Do Nothing To Confirm
I wanted to call my text “Don’t Make Me Click” but then I googled it and found this great talk by Aza Raskin, pretty much on the topic I wanted to explore. I almost resigned from finishing the article but finally, I wrote something shorter but still a bit related to Raskin’s ideas. For better understanding, watch it before you go further. You will not regret.
When you analyze a lot why users leave a flow without doing actions you expected them to do, you may agree that taking any action is preceded by a mental struggle. The best interface is what Brenda Laurel calls NFI – “no fucking interface”. You give users what they want immediately as they approach. This is something Raskin elaborated well in his talk.
The class of problems he didn’t mention, is when it is you who want users to do something and you cannot remove making a decision completely. Here are some good news – between doing nothing and making a strong statement by clicking a button, there is a middle ground of “acting by not doing”.
Great example are Kudos – invented by Dustin Curtis feedback mechanism which may be explained as kind-of an alternative to Facebook Like. Kudos work without a click. Move your cursor over the button and wait… done!
Some people find it a bit cheaty, the action cannot be undone – but hey! – repercussions are none so you probably don’t care. The barely perceptible effort fits perfectly a low importance of the decision.
It’s like an old good “say nothing to confirm” when you want to get a confirmation about something people don’t want to talk about. You have your confirmation, they have their alibi (at least they didn’t say anything).
Probably the most popular example are Youtube videos which load 10 seconds after the end of the previous watch. In fact, I’m not a fan of that implementation. It implicates making decisions by user when they may be away from the screen. All it misses is a recognition if anybody is actually watching it. But I believe sites asking for the access to your camera device are coming.
Recently, I worked on a simple feature allowing users to sent their current location to somebody from their contacts. I wanted to make it as simple as possible but removing some kind of confirmation completely wasn’t really an option. Especially on a touchscreen, you should be able to cancel an accidental touch. The process was already 2-tap long and I wanted to keep it that way. Here’s what I did:
You can argue with that, but a few seconds for canceling works enough for me and the flow is still 2-tap long.
It took me another week to find that I can skip one more tap by opening the popup automatically when there are no other actions possible. Initially, I wanted to start on the map view because it gives users a context for what they do. For now, I ended with the app starting at this intermediate state:
From here – with a vertical swipe – you may go either to the map or the fullscreen contact list view. But connecting your latest contacts takes only one click without compromising too much.
Let me know if you know some other examples of that “passive confirmation” method.