Six projects to do before you do a webside redesign: Learn what the user wants to do

[This is the first part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]

In the twelve years since I've been doing agency work, here is my entirely unscientific list of the most common reasons people redesign their website:

1. "It looks 'old', we need to update our look" also "We're changing our logo and colors".
2. "We want to use some new technology (usually a content management system)."
3. "We want to create a more engaging experience for the user."

None of these really speak to a deep understanding of the user's desires.  Furthermore, once you understand the user's desires, you may not actually need to do a redesign.  You may need to create more or different kinds of content, or you may need to conduct your work in a different way so you have something substantively different to talk about with your users.  Or you may need to invest in a piece of functionality that would allow your users to immerse themselves in your mission in a more engaging manner. 

But all of those projects are smaller and cheaper than a redesign, and that's not in the agency's interest.

Let's take the stated reasons for a redesign one at a time:

It looks 'old', we need to update our look / We're changing our logo and colors

This is the impetus for a lot of redesigns, but the magnitude of what gets changed is way out of whack for a simple graphics change.  This is one thing to do along the way to a redesign, but it alone is not a good reason.  (These reasons pile up though, as you will see)

We want to use some new technology (usually a content management system)

This is a big one.  It's worth noting that a new CMS doesn't necessarily make things any better from the visitor's perspective.  It's really just about making it easier to post content.  Yes, you may post a little more content than before, but ultimately a successful CMS integration project is invisible to the web visitor, which means they get little, if any benefit.

We want to create a more engaging experience for the user

Probably the most powerful thing you can do before you redesign is to understand what the user wants to do, but in actuality, people go by their gut instinct.  That's not necessary, of course, and the way to do this is to survey your web visitors.  Assuming you don't have a lot of money, it won't be scientific.  However you are not in need of scientifically accurate results to understand what they want, you just need to ask a lot of them.

So how do I learn what the user wants?

There's three tools/practices I'd recommend that aren't going to bust your budget:

1. Random sampling of web visitors for task analysis
2. In-depth telephone/in-person interviews
3. Web-based surveys

Random sampling of web visitors for task analysis

This is "stupid easy" to do.  You want to know the answer to a simple question:

"When the user was on my site today, what were they trying to do?"

Not a hard question.  The easiest and cheapest way to do this is to install the now ubiquitous tool 4Q.  (For a crash course in using 4Q, read my article "How can I survey my users and keep tabs on their website satisfaction for free?")

Once you've done that, you're going to get a large spate of responses to the question "What did you come here today to do, and did you accomplish it?"   Here's a collection of responses to the survey I keep on my website, truthypr.com about why people came to the site today:

This graph said that 80% of the people that came to my site in April were looking for a specific article.  Here's the text of what they said, which I think is the most valuable:

Ah-hah!   While these seem a little like platitudes, the first and third comments suggest that I need to continue working on my writing to carefully translate technical concepts into non-technical explanations.  The third comment emphasizes that frequency is important.

Alone, one or two comments is not anything to base a future decision on, but it does reveal that I should be asking my users "how often do you expect to hear from me?"  "Is weekly enough or not enough?"  And "what tools are useful in communicating these concepts?  Screenshots?  Videos?"

That last comment suggests that when I'm thinking about verticals, I should think about nonprofit charter schools, which is something that never occurred to me before.

Here's another one from a previous month:

The last one emphasizes an "arts nonprofit" vertical.

Once you've got a flavor for why people there, you can formulate questions that will begin to tease out whether or not you're missing something in your offerings, which leads me to...

In-depth interviews by phone or in person

These are the most valuable information you can get.  You want to make sure you get people who are honest and candid with you, which often takes an in-person interview if that's logistically possible.  You're interested in the initially obvious questions:

1. What do you like about the website?
2. What parts do you use most, why?
3. What frustrates you about it?

But then you want to move onto other, more important positioning questions, such as:

• What other organizations in this space do you follow?

I want to stop there, because the value in this is not to simply write down the name of a competing organization and move on.  You already know who your competitors are, and what your interviewee is going to say when you ask that question.

You want to know why they follow them.   You want the interviewee to grade both your organization and the competitor organization on the common mission, and then learn what they're doing better than you.

And then steal those ideas and improve upon them.

This is the most valuable thing you can learn, and will often go beyond the website's execution and to the way that your organization actually conducts their work.

Another question you should be asking is this:

• If you ran organization XXX with a mission to do YYY, what would you do?

You need to let the participant think big, and approach the mission focus with fresh eyes.  If you get a pat or political answer ("Just what you're doing today") then you need to find another participant.

Web-based surveys

I'm not a big fan of using these to get substantive free-form data from the visitor.  Whenever I read raw survey data (yes I am a nerd) I feel like the person was just filling out the form as fast as possible to get it over with.  That's not valuable to you, and it annoys the user. 

And what will I learn from this study of the user?

I suspect that you'll learn something like the following things:

1. Our competitor has a program that resonates better with supporters as being more mission-focused than something we do.
2. Our good work is not being explained well enough, and we need to do a better job of that.

Both of these things can be solved with business changes and editorial changes, but neither requires a whole site redesign. 

Wait, if learning what the users wants to do might not lead to a site redesign, why do agencies always make it the first step of a redesign plan?

Because web agencies like to sell redesign projects and clients like to buy them. 

Redesigns are big dollar engagements and make great portfolio fodder.  Also most sizable web agencies employ a marketing intern whose job is to submit client work for awards, and web redesigns are great award show material, which makes for great publicity.

Unfortunately I've yet to meet a digital agency that does their research and says "You don't need a redesign, let's give back that money".  Usually the agency concludes that the visitors are desperately craving a "better site" and charge full steam ahead with the redesign.  The client is covered from criticism because the research proved they needed a redesign, and the vendor feels justified as well.

If you went to the doctor complaining about a pain in the butt and he said, "We're going to run some tests but regardless of what they say I'm going to start planning a round of chemo for colon cancer", you'd freak out and go find a new doctor.  Why don't you hold your web agency to the same standard?

Do the research, figure out what the user wants more of, and then execute that.

[This is the first part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]