3 Proven Techniques for Improving Your Websites Usability

The number one factor that makes or breaks your website is whether or not people can use it.  This is typically referred to as your website's usability.  It seems simple: if people can't do what you want them to do (buy things, subscribe to things, request a call, etc.), they won't do it.  Yet, because websites are so easy to change, several companies just create websites, web applications, e-newsletters, etc., and hope that the changes will help their business.  

What further confuses this is typically a lack of clear insight into your website's performance.  For example, how well does your website convert visitors into buyers?  What are the key decisions that visitors must make on your website?  Do you give them the information and tools necessary to make those decisions?

This article will help you focus on 3 proven techniques for improving your website's performance: website analytics, usability testing, and personas.  Exactly how you choose to implement these techniques is obviously up to you.  However, one thing is guaranteed: all three techniques help you get closer to the people who visit your website: their needs, their desires, and their behaviors.  This information is critical if you plan to optimize your website's usability to achieve your goals.

Many companies mistakenly install a standard "website statistics" program and only get a group of standard reports.  Typically, these reports do very little to help you judge the true effectiveness of your website.  

Want to get a jumpstart on creating your website analytics?  Just follow these 3 simple steps:

Measure Progress with Website Analytics

1. Begin with the end in mind ñ start with your objectives.  Define your website marketing strategy objectives (i.e. "Increase the number of qualified prospects coming from web search engines"), and what you want your website visitors to do to reach those objectives (i.e. ìSee our listing in the top 10 in Google and click on it)

2. Get in touch with your visitorsí behavior on your website.  Track how many unique visitors you get, and how long they stay on your site (including how many pages they view).  You want all of these numbers to be going up since that means you're getting more visitors who are staying on the site longer.  You are maximizing the odds that they will do what you want them to do.   

3. Develop your conversion rate.  Track how many visitors do the key action you want them to do and compare this number to your total visitors.  This helps you determine your conversion rate.  For example, if 15 out of 100 visitors requested more information from you (and that is one of your objectives), then your conversion rate for information requests is 15%.

Once you have these key website analytics in place, you can start to evolve your tracking and look for trends to optimize for.  Here are two examples:

Let's say you notice higher conversion rates on weekends.  Then you might want to spend more on online advertising on weekends and reduce your spending during the week.  

Let's say you need more visitors and embark on a search engine optimization project to improve your rankings.  Then you can track the increase or decrease in visitor flow from your project's activities.

Regardless of what you want to achieve, getting to website usability first starts with solid website analytics.  Why?  Because website analytics force you to identify those areas that matter most, and identify how well or poorly you are doing in them.  Once you know this, you are armed with key data that can help you focus your efforts and determine where things like usability testing can help the most.

Leverage Usability Testing

Usability testing is where you take people who would use your website, and watch them using it.  Typically, you ask the person to do things on the site, and you watch either over their shoulder, behind a one-way mirror, or via a second computer where you can see what is being recorded on the test computer.  

It is amazing how many things you can make better on your website just by watching people use it.  Yet, as you get into it, you may find that hiring a usability professional for a testing project can be unnecessarily expensive.  Usability professionals are helpful since they typically have substantial expertise in planning and conducting tests, as well as interpreting test results.  However, usability testing does not have to be fancy or formal: people are going to give you their opinion whether you're sitting in a research company or at Starbucks.  So be careful when hiring a professional that seems to make the testing process complicated or costly.  When someone does this, it's usually only for their financial gain.  

To successfully conduct a usability test, just follow these 5 steps:

1. Define your objectives.  Begin with the end in mind.  What do you want to accomplish with this usability test?  Do you have specific areas of your website that you want to improve?  If so, this is a great way to get ideas on how to make those areas better.  Are you planning on rolling out a new area of your website?  A usability test is a great way to do a trial run before the big launch.

2. Recruit the participants.  This will take the most time and can be the most frustrating part of the test process.  You have to find people to participate (which can be tough, particularly if you need to match specific demographic profiles), and then you need to schedule them.  Then, some will cancel, some wonít show, and some will be great test participants.  The best way to get a feel for the person is to talk to them directly more than once over the phone.  TIP: Be sure to call the person on the day of the test to remind them about it.  

3. Script the test.  You want to have an intro script, a test script, and a post-test survey.  The intro script serves as a checklist of things you want to be sure to cover with the person before you start the test.  TIP: During the part, try to focus on making the person feel comfortable giving their opinion, and reiterate that any feedback is good feedback.  The next part, the test script, is a checklist of the actual things you want the person to do.  This is followed by the post-test survey, which allows you to ask the person questions, and later compare those answers to what they said during the test.

4. Conduct the test.  This is the fun part!  You sit down with the person and walk them through the test scenario.  Some tests benefit from close ìhand holding,î while others benefit from letting the person do whatever they think is right.  It completely depends on the objectives and the information you want to collect.  In either case, the best thing to do is to record both the person and what they do on the computer.  TIP: Be sure to compensate the person for their time. 

5. Report the results.  The best way to report the results is two-fold:  First, do a quick, one-page or less recap of each session immediately after the test.  That way, the information is still fresh in your mind.  TIP: Include a picture of the user in your recap, since it will help make that personís feedback ìcome alive. î  Next, take the information collected during testing, and create 1 to 4 ìpersonasî ñ user profiles that explain the type of person, what they need from the website, what issues they encounter frequently on the site, and what can be changed to help them.  This will help you explain the results to others, and you can reuse these personas later when you are adding or updating areas of your website.

How many people should I test?  

For most usability tests, you can learn the maximum amount by only testing ten people.  Too many more and you start to see too many recurring patterns.  If you go less than ten, you might miss things or not see enough of a pattern.

Develop Personas

Let's face it - no one reads a 20-page usability report from cover to cover. It just doesnít happen. Usually, key decision makers ask for ìrecapî presentations, and then ìlatch onî to one or two key points from the study, quoting that point over and over again.

This presents a great opportunity: why not give those key decision-makers something memorable?  Enter personas.  

Personas are a way to get everyone involved and thinking about the actual people who visit your website.

What Personas are:

Fake people based on real data

A practical tool to maintain focus on your target customers

A way to make your data come alive and be more memorable

What Personas are not:

Every possible customer profile they are created from real data, like usability test results

A replacement for existing ways we design and build our website

Reporting user tests as personas is a great way to:

   1. Get key decision-makers on board with the persona concept

   2. Communicate website issues in the context of the people using your site 

Creating personas from usability testing data is time-consuming, but very valuable.  Just look across the data for key trends: what common roles, goals, and actions do you see?  Can you group the feedback along those things?  You quickly start to evolve a handful of personas which can be refined over time.  Add a name and a few pictures of that ìpersonî and youíll be on your way to creating a more user-focused website experience.

Again, exactly how you choose to implement these techniques is obviously up to you.  Even small steps can make a big impact.  You donít have to have super-sophisticated website analytics, test your website with 100 users, or develop extremely detailed personas.  Every step you take in these three areas, no matter how big or how small, will help you get more from your website and your website marketing strategy.