Trending Upward

Usability tests are awesome. They really are. So many actionable insights can be drawn from a good usability test. Usually you don’t need more than five testers to get some good data. It’s such a strong message when the powers-that-be are confronted with video after video of users struggling to complete their website goal. It makes you sad, makes you laugh, and gives you a smile on your face all at the same time.

But after the usability test, how can you know if your new redesign, restructure, re[whatever] is successful? I know you know the answer! : )

Hypothetical. Let’s say a graduate school website has received feedback that their site is not user-friendly. They decide to go through some user-testing and get some ideas for what needs to be re-designed (if anything).

Wait. Let’s step back for a minute. Before even attempting to redesign, install some analytics code on the *old* site for at least two months. Why? This will give you good benchmark statistics. How do you know if it’s successful if you have nothing to compare it to?

Wait. Let’s step back a bit further. What are your website goals? Ah, yes. I know I’m beating a dead horse, here, but it is so important to *write down* your website goals. This is a must! Why even have a website if you have no goals for that website? Don’t you care about your users?

For this hypothetical, let’s say one goal is to have users self-help (notice I *love* having users self-help!) when trying to figure out the status of their recently submitted application. So this is how we might set up the process.

1. Install some nice analytics code (any tool will do … there are many free tools out there). You’ll need to have access to your source files (hopefully you are using a content management system so all you need to do is plug the code into your footer!). Remember, you’re installing this on your old site first! Benchmark before creating something new. Then sit back and let the tool collect some data.
2. While you’re doing that, run a quick user test (remember to use members of your target audience, not your internal staff). Ask them to find their application status (of a faux application they just submitted).
3. Did they have trouble? Let’s just say for the sake of this post that they had loads and loads of trouble. Oh no! Users have no idea how to check the status of their submitted application!
4. Now look at your analtyics data, taking two simple examples:
   • What was the conversion rate of visits to those that checked application status? For this you can simply take your visits to the application status page/overall website visits x 100. Let’s say that’s 5%. 5% of your website visitors visited the “check your application status” page.
   • Now let’s look at your clicks-to-page of your application status. Warning: be careful with this metric. Taken out of context it can be very misleading. Let’s say right now that it takes, on average, 6 clicks to get to your “check your application status” page.
5. After you redesign (or it could be a simple restructure) your website, run the above reports again. Did your visit-to-”check your application status” conversion rate increase? It did? Yay! It’s working. Did the amount of clicks it took to get to the page decrease? It did? Wow! That redesign was well worth it!

If you can measure against your benchmark data, you can clearly begin to see if your website redesign or restructure is working. If those numbers didn’t improve, then maybe it’s time to run another user test. Something is still wrong. That’s ok, though. Testing is a great thing! How else will you know if something is truly working?

Of course that was a very simple example, but I hope you can see how usability and analytics go hand-in-hand. Why go through all the trouble of a redesign if you aren’t measuring the success of that redesign?

Trackback URI | Comments RSS