why web best practice isn’t good enough

August 16, 2012 by
web communications research & testing

have you got research & testing in your quiver?

Ready for a glimpse into a nightmare?  Let's see if this scenario gets your palms sweating and your heart beating just that little bit faster.

Imagine that you're part of a web communications team that's been asked to overhaul your university's website.

You know what to do, because you know web best practice backwards and forwards.  So over a period of months you and your team

  • Bust your chops getting the copy just perfect
  • Carefully insert high search/low competition keyword phrases in all the right places
  • Develop a state-of-the-art content strategy
  • Painstakingly refine your design, navigation, wireframing, and information architecture

Then, finally, you launch.  The new site works.

Sort of.

It's not exactly a failure - the site is performing better. But you're not sure the marginal improvement was worth all the cost and effort.

What went wrong?

top-notch web communications require research and testing

Best practice is an essential part of success on the web.

But without a solid foundation of research and testing, you'll always be on shaky ground.

That's because best practice guidelines can only ever tell you what generally works.  If you want to ensure success for your site and your specific audiences, you've got more work to do.

target audience research is the place to start . . .

There are web strategies that work for 'students,' and then there are strategies that will work best for your students  (or faculty, donors, staff, etc.).

Okay - this part is a bit cheeky, since I've already talked about target audience research as central to best practice . But it's so important I'm willing to repeat myself.

And give you a couple more ideas about how to get inside your audiences' heads. These examples are specific to students, but the same logic can be applied to any other group you're targeting.

  • Read student reviews.  Find out what students are saying about your school, and get a feel for how they talk.  Here's an exhaustive list of US sites that feature student college reviews for you to mine.  For UK schools, check out The Student Room  and What Uni?.
  • Read what your students read.  Start with your school's student publications, then find out what magazines they're reading, where they're getting their news online, etc.  Not only will you connect with what they're thinking, you'll get a feel for the types of content and design that they prefer.

but not the place to finish

As Eric Sickler explains in a recent webinar for new recruiters, you need to gather other kinds of 'intelligence' as well.

  • Product intelligence.  In addition to understanding the basics of what you do best, a good sense of history, a stash of interesting facts, and some great stories will serve you well.  Keeping good connections with people all over campus will help you accomplish that and more.
  • Competitive situation.  You need to understand your closest competitors well. Very well.  Because without that understanding, you can't distinguish what makes you different.  Or in marketing-speak, your 'unique value proposition.' It's the key to the castle in getting prospective students to pick you instead of 'them.'

Okay, so if you accomplish all that, and have applied everything you learned, you get an A+.

But no, you're still not done.

always be testing

You simply can't know how your carefully nurtured baby will fare once she grows up and goes out into the world.

Will people love her?  Diss her?  Completely ignore her?

There's no point in speculating.  Your only useful option is to watch what people actually do.

And that's where testing comes in.

* Usability testing

This is where you literally observe and listen to people as they try to use your site.  It could be a recorded screen capture, one-on-one testing, or variations on a theme.

Jakob Nielsen's usability report on 'College Students on the Web' gives a great overview of student web preferences, and is highly recommended.

But to get specific information about your site, conduct your own usability testing with even a few students and other site users on campus - ideally before you launch.  Steve Krug's Don't Make Me Think gives a simple version of how to do this.  Avinash Kaushik details a more elaborate 'scientific' approach.

* Web analytics  

Using analytics involves tracking how visitors interact with your site using tools like Google Analytics.

Can be complicated.  Can be overwhelming.  Can mean people don't quite get there.

In fact, a recent survey shows that while higher ed is adopting web analytics, it's . . . happening . . . slowly.

But if you're not employing analytics tools, you're relying on hunches rather than data to inform the design of your site and its ultimate success.  Karine Joly makes a persuasive case about why you should embrace web analytics and provides useful tips on how to get started.

Or if you want to take things slowly, you could begin by

  • Doing some A/B split testing to test specific elements of key pages on your site.
  • Following Alan Etkin's advice on what he would do if he only had five hours a week to work on analytics:  'I’d automate as many reports as possible using the custom dashboard functionality of Google Analytics. . . .   For analysis, I’d focus on the key conversion events you’re able to track, whether they’re registrations or requests for information.'

Are you given the time and resources to do as much research and testing as you'd like?  What have your experiences been?

Photo credit: .Creative Commons License Neeta Lind via Compfight

Leave a Comment