Promoting UX in Your Organization: What works?

A story on Userfocus today, Usability ROI: 2 measures that will justify any design change, recommends two techniques for convincing management to pursue user experience deliverables in their projects.  Both involve quantifying, in dollars, the value of the change.  For example, a redesign of our eCommerce site improves sales by 5%, which results in X dollars, or we shave 15 seconds off a common task performed by employees on an intranet, which results in Y dollars of savings.

I haven’t found this to be universally true.  I did something like this with a client.  “Look…if we redesign your call center application like this, we should shave 10 seconds of each call, which will result in $6 million/year savings.”  I thought it would be a slam dunk.  Not so.  They said, “Thank you very much for that insightful analysis, ” and nothing ever changed.  I was stunned.

Looking back, I think there are two reasons why my argument didn’t fly, and why you might be cautious with the advice in the previous article.  First, those ROI figures sound unbelievably large.  I know the math is solid…so do you.  But I think Senior management hears those types of numbers and think, “yeah right…there’s no way moving this field from the left to the right will double my sales.”  Second, the user experience of the product is rarely the only consideration in a project, and even more rarely the most imporant consideration.  Managing a large web project is hard, with lots of people, teams, resources, stakeholders, and some inflexible and unreasonable deadline.  There may be absurd-but-nonetheless-real reasons why the manager can’t act on your recommendation: the technology platform has already been selected, the CIO said it has to be a certain way, the project plan has already been set, etc.  Plus you’re telling some grizzled project manager who’s learned to rush through requirements to maximize development time that you’re going to delay development while we extend the design process.  Tough sell.  (Yes, I know we would suggest we do agile/iterative development instead of waterfall…I’m just saying not everyone in every company is there yet)

I think Jim Nieters and Laurie Pattison said it well in UXMatters:

A senior VP once said that the cumulative savings all the teams proposed, telling him how much he could save by implementing their ideas, would add up to more than the revenue his company earns, which is clearly not reasonable! You can’t save more than you earn. Senior leaders frequently consider ROI arguments funny money. The issue of value really comes down to personal accountability. Remember, that trust thing: If leaders trust you, they’ll listen. If not, they won’t.

It’s not that ROI arguments never work. In fact, it’s useful to have some of the best ones in your back pocket to pull out if someone asks. It’s just that usually when someone asks What’s the ROI for user experience?, what they’re really asking is Can you give me examples of what you’ve done in the past? If you can generate a portfolio of your UX group’s wins, you’ll be in a strong position, because your arguments are no longer theoretical. And that’s the main issue. If you can show what you’ve actually done, executives can see practical value and extrapolate how you can help them, too. Don’t talk about what others have done. Show what you have done.

I don’t think you can rely on ROI arguments to sway everyone to your point of view.  You need to think more in terms of a seige than a single strike/snipper shot.  I think first, you have to establish your credibility in the organization.  To that, you have to show them some actual results that you have achieved within that organization…none of this, “it worked well when I was at my old job.”  Second, you have to be extra transparant on your work — how you do it, how long it takes, and when you’ll be done.  Don’t wait for someone to come and ask how you’re doing…be proactive in your communication to the project team.

I think UXMatters has a number of good articles on promoting and evangelizing UX: