commit: ff40dd1 - prod push (2014-07-23 17:01:37 +0100)
Ask me anything. Conversion optimization. Usability and UX. Web design. Better ROI from your marketing activities.
If there are questions already asked that you're interested in, please upvote them.
Feel free to answer questions others have asked! I would love to have everybody contribute to the discussion.
I'm currently in Bangkok, Thailand - and there's "slight" time difference with the US. So I'll be online and answering any questions from 9am to 1pm EST and then again 9pm ET - 12am ET.
I'll promise to answer all the questions as well as I can.
Looking forward to the discussion!
http://www.lingscars.com/ and to a lessor extent http://www.ryanair.com/en seem to defy almost every key principle of usability and conversion, and yet both are ridiculously successful websites by turnover. Lings Cars turns over £50m+ and Ryanair is one of the largest airlines in Europe.
Is there something in designing deliberately bad usability? Or does it make you tear your hair out? :-)
Ed, Lingscars should have a warning before arriving to the site. I wonder how many heart attacks occur because of that site each year.
Well RyanAir does it for branding reasons. They're cheap. Their website looks cheap. You have to work more, but you'll save money. If they'd charged same prices as say Lufthansa, they wouldn't get away with it. Lingscars - today it's a brand as well. They've done a good job building it up and making it known. 99.9% of websites that will try to succeed by doing everything "wrong" will fail. There probably have been thousands of other Lingscars that have utterly failed and that's why we haven't heard of them.
I bet they do their fair share of testing as well. I think it used to look worse, and their usability has actually improved.
What's the role of in-real-life humans and feedback when it comes to conversion optimization?
If I understand you correctly, it's this:Before any conversion work is done (well, if done right), there's research phase. Depending on the project that includes surveying and interviewing people that match the ideal customer profile and past customers. The goal is to learn how they buy, why they buy, how it all fits in their life, the kind of doubts and hesitations they had before buying it and so on.
Based on the research (which also includes analytics, looking at whatever other data available, talking to the client) we develop customer personas and use cases. Both of these have to be based on data.
So all of the site is worked on with the user personas and their use cases (= user scenarios) in mind.
Another important part is usability testing (usually done with 5 to 10 people) where we observe real humans interact with the website, trying to complete a set of tasks while commenting everything out loud. This helps us identify any bottlenecks, clarity issues and other confusing stuff that have to get fixed in order to improve conversions.
If it's an exiting site, we often conduct usability testing before CRO work to identify problems, and with brand new sites we do it as soon as the first version of the new site is made live.
If this was not the answer you were looking for, please clarify :)
I left it open-ended on purpose, and you gave a phenomenal answer. Thanks :-)
Hi Peep,How did you get into CRO (we love story-tellers), and why are you "currently" in Bangkok?
Like a lot of online marketing people, I started with SEO. In 2005 (8 years ago, jeez!) I was the marketing manager of a real estate portal in Dubai. In order to drive more business, I started to learn SEO and PPC. In 2007 I founded my own SEO company. It was then where I learned that my customers were interested in money more than traffic. So I started to tweak their sites here and there, learning stuff as I went along. I read a bunch, took courses, built stuff for clients - and got results! And I've been on a learn-build-measure-learn cycle since.
My wife runs a gemstone jewelry business. In order to grow her business (+ other various reasons) she needed to get a gemology degree. Thailand is one of the centers of the gemstone world. So she is now getting a gemology degree in Bangkok. I'm location independent and here for the ride.
What are a few ways my clients can get their feet wet in CRO without spending very much money?
If you're their service provider, you can of course do it free ;) The main cost of CRO is people (and their time). The tools - for research, wireframing, testing etc - are not expensive (and there are free options available). CRO is 100% ROI driven, that's the only reason people do it. So looking at it from a purely cost perspective is incorrect. It will (should) pay for itself relatively quickly (more so for larger websites where a 20% boost means lots of $$$ in absolute numbers). It also depends on the current state of their website. If it's really terrible, it's easier to find quick wins and get a significant boost faster. The better the site (in terms of conversions), the more research and testing needs to be done to find the sources of growth. Not sure if this answered your question - so if you can be more specific, I'd be happy to provide a more specific answer.
Thanks for the reply, Peep.
We'd be more than happy to provide them w/ CRO recommendations but they contract their web dev and are very cost conscious. W/ that in mind, I think I'm looking for some CRO tools that don't require a coder but will integrate w/ a custom e-comm platform.
Thanks for any suggestions you have!
What kind of tools? What are you looking to do? Adding a/b testing tools like VWO or Optimizely take 1 minute to set up (need to insert 1 script to the site - once).
What Wordpress plugins would you recommend for someone trying to monitor and improve CRO? (A/B Testing, Pop Up management, etc.)
For email capture scroll triggered boxes work superbly well. They're not annoying like regular popups (which are still more effective than say static email capture boxes on the sidebar or what not) and are only shown when the user has demonstrated some interest in the content by scrolling down. This plugin is awesome (use it myself): http://wordpress.org/extend/plugins/dreamgrow-scroll-triggered-box/
A/B testing involves complex algorithms and hence you want to be *really* sure the tool provides accurate data. I wouldn't trust any random A/B testing plugin. What's a A/B testing tool good for if it provides false data?
Most top A/B testing tools that you can trust (VWO, Optimizely, Convert.com etc) have a Wordpress plugin available.
And of course have Google Analytics set up and goal tracking configured (lots of plugins for GA).
Selfish bonus question - what do you make of: http://darkpatterns.org/
First of all it's totally unclear what the purpose of the site is. I read it, and then what? What's the business objective?
The slideshow is a bad idea, most people will not use it and those who do will not move past the first slides. If there are messages you want to get across, they need to be in plain sight on your website - and not behind a click or 27.
Sorry, I think you and Martin may have missed the point of that. It's not my site - I was interested if you've seen the slideshow on it about how some subvertive web design elements can help increase CRO. Just to chew the fat a little.
Aah. No I haven't seen it.
I'll give you a few ideas as well :)
The largest headline makes no sense to someone who is completely foreign to the concept. The three "levels" of copy (headline, subheadline and the paragraph) you have there should in my opinion be formulated into one clear and concise value proposition. It needs to be waaaay shorter. The less words the better.
Secondly I don't really buy into the slide show thingy. My immediate response to this is.. god, I need to click and endure a long presentation that may or may not even get to the point. I guess if you turned this into a short video (and display the length beforehand) a lot more people would actually watch this thing.
I'd also display a few of the designs (or graphics explaining some concepts) right on the first page so people actually can get a perception of what a Dark Pattern is ;) And formulate a call to action to connect with it.
Because currently there's no call to action on the site apart from the links to some more presentations (which people are not ready to watch yet because they aren't sure whether to not invest a significant amount of time - people perceive presentations to be longer than simple concentrated intro videos - to go through them).
Hopefully this helps a bit :)
As above, not looking for a critique on the website itself, more the content matter in the slideshow on how some dark arts of web design can allegedly push people towards converting.
I just checked it out and it was probably the best slideshare doc I've seen (that isn't particularly difficult tbf)
That was a seriously awesome presentation. Thanks for sharing this.
"Optimise everything" would be awesome, but difficult with limited time. What's your process for choosing which specific bits of a funnel to focus on first?
First you look at the site and identify each page in the funnel. Experienced eye will immediately spot questionable or missing items on the critical pages, but of course you use actual conversion frameworks for the analysis. Then you dig in the analytics and look at each page performance + funnel performance to see where are the people dropping off.
The goal of the home page is a) communicate what it is and why and b) get people off the home page (down the funnel). So you evaluate the value proposition and how good of a job it's doing about getting people to click through to something.
Then you look at pages where people are evaluating products / services at (e.g. product pages) and also all the pages that are used to get the conversion (cart, checkout, lead gen form etc).
What are the most common CRO-related issues you come across when speaking to businesses?
Please clarify the question. Do you mean main objections? Or main conversion problems? Or something else?
Sorry - problems.
1. Missing or poor value proposition. By far the most common issue, and so important to get right.
2. Being vague ("I'm a digital storyteller") and/or hype-y ("Our Point of Sale Systems Integrate Hardware, Software and Internet Social Media Marketing Into One Giant Revenue Super System."). You have to be specific, always. The #1 goal is to be understood. People won't buy fancy words, they buy what they understand.
3. Lack of focus. You can click here, or here, or how about these links and these banners and so on. No. You need to have only 1 main CTA on every page, one clearly defined 'most wanted action' for each page.
4. Uninteresting lead magnet. People won't join lists, fill out quote request forms and so on since it's not been made interesting.
5. Inadequate evidence. "The best solution for X", "The easiest way to Y" and so on. People are not idiots. Any claim you make has to be backed up with proof.
6. Not testing! Leaving the website exactly like it is the day the developers declare it "done". Millions of unearned dollars there...
Are there any common conceptions about CRO, for instance certain methods or "recommendations", that you don't think are true, given your experience?
One "recommendation" that's often misused is about calls to action. CTAs are necessary and very effective when used properly, however people often don't understand buying stages. Putting 3 lines of text on the page and a CTA "buy now", "download" or "join" will not make that many people take action. People need to understand stuff first. In fact in many of the tests I've done (and also seen others publish) you see more often that burying the CTA buttons (showing it later) results in increased conversions.
It all goes back to the dating analogy. You can't ask for action too soon.
Is there a "generally the best way" to list product benefits and features? I've seen a lot of examples (Apple being the most famous) listing them with a an image and a short paragraph of copy alternating left and right. Any thoughts on that?
They alternate the side to keep maintain your attention. Attention is a very scarce resource and a key principle for a high-converting UI is to conserve attention at all costs!
Nobody will just read a chunk of text. Not even your mom. That's why you need to breaks it apart with imagery, text structure, interesting typography, icons, unexpected stuff and so on. It's all about novelty.
More here: http://conversionxl.com/how-to-grab-and-hold-attention/
This answers much more than the original question. Thanks.
What reports, custom reports or advanced segments do you use in Google Analytics to identify areas of a site that can be improved?
A huge chunk of it is about traffic: which channels bring in the results.
1. Top converting traffic sources2. Organic traffic report to see which keywords convert3. Ecommerce traffic report - where are the dollars coming from?4. Email signups per traffic source5. Conversion rate by device
I also track which landing pages / entry points do the best job at aiding the final conversions and which geographical location the converting traffic is coming from (good to know for targeted PPC campaigns).
Other stuff I look at:
- visitor flow: how people are moving through your site, very insightful stuff in there- top visited pages to see which pages people are interested in and what not (to spot menu items that should be removed etc)- top landing pages to see what's the first thing most people see about the website- % of people on mobile devicesIf you've got 25% or more mobile traffic, you need responsive design. The trend is only up.
Just found a gem in the location report. I am going to test something with my PPC account, I will let you know if it works or not.
What's the best CRO tool that most people don't know about?
Also, what's the best usability testing service you have come across?
Lesser known CRO tools:
Qualaroo - for surveying website visitorsGoogle Docs forms - creating online forms for user surveysTypeForm - same same, but beautiful http://www.typeform.comGoogle PageSpeed Insights - analyzing how to make website fasterCall tracking tools like Mongoose Metrics, Freespee and others
Usability testing services:
I use OpenHallway, TryMyUI and Usertesting.com. I use testers that I recruit myself as often as I can (testers need to be represenatives of your actual target group) and hiring testers from OpenHallway or usertesting.com marketplaces will still give you quite a random bunch. You can use their testers if you've got a mass appeal site (e.g. selling jeans), but you shouldn't do it if it's a niche product or targets a specific audience.
I use ethn.io for recruiting my own testers.
You also need a great test protocol, so put a lot of thought and care into the tasks you want people to complete on the site.
I just found some new tools.... thank you.
Hey Peep. I have an ecommerce site selling art and photography. The majority of the products have similar features and benefits beyond the uniqueness and story of each piece of art.
How do you handle situations where features and benefits are similar from product to product when creating product copy and descriptions?
Any insight or strategy would be appreciated.
If it's art and photography, it's not about the copy. When people walk around in the Louvre, they only care about the "copy" once they love the piece of art.
So I'd make the product page ALL about the image - make it stretch across the whole site. The bigger the better (lots of research there about the impact of large imagery).
Underneath the image tell the story. People love stories and connect with them. Why did you take this photo? What inspired you? Same with art.
Of course, you should also have a just-the-facts-ma'm part: description of the photo / piece of art + dimensions and material of the frame if applicable.
Great Peep. The product pages use the standard two column product photo left add to cart right.
The add to cart select options allow the user to select size and framing.
So there are bits of meta data built in the page but not necessarily in the custom product description I write.
I won't share an example link because I don't know if it's appropriate here.
Do you have any links to resources about the research of making a product page specifically selling art? Or what you referenced about making a full width image?
If I look at the majority of major art ecomm site they take the two column approach. I'm guessing they are testing.
I definitely don't have the traffic to run any meaningful tests.
You shouldn't do much because of how others are doing it. Most are probably not testing.
I looked at your site (mattmikulla.com/shop/ ?) and see no reasons why you couldn't make the featured product image stretch the whole width. I'm also not sure about this automatic framing thing you've got going on there, it drastically reduces the image size.
In the end people buy the photo / art because they love it, right? And they can't fall in love without seeing it properly.
Large images: there's quite a lot of research on this out there. Some of it is detailed in the book Neuromarketing. The main concept is that our brain reacts to large imagery. The bigger it is, the more people stare at it. You'll probably also find stuff on it on Google.
I'm not sure about art-specific studies out there.
That's the site.I can easily make the unframed image the hero and make the framed photos as the alternative secondary thumbnails below.
I developed and designed the site so these changes will be easy.
I love the first paragraph in this answer - such an incredible statement! :-)
If I am going to build a lead management tool for websites then what are most 5 valuable suggestion which you will give me so that user love using it. what CRO based feature should I integrate in my tool to help Clients.
A tool which will help website owners to manage leads through their website, it will show basic needed analytcis, assign leads to their sales person & follow up, various reports like keyword ranking report, ppc report, lead scoring, etc
Some useful features could be to show which page the lead came from, visitor to lead conversion rate per page (identify top performing and low performing pages), traffic sources that send the most leads (so they could know which channels to invest it).
Hey Peep, first congratulations. I read your blog and has served me well.
I have a client right now, he wants to increase the page views of your site, your site now has 1.68 page / views, average. The goal is 5 P/V. / Day.
- Time, I have very little time to reach the goal.
- Tests on post view.- More "Related Articles"- A gallery for tops or lists.
What do you think is the best strategy to increase the number of Page views?
What do you think is the prudent time to measure results? For this type of campaign.
What do you think is the importance of content in these cases?
Thanks to Peep, and to everyone else.
(Sorry for my english haha)
Thanks for reading my blog!
It's hard to say without seeing the site. The key factor - content needs to be interesting! You can't boost pageviews when the content is uninteresting to the audience :)
If it's a content site, using things like "related articles" below articles and "popular posts" in the sidebar + other similar lures will definitely work well. Add thumbnails to those things to make them more inviting.
Scroll triggered boxes that invite to read more stuff will help you as well.
Sites that sell pageviews split articles onto many pages. It will increase pageviews, but will also make the user experience poorer.
If the ultimate business objective is to sell stuff or get leads, then tell your client to not worry about pageviews at all.
What is the best way to becoming a conversion rate optimization expert?
By optimizing conversion rates :)
Build your foundation by reading say 5 on CRO, but also on usability, information architecture, UX and stuff (to get a complete picture). Books I definitely recommend:
You Should Test That - Chris GowardYour Customer Creation Equation - Brian MasseyDon't Make Me Think - Steve KrugThe Lean Startup - Eric RiesWebsite Optimization - Rich PageHow to Build Websites that Sell - myself
Learn the tools - Google Analytics, A/B testing and MVT tools, heatmaps. Learn to measure everything.
And - find people that want to improve conversion rates - and offer to work on their sites for free - to build up experience.
Thanks for mentioning my book Peep! Great job with this Q&A session...
What books do you recommend for a beginner to start reading to do SEO?
I haven't been doing SEO myself for a few years and hence I'm unsure what's the latest and best SEO book out there. You'd be probably better off reading stuff online as books might have outdated content.
There's a free course by SEOMoz people here https://www.udemy.com/whiteboard-seo/
I have website and i got more than 2600 traffic for my website but i am not able to generate leads from through seo. what should i do for leads generation?.
SEO is for traffic - getting the visitors to convert to leads is up to your website. What kind of leads are you after?
If it's email capture, offer lead magnets (http://conversionxl.com/lead-magnets-email-list-building-on-steroids/). If it's something else, make sure the landing page is 1) focused on a single action 2) the lead generation has as few fields as possible 3) there's some instant gratification involved 4) the copy on the page is user-focused, not ego-centric.
Also pay attention to the keywords used when landing on your site. Are the keywords directly related to whatever you're offering? Hard to comment specifically without knowing more details.
Hello Peep, I couldn't agree with you more about making sure that the page is focused on a single action. With how web designers and even webmasters are keen on aesthetics, they fail to realize that it can actually hinder their conversion.
One of the things that I emphasize to my clients is to remove anything that might shift their visitor's attention from what they ought to do on that page.
Why does bad design (e.g. MS Pain ads) convert higher? Retro > Everything else on the web?
My question relates to buy-in from management.
What process would you recommend for in-house marketers to get buy-in from higher management for CRO?
e.g. a site I work on is http://www.adverts.ie - I believe CRO could reap significant benefits but it's difficult to convince management that it's worth allocating development time to CRO.
Hey Mark. Paint them a picture about the glorious future with more money :) Start with analysis and research to identify problem areas. Dig in your analytics, conduct surveys and run a usability test. If needed, involve a CRO expert to analyze the site. These things are not necessaily expensive. Compile your findings, and now you have a data-backed case.
If you can show the management where the bucket is leaking from, they're much more likely to throw money at patching the holes. Prove that the problem is real.
Thanks Peep for taking the time to respond. Great tips! I'll let you know how it goes :)
Hi Peep, what are the 5 most important things a service website should have in their design so as to make a conversion, here is my company's website can you please grade it on a scale of 1 to 10 http://bit.ly/7DiJLR
Sorry Neeshu, but it's around 2 or 3. No text, no value proposition, terrible clarity, unreadable text, random cheesy stock photos etc. Nothing worth keeping.
Thanks for the reply Peep, despite all those drawbacks I have been given an arduous task of ranking the site. If you can guide me about those "5 most important things a service website should have in their design so as to make a conversion" I can pass it on to the design people who are doing these innovative things.
Hi Peep,Most of the service offering pages on my site are quite text heavy ... and I want to do CRO for these set of pages.What is the %age of copy that I should change? Or should I simply stick to changing the layouts, changing the text and color of the CTA.
Content-heavy is okay, but you need to think about readability. That's why you need to break text apart, use lots of images, lists, sub-headline and quotes and stuff to keep the novelty up.
Color of the CTA is always the very last thing to test as it extremely rarely brings any result. Think about the user and the value is communicated to the user - that's where the conversions are.
Do you have some samples landing pages which you would recommend?
I just want to say thanks for the tips. I have read through many of these and picked up a few new ideas and tips. :)
Hey Peep,I have built a business referral site, where users can get and provide referrals, in addition to being able to make cold calls.The site URL is www.bizcoterie.comI have a two stage registration process -- first stage- first name, last name, user name and password is collected and second stage- the user creates a profile.
I have used a couple of CTAs with the focus being on signup. All on the landing page.How would you rate it for conversion? How do you think it can be improved?Any insight or strategy would be appreciated.
Thanks for all those personal replies Peep have a great day.
No questions, Peep. Just want to let you know how much I love reading your articles and insights. You convey the most unique ideas and I'm a big fan! :)
Great! Head over to login and use your credentials there