoid being overly long/verbose, unless you're selling a product that warrants more nurturing/needs more time to build trust.

4. Anticipate the context when writing CTAs; in many cases on mobile, people will want to call in, and will have very little appetite for long, demanding forms they need to mash their thumbs on.

@KateGwozdz be fat finger friendly :)

On a serious note, try to utilize mobile technology wherever you can (of course test this). Make things tapable, make quantity scroll boxes work with mobile, etc.

@KateGwozdz 

Here are 5 I learned at a recent conference

• Action Directives: The bigger the bolder the better. This applies to everything ATC (adoption to cart) and ALT ( adoption to lead) related, especially the search box and your action directives (per page). With action directives, the rule of thumb is one per view… What is there to do? What is there to click on?
• Navigation counts for 80 – 85% of mobile success. You have one view to make the magic happen – the page they come in on. It boils down to 3-8 items and a text search. Pages where the navigation is weak are dead ends. Navigation should not match the traditional site, the task they are willing to do on mobile are different.
• Email traffic is becoming more mobile traffic. Develop specific mobile landing pages for email.
• Prioritize your site search results (e.g. ecommerce site) – the order in which the search results are presented is typically far more important than the search itself. Too many results is a hindrance on mobile.
• In mobile we make our decisions in less than half a second on whether we will stay on the page or leave. So on mobile visual match and word connect are even more important.

@KateGwozdz 

Here are a few resources I've done on mobile CRO:

• Mobile Marketing Conversion Optimization (Star Wars Edition)
• Mobile Landing Page Optimization: 10 Keys to Success

I'm presenting on mobile CRO for Conversion World in a few weeks, so I'll shortly have some cool, fresh goodies available.

Other great experts that I'd recommend: Craig Sullivan (@optimiseordie) and @taliagw

Great AMA!
1) My first question is: in your experience, closed subjects catch's (like "You won't believe in this!") are more effective than open up proposals? I always wonder if people aren't tired of this bait strategy and they end not interested. Of course, AB test it and everything, but tell me more what you believe are the benefits of each strategy, and which one sounds like a winner for you. :)

and more personally,

2) What do you believe is the best part of working with CRO?

Cheers!

@eduardomarcs Hey Eduardo!

1. Honestly, I'm not sure how other people approach it but I tend to avoid gimmicks or overly link baity closed subject catches because my main appeal is trying to be genuine and straight to the point. If you can offer something that can truly help someone, you don't have to go above and beyond to try to catch their attention if you can word your copy the right way to exactly what your target audience is looking for.

2. The best part of working with CRO for me is the ability to optimize existing channels instead of only exploring new growth channels. I've had more success optimizing channels that work or that could use improvement than searching for new channels for growth. I think you need a healthy balance but I believe CRO gives you a repeatable systematic way of growing your company.

@eduardomarcs 

1. Promise them an outcome their interested in, or pique their curiosity - but by the gods, be relevant to what they care about. Hyperbole is quickly being learned to be ignored, so if it's not clear that they really. really DO want to open that email, they're gone.

2. Everything is testable, and assumptions can be proven wrong in bizarre and wonderful ways. Human behavior is weird, and I love trying to understand it.

I have two questions:

1. What are the biggest mistakes you see people making where they waste time running the wrong tests and not getting good results? Example: Not getting feedback from actual customers and diving straight into A/B testing without generating customer-influenced hypotheses.
2. What steps would you recommend to avoid these mistakes? 

@JosephPutnam I mostly partner with agencies who are running the tests/measurement, but my $0.02:

1. Doing way too much micro-testing and not enough testing of hugely different variants. Too many people get hung up on button color or changing very small elements of the page - the best thing to do is to test very, very different pages as you'll learn a lot more, a lot faster.Otherwise, you've sort of painted yourself into a corner.

2. Boldly create wildly different variations right from the start - one that is in line with the 'best practices' you assume apply to your industry, and one that intelligently challenges them in creative ways.

@JosephPutnam 

1. What is success? it's not enough to have an experiment confirm a hypothesis, there needs to be an action plan.
2. Create a carefully-crafted strategy in which there are defined stages, from formulating a hypothesis, to testing it, analyzing it, forming an action plan, implementing the action plan, and going back through the cycle as to avoid being trapped in local maxima.
   
   

Hi All and thank you for taking the time to do this. How do you present reports to less technical stakeholders? In what ways do you make it useful to them?

@Dino_Quarin Executive reporting is something we focus on a lot--if that's what you're asking about. The key things we try to deliver in our reporting:

• Is my testing program driving positive ROI? (Projected incremental revenue gains) 
• How healthy is my testing program? Testing velocity, win rate, testing pipeline, execution against roadmap
• How is this helping us hit our strategic objectives as a company / department?
• What did we learn? - High level summary of insights that can be shared across the company
• What's next? - Where are the biggest opportunities on the horizon

@clairevo Great break down. Thank you! 

@Dino_Quarin I tend to use lots of visual elements. A before picture, an after picture and how major KPIs were affected usually does the trick. This "which test won?" format has usually helped me get my information across in an easy to digest & fun format where people can walk away with repeatable information like, "That headline test increased conversion rates for that paid search campaign by 12%." 

@bellastone That's a great idea Tiffany - people are comparison engines. Thank you!

@Dino_Quarin 

Speak their language. One of the most dangerous words in conversion optimization is, ironically, "Conversions" - I've been surprised to find most stakeholders don't think in those terms. They think in sign-ups, sales, etc. - so watch your verbiage really, really carefully.

I find even non-technical people understand strategy - how well are we hitting our goals, and what do we need to change to make this work better? They often don't need to know the technical mumbo-jumbo at all - they just want to see how they're doing, and know there's a plan for improvement that's supported by the facts.

@JoelKlettke Definitely good points. Personally, I used to overwhelm with data but learned over time that plain terms and language(or theirs) backed up with data, instead of leading with it, seems to bring it home. Thank you!

@Dino_Quarin If you talk to the C-level, you talk about money.

If you talk to marketing people, you talk about insights you got from the data. 

Nobody cares about raw data.

@peeplaja for sure. No client should get raw data. Insights are gleaned from data which you pass on to them. 

@Dino_Quarin 

Keys:

1. Keep it simple. 
2. Share only what's relevant to the individual's role/goals (with more available if they ask) 
3. Adapt what you share based on the audience

The higher up in the company, the less likely the individual will be interested in what you've learned or the specifics of any AB Test. Show high-level how the efforts impact business goals. (I typically use strategy map diagrams like the one I depicted in this article: http://j.mp/somostratplan)

@aschottmuller Thank you Angie!

In my experience, a common place for users to abandon a funnel is when they are asked for their contact information. What are some things that can be done to prevent people from abandoning a step in your funnel when they are asked for their contact information?

Edit: What would the differences be in optimizing this step for lead-gen sites vs e-commerce sites, if  any?

---

@Mciocca89 

1. Ask at the right time. 
You'll always have some contingency of window shoppers with no intent to make an immediate purchase, but if MOST people are abandoning, you're probably asking too soon and haven't proven your value and credibility. 

2. Reduce friction.

People are nervous about giving up their details. They want to know what will happen when they do. They want to be assured their information won't be used to harass them. Statements that prove your intentions ("We'll send you 1 - 2 emails a month") can help ease the pain of submitting; if your product deals with security, letting them know their details won't be shared can be helpful.

You can also give them a radio box that gives them control over whether or not they're signed up to receive features and updates; when they can de-select this, they feel safer giving up their info knowing you won't abuse it.

In some cases, social proof can be used to reinforce the value of your service while confirming that real, live humans took the action the person is about to take and were happy with what they got back in return. 

3. Take only what you need.

Don't need a phone number? Don't take it. Don't need a full name? Don't ask for it. Unless the information is absolutely, positively mission critical for the next stage, DON'T ASK FOR IT. 

@Mciocca89 

Test minimizing the fields/combining fields where possible/cutting the fat.

Depending on what your demographic is, test solutions that autocomplete addresses and various fields where possible. I know that in some cases, this type of technology can harm conversion rate due, so test and take caution. 

This is really nice thing i have seen. You are doing great job.

@ItzGurpreet If this comment is directed to the Inbound.org team, thank you.

How do you show clients that the tests, cro research, work, really improved the bottom line in terms of profit for them? This isn't data which you can get from Analytics. Do you ask them upfront what there average margin is on a sale?

@PieterBaecke the basic calculation we use is:

Annual Incremental Revenue Added = (Monthly Traffic x New Conversion Rate x New AOV x 12) - (Monthly Traffic x Old Conversion Rate x Old AOV x 12)

It can get much more complicated than this, but that's the basic idea. You can do the same thing with profit, etc. As @JoelKlettke said you should be ale to get this data from your customers or from their analytics, depending on how its set up. 

@PieterBaecke 

1. Their sales team should be tied into the process/closing the leads; the sales team should be able to report an increase in leads and track/measure the quality of lead and whether or not it closed.

2. Sometimes you actually CAN show this by being tied into their ecommerce or other platforms.

3. Sharing data like this pretty much has to be part and parcel of the relationship you have, even if NDAs need to be signed. If you don't know the outcome/impact, how can anyone know it's working?

@PieterBaecke 

I would rank "what is the value or effect of a conversion here" as something that's very important to get an estimate of before starting a test. If it's not a meaningful number to the company for reasonable improvements to conversion rate, then it's not something you should be testing!

(If it affects user experience, change it anyway, and in favor of the option which makes life better for your customer.)

As @clairevo mentioned, the most basic calculation to look at is traffic at this point x conversion rate x expected value, which gives you an estimate for total realized value. You can set this formula up to look at changes in traffic, conversion rate, or value – which lets you handle changes to lead acquisition, conversion rate optimization, and pricing or cost changes.

You can also run a given value in that equation as segmented pricing tiers (run it once for each segment, sum the results), or chain together multiple conversion steps (odds of a landing page viewer of giving you an email, odds of an email subscriber becoming a sale).

In many cases, the estimate is simple and qualitative – "this is our main product landing page, all sales for this channel come through here, we have $x in sales for this channel per week" (although even then, you may have seasonal effects etc. to worry about). Then there's not much question that it's worth doing tests against. Other places on the site...do you need to test this? Check analytics, talk to the team, talk to customers. Test things which you know matter.

Then, exploit that information when presenting results (and planned tests) to clients or stakeholders. If they won't give you average order value or customer lifetime value, make up a plausible number and write a slide showing the guess you used. They'll either correct your guess or they won't, but either way you'll have made your point about the value of your results.

Do you have any examples of well-designed, CR-optimized law firm websites? A lot of my clients are personal injury law firms and I'd love to know if there are any examples of law firm sites you've seen/worked on that have been designed with very good CRO success and optimization. And any unique CRO strategies for law firms? Thanks!

@scarselli I can't speak to individual examples, but in my CRO tools study for VB Insight I did find it interesting that when I analyzed 3.1m US businesses and their CRO tools usage, law firms were in the top ten in terms of market penetration. It was a real surprise to me.

Sure - Internet businesses, software houses, and media were in the top slots as usual - but law firms (especially consumer-facing, such as personal injury firms) appear to be heavy CRO tool users.

One type of the CRO toolset I am inexperienced with is the qualitative-type tools. What is a good way of approaching the evaluation of these tools, so I can determine which will deliver the most utility for the businesses I work with?

@StevieHamilton the most important aspect is "how easy is it for the respondents to use this?"

There was (well, still is) a user testing software that was so complicated to use that people who were supposed to complete tasks on a website just couldn't figure it out. Horrible response rates.

Usertesting.com, Typeform, Qualaroo, Hotjar, Inspectlet are great in this aspect.

Second is data analysis. How easy / hard is it for you to analyze the data you get? SessionCam is a great tool for session replays and what not, but absolutely horrible UI, everything takes forever.

I also dislike SurveyMonkey for analyzing qualitative responses, much prefer Typeform.

@peeplaja Thanks! I'm unfamiliar with most of the tools you've mentioned, but these two points make a lot of sense for working out how to evaluate what ones we want to use.

@StevieHamilton 

The tools can be arbitrary, honestly. I recommend Qualaroo to collect in the wild feedback, and whatever you can get your hands on for survey tools (though Typeform does add a phenomenal level of polish to the survey experience.) 

The thing about qualitative feedback is less about the tools, and more about the questions. More specifically what questions are you asking what people, and when. 

If people are abandoning their cart, for example, an exit intent form asking "what's preventing you from buying today?" is going to provide you with more than just a general questionnaire.

If you're sending a survey specifically to your most frequent buyers, you could be asking questions about why they buy from you so frequently, and what kinds of programs they'd like for frequent shoppers. 

It's mostly about the questions you're asking and to who. 

@tommyismyname Thank You. Your answer makes a lot of sense - I'll focus on working out the questions I want answered, and by whom, before focusing on what tools to use,

Since conversion doesn't always have to happen on a landing page, I have one about converting in email campaigns. Have any of you compared buttons against the written-out URL? 

NB: the URL would be hyperlinked to itself of course, so your email marketing service can track it.

@stephanhov 

I saw this study a while back that found buttons resulted in a temporary boost in CTRs, but ultimately regressed to the mean. 

My own personal experience found no noticeable difference in my email tests, however, I'm also incredibly lazy about my email list anyways. 

BUT, the thing about when you see no noticeable difference is that you go with the thing you prefer, because in the end you're going to be happier about it, and that will show up in other areas. 

@tommyismyname Thanks Tommy!

@stephanhov 

It depends, I've seen results go either way. Test it. Be aware that the tests going one way or the other in a particular email won't hold true for all emails you send, or even for all links in a given email.

@stephanhov 

I can't point to a specific test result, but in my experience either one can be effective - it's entirely about what precedes the link/button and whether or not it is 100% obvious what will happen when they click that link. Any mystery, and they're not going to click. Buttons have the advantage of occupying more visual space and are less easy to miss (IMO), and with links, you usually need to be careful to explicitly tell people what to do next: ("click this link right now" type thing).

@JoelKlettke thanks, and a good point about looking at the link spatially in context with the content around it. I learned a bit about that today in Oli's CCD webinar.

@stephanhov Most of the time URLs are long, complex, and not readily scannable. I would rarely present an actual URL within an email or web page unless it was extremely simple and added value as something I wanted users to remember and share. Some recent examples that come to mind are CROday.com and roadtrip.unbounce.com. Note that these instances are also landing pages for social events, and not traditional product, blog, or contact us web pages.

I typically include the primary call to action in an email at least twice -- once as an an obvious button and the other as an inline (part of a sentence) text link. Depending on the email length, repeating one or both lower on the page is wise as well. The diversity of button vs. text link is helpful, and email metrics help you discern what gets the most engagement. For the most part though, it's probably best to avoid actual URLs as the text for links unless it adds value and doesn't impair scannability. (This also assumes you have a text-only email version that accommodates links accordingly.)

@aschottmuller great input, and I'd agree; for me, part of writing out the URL is to give a reader the opportunity to commit it to memory, which speaks to your examples. If conversing on social media, for example, being able to recall the URL instead of thinking "oh, it's in that email from Unbounce, let me go find it, click the link, and then copy the URL from my browser." Hardly efficient.

Hey, and thanks for the AMA and your valuable time!

What tips would you offer to someone that wants to increase the conversion rate of visitors to social media followers without using popups? Or are popups, in some form or another, a must (especially for a young site)?

@AmitiRothstein As with all A/B split tests, A/B(n) split tests, and multivariate tests, you have to try this for yourself and see what converts best. Your audience may react differently to a 'rude popup' than mine. Your audience might like the colors, design, and copy in your device, and mine may not. The simple answer is to create a hypothesis, set up some tests, and measure, measure, measure until the data tells you which works best for you.

Hello!

For years I try to sell CRO to my SEO, blogging etc. clients to no avail in most cases. I even sopped explaining it by now, I just try to add it to the overall site audit but even then people will say things like "this is not SEO, I don't want this", "this won't bring me new traffic". How do you sell conversion optimization, landing page optimization, A/B testing etc. to the uninitiated?

@onreact_com If you have 10,000 visitors per month, and convert 4% of them to leads (whatever a conversion means to you; white paper download, contact form etc.), and you want double that amount, you have two options.

• Get to 20,000 visitors a month
• Double your conversion rate to 8%

The former costs a lot of money. The latter is, largely, free. Or at least cheap in comparison.

In my latest 17k-word study of CRO tools, 3,000 users told us that - on average - they get 223.7% return on investment from using CRO. And many of those people invest zero in those tools.

So, how to sell it? Use the data.

@onreact_com This has a whole lot to do with how you're positioned from the outset. If people only know you as an "SEO" then the only thing they want from you is rankings and traffic. Period. YES, what they really want is more sales, but in their minds, you're making that happen by bringing more people in, and they see your expertise as traffic generation.

So when you pitch CRO, because you haven't proven that it's something you know a lot about, they'll balk. It's not what they came to you for.

If you want to sell it, I'd first address how you sell your services so it's not a left-field proposition that looks like an add-on. 

I love @therealsjr's suggestion of using the data and think it's important, but I also think a basic explanation that says, "I can drive you more traffic, but if you're site's not built to sell, the effort won't have the impact it should" can be a door-opener for discussion - as long as you can then step in and prove you know how to turn more of the traffic you're generating into paying customers.

should I bear in mind revenue metric of the variations with an uneven split test if significance is 100%?

@Nikalytics Always go back to revenue. In most cases it's the primary metric and what we're really optimizing for.

should I bear in mind revenue metric with an uneven split test if significance is 100%?

@Nikalytics Hi Mónica, not entirely sure what you're asking - can you expound?

How to increase conversion rate in google adwords?

@infanc you absolutely have to attend my webinar at 3PM EST today ... will be covering this in great detail.

but some ideas include use of remarketing to go after non-converters, and use of call extensions to eliminate landing pages all-together. I'll be showing how to set up these and 3 other crazy PPC-CRO hacks.

@larrykim Call extensions look AWESOME....but how can someone based in Dubai use them with clients based locally and in Asia? Everybody talks about call extension but I couldn't find even one supplier to work with where I am from.

Thanks,
Blasko

@saxestmedia yeah support for this stuff isn't quite 100% coverage across the world. i'd ask @adwords about that specific one...

@larrykim Thanks for the answer. I shall do as you suggested..

@infanc Really, really broad question, so I'll give you an equally broad answer:

1. Write better, more compelling ads.

2. Improve your landing page to ad relevance.

3. Eliminate negative keywords and irrelevant clicks.

There's a whole lot of factors you can play with here - got a more specific question? :)

Hi guys!

What strategy is better to make simple and quick experiments (e.g. CTA button's colors and shapes) with projected minor CR improvement or to go into more complex experiments (page structure, interaction flow, etc) to search for significant improvement?

Thanks!

@ANaumenko 

@larrykim I would argue that fraction of the percent increase in CR after 10 seconds change of button color and 1 day of split test might earn funding for your complex long test:)

@ANaumenko ok but i just think the button color stuff is "table stakes" and should not be confused with having a "real CRO strategy" (which should go above and beyond button colors and other small tweaks, imho)

@larrykim It depends on the stage of the company, the sophistication of their website, the market, and their internal ability (or need) to change. Large changes have significant opportunity cost, especially for a well established player for which incremental optimization can drive significant growth (given the $ multiples you're working with at that level.) Startups: it's not your button, it's you. Enterprises: it may be the button. 

@clairevo But the question is, how are you defining "enterprise" and how are you defining "big" changes too. 

Does a button color test trump a value proposition that adds more clarity, or a product description that evokes more emotion? 

Even at the enterprise level, prioritizing a button color over something that helps the visitor understand the offer better, or even the offer itself, seems like changes that are being tested because of internal politics. 

I'm all for incremental gains, but button color (unless it's blending into the rest of the page and isn't clearly the CTA) seems like it should be at the bottom of the list. 

@tommyismyname I you're focusing on the specific use case of "button color" when I'm really talking about incremental changes (small, measurable, specific") vs. radical redesigns of experiences. Agree with you on the specific use case of button color vs. value prop, but still believe strongly in incremental, continuous testing.

@clairevo 100%. 

The compounding interest of incremental gains is (to me) the way to go. I mean, you have to test big changes too, but that kind of buy-in isn't always easy to come by. We're totally on the same page. 

@tommyismyname, @clairevo,  Thanks for your opinions! I am glad you put it into incremental vs. radical discussion. I regret I used "button colors" as an example of the most simple design change. That was a bit misleading.

@ANaumenko I prefer continuous, incremental design changes. I think that it's a myth that small changes = small wins and big changes = big wins. The biggest wins I've experienced in my career have been from small changes--and the added bonus is we know exactly what drove the result. 

@clairevo respectfully, i think it's the complete opposite. small changes have small changes. big changes yield big changes. my webinar today at 3PM EST will talk about big changes that yield lasting 3-5x improvements rather than 3-5% improvements.

@larrykim We're tracking performance of radical vs incremental changes by vertical/stage/market right now,  so by next year's CRO day we can have a data-off. For now, agree to disagree.

@ANaumenko Speaking only for myself, I like to test two wildly different variants and slowly converge to one that's in the middle. When things are SO similar (button color change only), you're really not giving yourself a fair shake at figuring out whether or not you're even in the right ballpark. It's kind of like changing a sparkplug vs. upgrading your engine.

Scenario: Client with limited budget has an outdated website asking for more website traffic and higher conversion rate.

Questions:

1) Based on your experience, what would be the first set of actions that you might take in other to start convincing the client that some changes might need to be made to the website.

2) What would be the key indicators and reports that you will start monitoring in order to identify easy-to-win opportunities.

Thanks!

@soughtwords

Personally, I wouldn't take the client. I've long since gotten out of the game of trying to convince people they need something they sought my help out for. 

I starved for a while to make that happen, but being on the other side of it (and making significantly more money) it was the best decision I ever made. 

When you can spend your time learning, not selling, every step of the way, you get to focus on the work. The more selling you have to do along the way, the more warped their expectations become, and the more unnecessary pressure you put on yourself, which can muck up the entire work process. 

@tommyismyname 

This. Emphatically, this.

Not every lead can be a high-value customer, no matter how much effort you put into it. Save your time and effort for the leads that have better odds of turning into a deal you want.

The best extension of this is to start teaching at scale (blogging, email newsletters). It attracts good leads, and it will convert some ratio of lower-quality leads into good leads just by doing something you're already doing to create and prove value for your actual target customer.

Spending a major slice of your limited energy and attention struggling to turn people who aren't your target customer into someone you can help is usually a waste of your time and sanity.

What are the go to resources about CRO knowledge for you for the various topics:

• Design
• Psychology and use of language
• Data gathering and understanding

Also, which tools do you recommend I should try to improve my CRO? Optimizely, Hotjar? What is the best tool out there according to you?

@HooijerErik So many amazing resources out there. I'll point you to a few that I like to reference in my work:

Psychology and use of language - definitely follow the work of people like @TaliaGw and @BartS - they really know their stuff.

Tools? There are so many worthy of a mention. I reported on Hotjar coming out of beta and going to a full launch yesterday at VentureBeat, and you'll find a lot of information in the free sample edition of my CRO Tools study. The full version offers up detailed profiles of 36 vendors.

@TheRealSJR that is a great response for sure, just what I was looking for. Thank you! 

Anyone else willing to weigh in?

hi guys,

this is fantastic!

i'm very new to CRO & don't have much knowledge about it. my questions are.

1. how to describe cro to my seniors?

2. how to sell cro to potential clients? how can i convince them?

3. how cro aligns with seo? 

4. what not to add on homepage to distract potential customers?

would love to see your comments.

thanks.

@hyderali_ A repeat of the answer I gave to @onreact_com in many respects, but I'll expand a little.

How to describe it: CRO is money for free

How to sell it: If you have 10,000 visitors per month, and convert 4% of them to leads (whatever a conversion means to you; white paper download, contact form etc.), and you want double that amount, you have two options.

• Get to 20,000 visitors a month
• Double your conversion rate to 8%

The former costs a lot of money. The latter is, largely, free. Or at least cheap in comparison.

In my latest 17k-word study of CRO tools, 3,000 users told us that - on average - they get 223.7% return on investment from using CRO. And many of those people invest zero in those tools. 

How CRO aligns with SEO? There are elements of CRO that cross over. For example, it is completely reasonable to perform tests before and after improving site speed to see how that affects conversions. I've seen case studies that show significant improvements in conversions after increased site speed. Of course, that carries a potential SEO benefit too, since Google takes site speed into account.

@TheRealSJR Thanks Man! You are best :)

@hyderali_ we did a session this morning that totally addressed your first 2 questions. I think the recording will be made available somewhere. If you wanna see it, follow up with Talia Wolf of Conversioner. 

What are top 5 (easily implemented) best practices to convert referral traffic from industry news websites?

@XanniRed Best practices are a dangerous thing; I'd encourage you to think more about who those customers are, what they care about and where in the funnel they are after coming across as a referral. What are they arriving looking for? They may not even be ready for a hard conversion - they may want more information; they may be looking for credibility signals.

Everything starts with the customer - who they are, what they want, where in the funnel they are, what their pain points are, and why they care about getting those pain points solved. After that, it's about overcoming objections and obstacles - do they need to convince someone else internally? Are they even in a position to buy right now?

So start by analyzing that.

Then, understand who you are and how you answer all those questions in the best possible way for that lead. Explain it to them with killer copy, use a CTA that describes an action they already want to take, a promise them an outcome they really want (that you can deliver on).

Not the clean-cut answer you wanted, but helpful, I hope.

What is the best landing page you've ever seen for an on-demand service, like Taskrabbit or Homejoy?

@robpierson That's the beauty of CRO.

My opinion of someone's homepage is irrelevant.

Their CEO's opinion of their own homepage is irrelevant.

The only thing that matters is the data.

@TheRealSJR By best, I meant best performing :) 

@robpierson If only they gave me all their KPI data when I asked for it... :)

Where do we ask a question?

@MalleeBlue Right here!

@TheRealSJR Thanks Stewart. Sometimes I just need to be told.

Do you think cartoon style product-oriented video has a better potential for higher conversion rates on a landing page, in e-mail campaigns and on social media (esp. FB) as compared to text, images and hyperlinks? Thanks for hosting this session!