Be aware of assumptions
One of the most dangerous things in product design is blindly trust that the user knows what they want. Henry Ford probably would’ve been busy training horses in order to make them faster if he’d listened to his customers.
If I had asked people what they wanted, they would have said faster horses. — Henry Ford
Users often don’t know what they want, but they do know that they’re struggling with something. As we all know, it’s really difficult to follow what all your users are doing, so if you receive feedback you treat it like it’s really special.
During my graduation internship at Bundle I learned this by having a very motivated mentor (thanks Pieter-pleun) who saw the product designer in me. Bundle is a startup and startups cope with a limited amount of resources, resulting in very difficult decisions sometimes. To make these decisions easier it’s recommended to validate your assumptions and intuitions as soon as possible, and back them with data or feedback from users.
Most of the experiments I ran at Bundle were not necessarily connected to the main purpose of the app. But it would help the app grow to be a serious mobile app.
I got used to validating assumptions and it became part of my design process. So much that it turned out that I went a little too far and made a poor decision. This post outlines that decision and I hope that those who read this will learn from it.
For those who don’t know me, I’m the guy who designed Iconjar. An app for Mac OS that allows people to store their icons and logos in one place, just one click away. Icons are automatically tagged by their name and we allow people to add their own tags to make them easier to find. Therefore search is the one of the most important features of our app.
Redesigning the way search works
We received a few complaints about the way search results were shown. They used to be ordered by their set, which resulted in a very long list of search results. Some sets contain four icons that match the search criteria, but some only contain just one icon. We understood where people were coming from, we assumed that they just wanted to quickly search for an icon, drag and drop it and continue where they left of.
Based on this assumption we changed the way search shown its results. No more ordering by set, just a list of search results. Awesome, right? Not really. We got new complaints from people who actually liked the way search used to show its results. I cried a little, maybe a lot, but soon realised what went wrong by talking to our users. Our app is used by different kind of personas who all have different goals. Some people have a selection of icon sets they really like and some just want to put an icon in place. Our assumptions were wrong, because they left out the different personas. One size didn’t fit all.
So what’s next?
It would be easy to just revert back to the way things were and act like nothing happened. We didn’t want that, because that would be too easy and some people liked the new search too, hence the different personas. Based on the different personas we decided to let people choose which way they prefer search results to be shown. We chose to set “order by set” as default because people are familiar with it, and optionally allow people to combine all search results. Example: https://vimeo.com/187064866
In the end this mistake lead to great new findings about our users, but I don’t like the way we learned about them. We could’ve lost a lot of users, but thankfully we didn’t and released a new version within the same week. Lesson learned (yes I’m a huge It’s Always Sunny in Philadelphia fan).
Interested in trying out Iconjar?
Download Iconjar | Follow us on Twitter | Like us on Facebook
