Take the 2-minute tour ×
User Experience Stack Exchange is a question and answer site for user experience researchers and experts. It's 100% free, no registration required.

We've been working on/off on an account/bar header for a large corporate site for nearly a year now.

We've got so many stakeholders that nobody can ever agree what should go there!

What tools/ways/arguments do UX pros have for sorting this stuff out!

We have quantitave stats on header links useage, but many of those could be false positives (ie, if you put a link in the header, then people are going to click on it!).

BTW - I'm doing some usability testing on the Beta site on Thursday - can I use this to somewhat get some clarity on the situation!

share|improve this question
2  
Card sorting might help –  Phil Dec 13 '11 at 16:15
2  
Is there disagreement on what the business goals are, or is it disagreement about how to achieve said goals? If the latter, you've got a good chance to introduce "science", but if the former, you have bigger problems to solve first. –  peteorpeter Dec 13 '11 at 16:37
1  
Beat them with a bat and run off with the money. Ahahahaha! No wait, something constructive. –  Ben Brocka Dec 14 '11 at 2:23
    
Sometimes, the battles aren't worth it. It is up to you to decide whether this one is. –  Schroedingers Cat Dec 21 '11 at 13:56
    
Unfortunately, we have disagreement about what the business goals are - and various stakeholders clamoring to see their areas represented. –  Richard Payne Jan 5 '12 at 7:39
add comment

4 Answers

up vote 6 down vote accepted

I think its more about politics and people than about statics and testings. Nevertheless those give substance to your arguments.

In mettings at presenting your design:

  • Keep in mind what history and interests your colleagues stand for. Be prepared for it and have arguments for that.
  • Identify who is against your arguments and why. Sometimes its best to have a single meeting and presentation with them before. so you can go more into detail. Take their concerns serious, this can work as a dooropener.
  • Invite stakeholders to your UX-Testings, so they can see how people struggle with the actual design.
  • Invite stakeholders to Design Session and get a Buy-In. Nobody would argue against own ideas - (actually you can spin ideas in right direction during the Session).
  • Try to get buddies. Make favours for some and get favours back.

More tactical:

  • Step back on some minor points and get your big point win.
  • Present minor designs first, so people can get their heads dive into them. Later or last present your most important design - people havent much energy now, got some winning fights and are more likely to wave your design through.
  • If nessesary make a special meeting for a very debated/unclear point. So the meeting is going on and not disturbed by just a single tpoic.

Your arguments/points can be supported by:

  • UX Testing Outcome
  • Interview quotes
  • Analytics
  • KPI improvements (if numbers matter)
  • ROI on new design (if numbers matter)
  • Sketches
  • Prototypes

And here is a good article about how to manage and prepare this kind of decision-making meetings at Uxmag.com.

Selling UX at UXmatters

Communicating the UX Value Proposition at UXmag

Where's the Value? Return on User Experience (ROUX).

share|improve this answer
    
Great answer, thanks Frank. I streamed a usability testing sessions which showed customers struggling with the current implementation, which has helped greatly. –  Richard Payne Jan 5 '12 at 7:44
add comment

Do A/B testing.

By getting direct knowledge of what works or what doesn't from the analytical data collected by the tests from your user's behaviours, data will speak by itself instead of each other's personal assumptions (both the UX designer's and the stakeholders').

It is important to stress that A/B testing must be done right in order to harvest correct conclusions, otherwise it could be contraindicated.

EDIT: UX Magazine published an article about winning approval in design presentations that perfectly fits the subject.

share|improve this answer
1  
Can you provide some more detail how Richard might apply this to his situation? –  Alex Feinman Dec 13 '11 at 19:20
    
@AlexFeinman I have elaborated on the answer. What do you think about the technique being indicated in this case? –  Naoise Golden Dec 13 '11 at 20:34
    
I agree A/B testing is great when done right. Alas, when used to convince bad management, one can assume bad management will also end up insisting on ruining any A/B testing. I've been in situations where I've had to prototype and test two obviously bad options. –  DA01 Dec 14 '11 at 20:29
    
by the way, that is an awesome article. "The best way to get a design approved is to let them think it’s their idea." is pure genius –  gustavofritsch Dec 21 '11 at 14:35
    
Thanks guys. I really (really!) like that article - a lot of the issues we face are around controlling the flow of a meeting, so that ideas and decisions can be fully explained before stakeholders scatter-gun questions or requests! –  Richard Payne Jan 5 '12 at 7:41
add comment

1) Establish users problem that lead to this project 2) Check if this problem was real (e.g. use event tracking in Google Analytics to check users paths)

If real: 3) Establish clear and measurable objectives 4) Check what solutions will be easiest and fastest to develop 5) Do Multivariate Testing. 6) Choose best solution 7) Check if this solution works really well 2 weeks after its launch using e.g event tracking in Google Analytis

This my work for you.

share|improve this answer
add comment

I just read a great article on this from UX Magazine:

http://www.uxbooth.com/blog/design-by-numbers-quantifying-the-subjective/

In short, map the task to audience goals.

If it doesn't meet a goal, then it doesn't matter what a stakeholder thinks.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.