Planning for last minute UX changes

Over the past few years as a Product Manager, I’ve run into several instances where I regretted a UX decision just before shipping. When you’re working out of Google Docs, Balsamiq, and Photoshop, it’s difficult to get a ‘feel’ for the final product until the engineers actually create it — thus, it’s hard to be confident about some UX decisions early on. There are obviously ways to alleviate these issues, such as through prototyping software and studying similar products. But at the end of the day, they may not be enough. I recently learned that even the best PM can’t make all the right calls from the start. The best PMs embrace and plan for these issues so that they’re not thrown off later.

Let’s say I’m designing a car. This is a low end luxury car, so keeping costs low is important, but I want it to feel very good all around. The engineers come to me and ask, ‘what material are we using for the interior accent trim?’

Well uh…we could go with leather…or texturized rubber. It’s kind of hard to tell. The mockups kind of look the same. It’s nice to touch the small square samples, but it’s hard to imagine the final experience from these squares. I guess we need a decision…okay…fine…let’s go with texturized rubber.


In your product spec, create a table of decisions that you’ve had to make where changes may be required after the product is built. Add this particular decision, and keep going. As you keep coming across UX decisions where your gut is queasy trying to decide, add to this table. As you work with the engineers to plan out the project, add stories or work items to account for changes to about 1/3 of these decisions. You should be pressured to make the right decision from the get-go, but you will have some time to make changes later on.

Now, instead of getting that sinking feeling of being wrong just before shipping, you should be ready for design changes when picking a ship date. If you have a hard deadline, you can at least do creative things like add another engineer to the project, or cut features. This method will hopefully eliminate those sinking feelings you get when an engineer has followed your spec, but you realize that — oops — you were wrong!

Have other ways to account for last minute changes? Add a comment, or tweet to me at @cgallello.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.