The problem with ‘Which UI is better?’ posts on social media

I bet you've seen social media posts from an author, a designer who posted two slightly different design versions of a UI solution they created and asked their followers to vote on which one they preferred. That’s a cheap way to gain attention that’s more harmful than useful.  

These posts are usually quite popular, with hundreds or thousands of reactions. I’m writing about why you shouldn’t share a post like this or vote on such posts. 

1. The post doesn’t give any context or cultural aspects for those voting or any explanation that would help you, the designer, as it is just a vote.

2. Deciding between two visuals without context is subjective and more about personal taste.

3. Making design decisions is not a sympathy vote.

4. Voters don’t understand the requirements or what users need

5. Voters can’t ask questions, and there is no reaction emoji for "it depends."

6. You don’t allow voters to think about the function and usability by providing them with data. They can just rely on their personal preferences, which doesn’t really matter. 

7. It puts back the old ‘graphic designer hat’ on you that you fought so hard not to wear. I’m talking about when designers were just simple executors, and an art director was constantly hovering over their shoulders and telling them which direction to go instead of relying on the actual users. 

8. Asking random people’s opinions on social media is not A/B testing. It is also not usability testing. Let the user tell you which works best for them.

9. Aesthetics are not usability. You might like the visual appearance of something that is not usable.  

10. It creates a bad habit. Other designers may ask a user or a client which UI they prefer. This is the worst thing you can do when you test your work.

There are far better ways to gather feedback on if an actual design is a good direction. For instance, set up a quick meeting with your colleagues and invite them to a brainstorming session, where you can share why you are choosing between two solutions by giving the proper context. They might devise a possible third solution better than the two you show them. You can also test the design with actual users. Only a quick interactive prototype is needed, and it will provide much more context than simply social media image sharing. Otherwise, you can build both and properly A/B test them, then conclude from actual funnels and usage data which one performed best and achieved the original goal of the design.