7 Ways to Increase Web Sales with Usability


Here are 7 ways you can increase your web site sales with usability.

Ecommerce web sites can use these seven powerful usability tools to help understand what behavior is occurring on the site, and most importantly WHY that behavior is occurring.

This is extremely helpful because managers can use this information to optimize the usability of the site – thus increasing conversion and ultimately sales.

Batman's Utility-belt

Just like Batman and his utility belt full of tools, usability experts have available a variety of tools designed to fight the crimes of task-flow error and poor performance.  These usability tools, used either separately or in conjunction with one another, can help determine where in an eCommerce web site there are usability issues that are causing lost conversion and sales.  Implementing optimizations based on the analysis of the “why” of these task-flow errors can help increase web site conversion and sales.

So here then is a list of 7 ways eCommerce web sites can increase web sales with usability:

1. Conduct in-person moderated usability testing of your web site

This is the traditional, and arguably the best method to learn about where there are usability issues that are impacting web site performance, thus sales.  I refer to it as “best” because it uses real users and directly answers the question of “why” your web site visitors are doing the crazy behaviors they do (like leaving order-flow pages early, or clicking the browser back and forward buttons continually part way through your eCommerce sales pages).

The “moderated” part of in-person moderated usability testing means the test participant and the moderator (and potentially observers) are present in the same room for the test sessions, and the moderator delivers the test and probes interesting behaviors with follow-up questions.

In-person moderated usability testing provides full interaction with and observation of the testing participants as they conduct the test.  This enables the moderator to follow-up or probe issues as they occur, and learn from the participant the “why” of behavior and task-flow issues.  In terms of actual usability data this method provides the most detail and best analysis of your web site, which eventually leads to optimization and increased sales.

Pros:  Arguably the best way to gather usability data about your web site.  Provides detailed observations and feedback via the in-person 1-on-1 interaction between the moderator and the participant.

Cons: Takes more time to set-up and administer than other methods.  Costs more than many other methods and can be difficult or almost impossible logistically due to geography / distance.

2. Conduct a remote moderated usability test of your web site

Click here to view latest job postings

Almost equal to the power of in-person moderated testing is remote moderated usability testing.  With remote testing, the participant and the moderator interact together while conducting the test, but are not in the same room, maybe not even in the same country!

By using technology such as screen-sharing software, phones, web cams, and audio-video recording, a live moderated test can be conducted from anywhere, with almost anyone.  The moderator could for example be in let’s say, Los Angeles, California, and the participant could be sitting at her computer in say London, England.

As with in-person moderated testing, remote moderated usability testing enables the moderator to observe and ask critical follow-up or probing questions with the participant.  This provides rich and detailed “why” behavioral data that can be used to make recommendations for optimizing the site, thus leading eventually to increased sales.

If a web cam and microphone are available on the participant’s end, this enables the visual and audio interaction that is obtained via in-person testing.  The only issue is that the same technology relied upon to deliver the computer-to computer or phone connection can sometimes fail (or technical term “hic-up”), causing issues or problems with the remote test.

Pros: Almost as good as in-person usability testing.  In some ways better than in-person testing because remote moderated usability testing enables testing of participants regardless of geography, and thus can be far faster and cheaper than in-person moderated testing.

Cons: The technology used for the connections (telephone and or sharing of computers) can sometimes fail, causing issues with the test itself.  In-person remote usability testing without the added visual component of a web cam means helpful non-verbal information is not obtained.

3. Conduct a remote un-moderated usability test of your web site

Often referred to as automated research (and by Scrabble champions as asynchronous research) remote un-moderated usability testing uses online tools to deliver usability tests to users without a human moderator being present.  The user is typically intercepted via a pop-up or related type of request to participate in the study, and upon acceptance the participant is walked through the study via written instructions – while behavior is captured via click-analysis and participant feedback gathered via survey-type questions at the end of the study.

Because actual users are part of the study, this is a very helpful source of data – especially when larger numbers of participants are used.  Data gathered from the analysis of the testing can lead to usability optimizations that increase conversion and thus web site sales.

Unlike moderated usability testing however, there is no ability to modify the test in real-time based on participant behavior or feedback.  Thus the “why” of behavior may not be readily available.  To help understand the “why,” automated usability testing has to carefully include sufficient questions to dig into participant attitudes and opinions about their behavior without leading the witness (so to speak).

The validity of results for remote un-moderated usability testing completely relies upon the skills and expertise of the person developing the test.  This is NOT the place for amateurs or do-it-yourselfers who have no usability training.  A poorly set up test with leading questions could actually harm web site sales, because the tainted results of a bad automated study if applied on a web site could negatively impact conversion and thus sales.

Pros: A very quick and relatively inexpensive way to gather large amounts of actual user behavioral data.  Testing can be run on almost any web site (including your competitors!) and analysis can lead to optimizations that improve eCommerce conversion and thus sales.

Cons: Developing automated tests that evaluate the right information but without providing give-away answers or leading questions is very difficult without the expertise of a trained professional.  The inability to probe participants and dig into the why of their behaviors while they are conducting the test means important feedback may not be available.

4. Conduct  an expert usability review of your web site

An expert usability review is an audit of your web site by a trained usability professional.  The usability review is an examination of the site against common usability best practices and heuristics, which are the 10 general principles for user interface design.

The usability review should  include detailed analysis including screen captures of the web pages with specific call-outs for where the issues are, what they are, and potential ways the issues can be optimized.  This is a very fast and efficient way to gather usability feedback about an eCommerce site.  The information provided can be used to optimize the site and thus improve conversion and sales.

The issue with expert usability reviews is they don’t use actual web site visitors for testing at all – instead the expert evaluates the site in much the same way a Doctor evaluates your health during a check-up.  If only one expert conducts the review, it’s unlikely that the expert usability review will find all the usability problems.  The important “why” of user behavior is not available, although the trained expert may give opinions about potential reasons for the “whys.”

Pros: Very quick and efficient way to learn about potential usability issues and opportunities for improvement.  Lower in cost than many other usability testing types.

Cons: If only conducted by one person, usability issues may be missed.  Worse, because actual users are not tested and observed, the “whys” of user task flow errors are not captured.

5. Conduct a click-stream analysis of your web site

Another handy usability tool is click-stream analysis, which can be used to map typical (or more interestingly atypical) paths through your eCommerce web site.  Evaluations of where users click on a page (or where they don’t!) can lead to testing of new placements, graphic treatments or related optimization of buttons, calls-to-action and related interaction elements.

Most click-stream analysis tools provide a snippet of code you place on your web pages, which enables the tracking of actual users clicks.  Because your real users are providing this real data, it can be very helpful when evaluating usability optimizations to improve conversion and thus sales.  Several tools also capture form entry data, which can be very useful when trying to evaluate why certain form fields have high error or abandonment rates.

However, because there is no “why” information of user behavior, click-stream analysis misses important user behavioral information and feedback.  This means that to a certain extent some guess-work is required to evaluate results and make recommendations, a dangerous proposition if the guesses are wrong and conversion and sales decrease instead of increase.  I typically recommend to my clients who are interested in using click-stream analysis to do so in conjunction with A/B or multivariate testing of resulting recommended optimizations.  It’s a safe way to hedge your bets and ensure you don’t hurt instead of improve sales.

Pros: A very handy way to evaluate actual user click actions on a page, or better yet across several pages.  Very cost-effective, and assuming you have the ability to add code to your pages can be set up and run quickly.

Cons: Doesn’t provide the “why” of user behavior.  Interpretation of results and recommendations for optimizations is completely reliant on the skills of the evaluator.  Can be privacy and or security issues if captured data includes form entries as well as clicks.

6. Conduct an eye-tracking study of your web site

Ahh eye tracking, the one subject that seems more than most to cause usability professionals to take sides and in some locations (picture a bar with a few drinks under their belts) might even cause a fight.  Some usability professionals swear by eye tracking, and some usability professionals swear AT eye tracking.

Eye tracking is a means of providing a participant with an apparatus that tracks their eye movements as they look at a web page or pages.  Typically the path of the users eyes as they move around the page is recorded (this is called “saccades”) as well as the amount of time users focus on particular places on the page (this is called “fixations”).  Aggregating multiple sessions of eye tracking can provide common visual paths users take as they view a page or pages.

Proponents of eye tracking  use the information to determine what objects seem to be capturing a users attention, and what elements are ignored or missed.  This can be helpful when analyzing and optimizing placement or graphical features of important objects on pages, which can potentially help improve usability and thus web site sales.

Opponents of eye tracking claim the data is highly artificial and potentially not valid because users are not in their normal environment and are required to use technological implements that they normally would not use.  In addition, opponents feel the data can be misinterpreted, causing potentially bad recommendations that could hurt web site conversion and thus sales.

Pros: Enables eCommerce web site managers to gather actual user visual data as participants scan web pages and objects on the page.  Analysis of what is capturing attention, as well as what is not capturing attention can be used to optimize placement or graphical features of objects on the page – potentially resulting in increased conversion and sales.

Cons: Can be expensive and time-consuming to set-up and run.  Opponents maintain that because of the technology required to capture the data, the user is not in their normal context and thus results may not be accurate.  Further, analysis of results and subsequent recommendations for changes are completely reliant on the skills of the evaluator.  They “why” of typical web site user behavior may also not be known as eye movement is only one part of the interaction a user has with a web site.

7. Conduct a simulated eye-tracking study of your web site

Because in-person eye tracking can be expensive (depending on the technology, number of participants and facility) simulated eye-tracking algorithms were created as a low-cost alternative.  These simulated tools in theory replicate a human eye-track based on object location, white space, contrast, size, etc. of objects on a page. This information can be used to infer where potential issues with typical visual paths are, and to make changes that seek to optimize tracking resulting in increased conversion and sales.

Because they are so cheap and relatively easy to perform, simulated eye-tracking can provide data almost immediately.  Test pages with changes in object location or graphical treatment can be run through the same test to determine if the changes improved the visual path.

As with in-person eye-tracking studies, opponents claim this is all a load of dingo’s kidneys (hat tip to Douglas Adams) and that not only is the “why” of user behavior data missing, so is the user.

Pros: Provides a low cost and fast alternative to human eye-tracking studies.  Data can be used to evaluate potential usability issues with object placement or graphical treatment.  New versions of placement or graphics can be tested quickly and compared against the original set to evaluate the potential usability improvement.

Cons: Does not use real users, instead uses algorithms to simulate typical human eye-tracking responses to objects on a page.  Does not capture the “why” of behavior.  Dependent on the skills of the evaluator in analyzing results and making optimization recommendations.

Conclusion: 7 ways to increase your web site sales with usability

As with Batman’s tool belt, it usually takes one or more tools used for specific means to provide the best usability testing results.  This is certainly true when dealing with the 7 usability testing tools mentioned above.  By using the correct tool or tools an eCommerce site can be evaluated and usability improvements can be made, which improves the conversion and thus sales of the site.


  1. Hi Tomlin,
    Spot on! It is an eye opener for me, a guy who has decided to become an enterprenuer that too an online service business.

    It almost sounds to me as a Bible.

    I would request you to keep on writing these kinds of excellent articles, which would be of great help to people like me out there wandering to know exactly what to do and how to do.


  2. Hi Craig,

    I don’t know how I missed this article until now; another great article from you, always a very good read!

    Thanks, and keep up the good work.

  3. Great article. I have a question about clickSteeam recording and analysis.

    I’m looking for a free wordpress click stream recording and analysis plugin. So far the closest options I find are http://www.mousetrace.com/ for $15/month. $15/month is not bad for a large money making site, but I’m really only interested in tracing a few VIP responders who are viewing my portfolio and $15/month seems like too much for something that can live on my php/mysql and I could build myself in 100 hours. http://getclicky.com/ also can be made to do this sort of with custom tags on each button but you need to do you’re own path construction and sort of create the analytics to make a click path, plus it’s still $5/month. So again, I think it’s maybe easier to build one myself.

    I just want to answer some basic questions like. . .
    1) did the person I emailed ever click on the link and visit my portfolio website.
    2) what portfolio pieces/pages did they look at and for how long (I want to know what they are really interested in and what’s irrelevant to them)
    3) Which buttons on my site actually get clicked. are their buttons that never get clicked and so I should remove them?
    4) The exact timing between events such as page changes is very important and tells me a lot. for instance it it takes about 3 minutes to read a page and nobody stays on the page for more than 1 minute I should remove some text.

    thanks again for the article and I hope to hear from you if you know of any good stuff.

Comments are closed.