13

Misleading Metrics and UX Tradeoffs

 3 years ago
source link: https://belkadan.com/blog/2018/04/Misleading-Metrics/
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
neoserver,ios ssh client

Misleading Metrics and UX Tradeoffs

Imagine you have a search feature, and you find that basically no one is using it. That is, the number of loads you get on the dedicated search page is tiny. You make a tweak to your UI, and suddenly way more people are going to the search page! Sounds like a win, right?

Except…the tweak to the UI was that previously you had search results loading inline in the search bar, and now it’s just doing completions. And your product is Facebook, which means that most searches are for people or pages that the searcher already knows; they’re just using the search bar for quick navigation (like Spotlight or the Start menu). And this change happened a year or two ago, but it still annoys me every time.

When I search for “mark” on Facebook, I used to be able to jump straight to one of my friends named Mark. Now, Facebook still completes their name, but takes me to a search page first if I choose that completion. I have to click on the first result to actually get to the friend.

(here's what that looks like)

What’s worse is that they clearly still have the old UI, where you can jump directly to a person or page, because that’s what they use for Recent Searches. (You can see this by clicking in the search bar and deleting all the text; no image here because blacking it out would mostly defeat the point.)

I called this post “Misleading Metrics” because I felt like this had to have been a metrics-based decision. It turns out I was wrong. Facebook did do qualitative user studies as part of making this change. But the design they settled on clashes horribly with how I actually use Facebook, and likely many others as well.

My original takeaway was that metrics give you numbers, but they don’t give you reasons, and you should always try to think of multiple reasons why the numbers could be what they are—and even to mix your metrics with some in-person user studies and qualitative feedback. That’s still a good lesson, but maybe I need a better example.

(I’m picking on Facebook because I use Facebook1 and because they’re big enough to stand this kind of critique, but this is the kind of thing any company could do. Also, feedback submitted via https://www.facebook.com/help/feedback.)

  1. Please don’t try to find me on Facebook unless we’re friends or coworkers offline, or at least Twitter mutuals. Separate persona. ↩︎

This entry was posted on April 29, 2018 and is filed under Technical. Tags: User experience, Rant


About Joyk


Aggregate valuable and interesting links.
Joyk means Joy of geeK