You all know that project...

After months of research, wireframing, prototyping, testing and workshops, there is that one client (normally the MD) that comes in at the end and knows everything better - wanting the site totally different to your thinking and those of your users.

Its an important client that you want to please, but at the same time, you don't want to put your name to a piece of work that contains massive usability issues.

How do you deal with these clients? How far do you go to appease them?

6 Answers 6


Quick anecdote: I was working at Intuit and redid the Quicken.com homepage. I added some icons and moved some stuff around. The marketers also changes of their own on their own version. Nothing goes live without extensive testing, so we all put our A/B tests into the hopper and waited.

My design won by alot, and made Quicken over $20,000 per month immediately. Then came the interesting part...

The marketer refused to promote the winning test to the site. She said, "I don't like the icons".

So what's the lesson? We live is a messy world with horrible people who are not interested in success or user happiness or even money. People are interested in their own egos. Personally, this is the worst part about knowing how to design. You know what will be successful, but we don't live in a meritocracy. We live in a bureaucracy.

I feel your pain deeply. One can not buy groceries on principle. You have to decide when to fight for what's right and when to smile and let the other person get what they want. Figure out what matters the most and everything else. Make sure other people decide "everything else" and you fight for what matters. I wish this answer wasn't so cynical. Hopefully, it was a little helpful for you to know you aren't alone.

  • Wow, I've had run-ins with marketers before but never this blatant. How sad!
    – Nathanael Boehm
    Apr 13, 2010 at 21:27
  • In my experience, marketing teams have a lot more power than any other. And, unfortunately, they don't usually have the users' needs in mind. (I've come to the conclusion that Marketing is the opposite of UX.) Nov 26, 2016 at 12:14
  • It's been 13 years, and this is still true Jun 18, 2023 at 22:17

I think the rationale for your output always has to refer back to the clients stated goals at the beginning of the project - "Our testing showed that users were more likely to buy if..." or "Research [quote evidence] suggests that revenue is likely to increase if you do this..." etc.

Ultimately what the user wants and what the client wants are one and the same. For example, the user wants to buy something easily and feel like they've had a good experience. The client wants the user to buy something and feel like they've had a good experience!


Yes Oliver I think we've all been there. Referring back to research always helps but I think in some cases you have to pick and choose the battles you want to win. Let the client feel like they are being listened to and have an input by conceding in areas that cause the lease amount of 'damage' - however much you hate it! But stick to your guns when you know that something is going to cause a major issue backed up with every piece of research you can find.

Ultimately they are the ones paying for the site; they own it and some think they know enough to make decisions. Warn them, educate them, and then let them decide for themselves. But make them aware of the consequences their actions may cause.


Another great tactic -- let THEM come up with the ideas. In truth, you probably went through a series of decisions to get to your final conclusion. When a client doesn't understand a recommendation that I've made, I like to start with their recommendation. The conversation usually plays out like this:

"Yes, I see exactly where you're coming from with that. In fact, that's something that I/we debated a bit on our side. Then, I/we realized..."

Talking slowly through the realizations that led to the current state often allows the client to pre-empt your suggestion and feel like they'd have come to the same one.

Also, you might take this as an opportunity to refine your upfront processes to protect yourself better next time Identifying who has final sign-off and who must be included at each juncture is useful, and the technique I've seen work best is the RACI matrix: http://en.wikipedia.org/wiki/Responsibility_assignment_matrix.


you need a good enough client relationship to look the client in the eye and say "you are off your rocker, trust me" and then hope to hell you have got it right


Came across this exact issue the other day, except it wasn't my idea I was trying to defend or justify, I was asked to act as an independent arbitrator to decide which of two competing designs for a design refresh would work the best.

The solution I recommended was A/B testing. This removes all the emotive subjectivity from the decision. I asked the business owner of the content what they would consider a measure of success, and also asked the designer what they were trying to achieve. It all came back to the call to action. The measurement of success will be what percentage of users visiting the redesigned page clicked the call to action, and then as a secondary measure, how many of those customers then went on to complete a transaction.

That's easy to measure.

The change is also small enough that two versions can be built. The test is going to run for about 4-6 weeks. Personally I think the designer's preferred option will win hands down. Then hopefully we can repeat the test and try out some different copy.

  • Really interesting - let us know what happens
    – Oliver Gitsham
    May 17, 2010 at 12:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.