The Ultimate Guide To A/B Testing

Advertisement

A/B testing isn’t a buzz term. A lot of savvy marketers and designs are using it right now to gain insight into visitor behavior and to increase conversion rate. And yet A/B testing is still not as common as such Internet marketing subjects as SEO, Web analytics and usability. People just aren’t as aware of it. They don’t completely understand what it is or how it could benefit them or how they should use it. This article is meant to be the best guide you will ever need for A/B testing.

What Is A/B Testing?

At its core, A/B testing is exactly what it sounds like: you have two versions of an element (A and B) and a metric that defines success. To determine which version is better, you subject both versions to experimentation simultaneously. In the end, you measure which version was more successful and select that version for real-world use.

This is similar to the experiments you did in Science 101. Remember the experiment in which you tested various substances to see which supports plant growth and which suppresses it. At different intervals, you measured the growth of plants as they were subjected to different conditions, and in the end you tallied the increase in height of the different plants.

A/B testing example1
Large version2

A/B testing on the Web is similar. You have two designs of a website: A and B. Typically, A is the existing design (called the control), and B is the new design. You split your website traffic between these two versions and measure their performance using metrics that you care about (conversion rate, sales, bounce rate, etc.). In the end, you select the version that performs best.

What To Test?

Your choice of what to test will obviously depend on your goals. For example, if your goal is to increase the number of sign-ups, then you might test the following: length of the sign-up form, types of fields in the form, display of privacy policy, “social proof,” etc. The goal of A/B testing in this case is to figure out what prevents visitors from signing up. Is the form’s length intimidating? Are visitors concerned about privacy? Or does the website do a bad job of convincing visitors to sign up? All of these questions can be answered one by one by testing the appropriate website elements.

Even though every A/B test is unique, certain elements are usually tested:

  • The call to action’s (i.e. the button’s) wording, size, color and placement,
  • Headline or product description,
  • Form’s length and types of fields,
  • Layout and style of website,
  • Product pricing and promotional offers,
  • Images on landing and product pages,
  • Amount of text on the page (short vs. long).

Create Your First A/B Test

Once you’ve decided what to test, the next step, of course, is to select a tool for the job. If you want a free basic tool and don’t mind fiddling with HTML and JavaScript, go with Google Website Optimizer283. If you want an easier alternative with extra features, go with Visual Website Optimizer4 (disclaimer: my start-up). Other options are available, which I discuss at the end of this post. Setting up the core test is more or less similar for all tools, so we can discuss it while remaining tool-agnostic.

You can set up an A/B test in one of two ways:

  • Replace the element to be tested before the page loads
    If you are testing a single element on a Web page—say, the sign-up button—then you’ll need to create variations of that button (in HTML) in your testing tool. When the test is live, the A/B tool will randomly replace the original button on the page with one of the variations before displaying the page to the visitor.
  • Redirect to another page
    If you want to A/B test an entire page—say, a green theme vs. a red theme—then you’ll need to create and upload a new page on your website. For example, if your home page is http://www.example.com/index.html, then you’ll need to create a variation located at http://www.example.com/index1.html. When the test runs, your tool will redirect some visitors to one of your alternate URLs.

Once you have set up your variations using one of these two methods, the next step is to set up your conversion goal. Typically, you will get a piece of JavaScript code, which you would copy and paste onto a page that would represent a successful test were a visitor to arrive there. For example, if you have an e-commerce store and you are testing the color of the “Buy now” button, then your conversion goal would be the “Thank you” page that is displayed to visitors after they complete a purchase.

As soon as a conversion event occurs on your website, the A/B testing tool records the variation that was shown to the visitor. After a sufficient number of visitors and conversions, you can check the results to find out which variation drove the most conversions. That’s it! Setting up and running an A/B test is indeed quite simple.

Do’s And Don’ts

Even though A/B testing is super-simple in concept, keep some practical things in mind. These suggestions are a result of my real-world experience of doing many A/B tests (read: making numerous mistakes).

Don’ts

  • When doing A/B testing, never ever wait to test the variation until after you’ve tested the control. Always test both versions simultaneously. If you test one version one week and the second the next, you’re doing it wrong. It’s possible that version B was actually worse but you just happened to have better sales while testing it. Always split traffic between two versions.
  • Don’t conclude too early. There is a concept called “statistical confidence” that determines whether your test results are significant (that is, whether you should take the results seriously). It prevents you from reading too much into the results if you have only a few conversions or visitors for each variation. Most A/B testing tools report statistical confidence, but if you are testing manually, consider accounting for it with an online calculator5.
  • Don’t surprise regular visitors. If you are testing a core part of your website, include only new visitors in the test. You want to avoid shocking regular visitors, especially because the variations may not ultimately be implemented.
  • Don’t let your gut feeling overrule test results. The winners in A/B tests are often surprising or unintuitive. On a green-themed website, a stark red button could emerge as the winner. Even if the red button isn’t easy on the eye, don’t reject it outright. Your goal with the test is a better conversion rate, not aesthetics, so don’t reject the results because of your arbitrary judgment.

Do’s

  • Know how long to run a test before giving up. Giving up too early can cost you because you may have gotten meaningful results had you waited a little longer. Giving up too late isn’t good either, because poorly performing variations could cost you conversions and sales. Use a calculator (like this one6) to determine exactly how long to run a test before giving up.
  • Show repeat visitors the same variations. Your tool should have a mechanism for remembering which variation a visitor has seen. This prevents blunders, such as showing a user a different price or a different promotional offer.
  • Make your A/B test consistent across the whole website. If you are testing a sign-up button that appears in multiple locations, then a visitor should see the same variation everywhere. Showing one variation on page 1 and another variation on page 2 will skew the results.
  • Do many A/B tests. Let’s face it: chances are, your first A/B test will turn out a lemon. But don’t despair. An A/B test can have only three outcomes: no result, a negative result or a positive result. The key to optimizing conversion rates is to do a ton of A/B tests, so that all positive results add up to a huge boost to your sales and achieved goals.

Classic A/B Testing Case Studies

Here are some case studies to give you an idea of how people test in the wild.

Writing Decisions: Headline Tests on the Highrise Sign-Up Page7
37signals tested the headline on its pricing page. It found that “30-Day Free Trial on All Accounts” generated 30% more sign-ups than the original “Start a Highrise Account.”

37signals A/B test8

“You Should Follow Me on Twitter Here”9 (Dustin Curtis)
This much-hyped split-test involved testing multiple versions of a call to action for Twitter followers. Dustin found that “You should follow me on Twitter here” worked 173% better than his control text, “I’m on Twitter.”

You should follow me on Twitter here10

Human Photos Double Conversion Rates11
A surprising conclusion from two separate A/B tests: putting human photos on a website increases conversion rates by as much as double. Scientific research backs this up, saying that we are subconsciously attracted to images with people.

Human photos vs. generic icon12

Google Website Optimizer Case Study: Daily Burn, 20%+ Improvement13 (Tim Ferriss)
A simple variation that gave visitors fewer options too choose from resulted in a 20% increase in conversions. The winning version was also much easier on the eye than the control in its detail and text.

Home page screenshot14

Two Magical Words Increased Conversion Rate by 28%15
The words “It’s free” increased the clicks on this sign-up button by 28%, illustrating the importance of testing call-to-action buttons and how minor changes can have surprisingly major results.

It's free screenshot16

Changing the Sign-Up Button from Green to Red17
Along with its other A/B tests, CareLogger increased its conversion rate by 34% simply by changing the color of the sign-up button from green to red!

Green v/s Red18

Single page vs. multi-step checkout19
If you have an online store, it is quite common to see visitors abandoning the purchase process at the time of checkout. This A/B test found out that a single page checkout process works much better at completing sales than multiple-page checkout process.

checkout20

"Mad Libs" style form increases conversion 25-40%21
Defeating conventional wisdom, in this A/B test it was found out that a paragraph-styled form with inline input fields worked much better than traditional form layout. Though the result was probably specific to their offering as it wasn’t replicated in another, separate A/B test22.

madlibs23

Complete redesign of product page increased sales by 20%24
A software product company redesigned their product page to give it a modern look and added trust building elements (such as seals, guarentees, etc.). End result: they managed to increase total sales by 20%. This case study demonstrates the effect of design on sales.

aquasoft25

Marketing Experiments response capture case study – triple digit increase in conversions26
Through a series of A/B tests they optimized the mailing list opt-in rate by 258%. Focus was to remove all distractions and require the visitor to only provide email address. For completing his/her complete profile, the landing page motivated the visitors with an Amazon gift card (which was again split tested).

marketing-experiments27

Tools For A/B Testing

A number of tools are available for A/B testing, with different focuses, price points and feature sets. Here are some:

Resources For Deep-Diving Into A/B Testing

If you’ve read this far, then A/B testing has presumably piqued your interest. Here, then, are some cherry-picked resources on A/B testing from across the Web.

Get Ideas for Your Next A/B Test

Introductory Presentations and Articles

The Mathematics of A/B Testing

(al)

Footnotes

  1. 1 http://www.smashingmagazine.com/wp-content/uploads/2010/06/abtesting.gif
  2. 2 http://www.smashingmagazine.com/wp-content/uploads/2010/06/abtesting.gif
  3. 3 http://www.google.com/websiteoptimizer
  4. 4 http://visualwebsiteoptimizer.com/
  5. 5 http://visualwebsiteoptimizer.com/ab-split-significance-calculator/
  6. 6 http://visualwebsiteoptimizer.com/ab-split-test-duration/
  7. 7 http://37signals.com/svn/posts/1525-writing-decisions-headline-tests-on-the-highrise-signup-page
  8. 8 http://37signals.com/svn/posts/1525-writing-decisions-headline-tests-on-the-highrise-signup-page
  9. 9 http://dustincurtis.com/you_should_follow_me_on_twitter.html
  10. 10 http://dustincurtis.com/you_should_follow_me_on_twitter.html
  11. 11 http://carsonified.com/blog/design/human-photos-double-your-conversion-rate/
  12. 12 http://carsonified.com/blog/design/human-photos-double-your-conversion-rate/
  13. 13 http://www.fourhourworkweek.com/blog/2009/08/12/google-website-optimizer-case-study/
  14. 14 http://www.fourhourworkweek.com/blog/2009/08/12/google-website-optimizer-case-study/
  15. 15 http://visualwebsiteoptimizer.com/split-testing-blog/ab-test-case-study-how-two-magical-words-increased-conversion-rate-by-28/
  16. 16 http://visualwebsiteoptimizer.com/split-testing-blog/ab-test-case-study-how-two-magical-words-increased-conversion-rate-by-28/
  17. 17 http://dmix.ca/2010/05/how-we-increased-our-conversion-rate-by-72/
  18. 18 http://dmix.ca/2010/05/how-we-increased-our-conversion-rate-by-72/
  19. 19 http://www.getelastic.com/single-vs-two-page-checkout/
  20. 20 http://www.getelastic.com/single-vs-two-page-checkout/
  21. 21 http://www.lukew.com/ff/entry.asp?1007
  22. 22 http://www.kalzumeus.com/2010/02/27/lesson-from-madlibs-signup-fad-do-your-own-tests/
  23. 23 http://www.lukew.com/ff/entry.asp?1007
  24. 24 http://carsonified.com/blog/business/the-business-case-for-ab-testing/
  25. 25 http://carsonified.com/blog/business/the-business-case-for-ab-testing/
  26. 26 http://www.marketingexperiments.com/blog/research-topics/response-capture-case-study.html
  27. 27 http://www.marketingexperiments.com/blog/research-topics/response-capture-case-study.html
  28. 28 http://www.google.com/websiteoptimizer
  29. 29 http://www.bingocardcreator.com/abingo/
  30. 30 http://vanity.labnotes.org/
  31. 31 http://visualwebsiteoptimizer.com
  32. 32 http://unbounce.com/
  33. 33 http://performable.com/
  34. 34 http://vertster.com/
  35. 35 http://sitespect.com/
  36. 36 http://www.webtrends.com/products/optimize.aspx
  37. 37 http://www.omniture.com/en/products/conversion/testandtarget
  38. 38 http://whichtestwon.com/
  39. 39 http://www.conversion-rate-experts.com/articles/101-google-website-optimizer-tips/
  40. 40 http://abtests.com/
  41. 41 http://visualwebsiteoptimizer.com/ideafox.php
  42. 42 http://elem.com/~btilly/effective-ab-testing/
  43. 43 http://exp-platform.com/Documents/GuideControlledExperiments.pdf
  44. 44 http://20bits.com/articles/an-introduction-to-ab-testing/
  45. 45 http://20bits.com/articles/statistical-analysis-and-ab-testing/
  46. 46 http://www.evanmiller.org/how-not-to-run-an-ab-test.html
  47. 47 http://visualwebsiteoptimizer.com/split-testing-blog/what-you-really-need-to-know-about-mathematics-of-ab-split-testing/
  48. 48 http://blog.asmartbear.com/easy-statistics-for-adwords-ab-testing-and-hamsters.html
  49. 49 http://www.cennydd.co.uk/2009/statistical-significance-other-ab-test-pitfalls/

↑ Back to top Tweet itShare on Facebook

Paras Chopra is founder of Visual Website Optimizer, the world's easiest A/B testing tool. Used by thousands of companies worldwide across 75+ countries, it allows marketers and designers to create A/B tests and make them live on websites in less than 10 minutes.

Advertising
  1. 1

    Great article. Thx. We are currently thinking of looking into A/B testing for our product, your article and resources are a big help!

    0
  2. 52

    testing the control first, yet test it simultaneously. Seems like weird advice in the Do’s and Don’ts. Which one?

    1
  3. 103

    Fascinating article!

    Of course, now I find myself asking, “Is the article supposed to be on the left side of the page? Is the Post Comment button ALWAYS green?” I’ll never look at websites the same. :)

    0
  4. 154

    Excellent POST!

    0
  5. 205

    You might want to add FairlyCertain ( http://www.fairtutor.com/fairlycertain/ ) to your list. It’s a lightweight AB testing library for ASP.NET and ASP.NET MVC, modeled along the lines of A/Bingo.

    0
  6. 256

    Nice article Paras! I was exploring the topic and found this one very nice. Thanks!

    0
  7. 307

    I recently set up a test using google adwords with a very basic landing page to test company names and value propositions but Google shut down the campaign after a few hours. How can you run tests without having an elaborate website? All tips and resources are much appreciated!

    0
  8. 358

    i know how to use A/B in to MKV to MP4 Mac, i Need to change my ideas,i read a similar articles .With the page layout optimization also need to continue to learn, thank you, good article refueling.
    How to keep customers, learned.

    -1
  9. 409

    Great beast of an A/B test guide = cool! I have put together a less detailed top tips guide for A/B and multi variate testing you might find vaguely interesting – http://www.searchmuse.com/blog/2011/7/23/advice-for-a-b-and-multivariate-testing.html

    1
  10. 460

    Hi,

    I would like to know if is there a way to do the A/B Testing in facebook games. Do you any tool for that? Because I investigated the Google Website Optimizer and it’s just for websites, right?

    0
    • 511

      Hi Mil, the HoneyLizer which is a social optimization platform (for Facebook) that serves personalized, ‘best-fit’ page options to each user, automatically and in real-time, can be used also for A/B testing.
      Feel free to contact me if you want more info – yaron[at]BeesAndPollen.com

      1
  11. 562

    Nice blog i like it thanks

    -1
  12. 613

    Hi Paras, I have a question for you that I’ve been struggling with so it would be great if you could help. Can you use GWO to A/B test bounce rate?

    1
  13. 664

    Thnks ,usefull article

    0
  14. 715

    I’ve created a ruby A/B testing tool called Split.

    Split is a rack based A/B testing framework designed to work with Rails, Sinatra or any other rack based app.

    It’s heavily inspired by the Abingo and Vanity rails ab testing plugins and Resque in its use of Redis and is designed to be hacker friendly, allowing for maximum customisation and extensibility.

    https://github.com/andrew/split

    0
  15. 766

    Thank you for this article! It is a great piece of work – not only for someone who wants to get started with A/B testing.

    0
  16. 817

    Thanks for a brilliant article. I have had three websites so far and have learnt from them all. Every website developer has raved about landing pages. It never happened as I would not relinquish content control to them. The link to ‘unbounce.com’ is such a kindness. Thank you. I have a self-build from a host for small companies. It is so slow to upload that it is pointless. Basically I want the perfectly built website. I will have a look around your site for answers. I will go and start. Thanks.

    1
  17. 868

    Unfortunately, many of the conclusions drawn from these a/b tests are quite flimsy. As a psychology researcher the flaws in the methodology are quite striking.

    Fortunately there are people like myself who audit such testing and let companies know whether or not the results are salient and actionable, or just junk. Indeed much of a/b testing is no more relevant than science 101 tests.

    0
  18. 919

    For example, the D. Curtis “experiment” listed above is tremendously flawed.

    You can read my breakdown of just a few of the problems with his testing here:

    http://researchaudit.wordpress.com/2012/09/18/flaws-in-the-d-curtis-you-should-follow-me-on-twitter-experiment-ab-testing/

    0
  19. 970

    Great post! Thank you :)

    0
  20. 1021

    Even 3 years after the post, highly relevant and useful. Especially when the Internet industry has begun its next boom in India.
    Well written !!

    0
  21. 1072

    A/B Testing is more like a marketing method than a method used for Search Engine Optimiztaion. After all, you are checking the behavior of people. You are not looking at how to exploit the weaknesses of search engines in order to grow the rank of your website.

    0
  22. 1123

    Multivariate testing and observation is highly necessary in any internet marketing process. As much as we have useful analytics and demographics, A/B testing is its’ ultimate completion.

    Den,
    searchzinger.com

    0
  23. 1174

    It is one of the deepest article about a/b testing. one more i have click there called 101 tips it is also amazing topics. For beginners like me it will help to do websites optimization so well. A BIg THANKS

    0
  24. 1225

    Even if I discover your article 3 years after its publication you’re still right, people lack information about AB Testing, and do not dare trying it. This is why articles like yours is very useful ;-) Thanks for the tips you give, I will use them a lot. Here is another article on testing that may interest your readers: http://www.atinsight.co.uk/actualite/test-procedure-7-steps-to-successful-testing/

    0
  25. 1276

    Visual Website Optimizer is a great tool!! Congrats!!

    0
  26. 1327

    i don’t want to comment, but just hit like or something..

    anyway – has smashingmagazine testet design/usability un A/B test?

    0
  27. 1378

    Good article, well written.

    Thanks Paras!

    0
  28. 1429

    Just wanted to thank you, Paras, for this comprehensive guide (and especially the added resources). I have a similarly-related glossary on SEO terms (in which I reference this guide, as well as the Visual Website Optimizer tool) if anyone is interested in taking a look and offering any input – http://tarungehani.com/seo-glossary

    0
  29. 1480

    any good ideas for APP AB testing? E.g. Android or IOS

    1
  30. 1531

    It’s easy to use and convinient option for finding better result.

    0

↑ Back to top