A couple of years ago, I was doing some work for a Series A funded, double-sided marketplace. One of the things that’s fun about staying hands-on with design work and writing is that you have ample opportunity to keep learning about people and technology.
The main way this startup recruited new customers was through a sign-up wizard. You know the sort of thing: an if-this-then-that ordering of questions leading up to a sign up or portal. You can build them pretty easily in either Typeform, with Intercom or with pretty much any other SaaS product you like.
Here, they were asking the sorts of questions they needed to know up front. On the face of it, this makes sense. That information was critical for their internal workflows, and they wanted to get to the point quickly. Speed and ease of use is important! But it’s surprisingly easy to get wrong: an early sign that you’re more familiar with the inner running of your organization than you are with how your users think about the problems you’re trying to solve.
For what it’s worth, it’s also the sort of thing you see in sign-up flows where the Marketing Department is looking for a lot of demographic and behavioral data. Or in public sector portals where you’re asked to give vast amounts of personal information for a task that really only needs a couple of names and a few contact details. (Yes, LA Parks and Rec, you need to know if I’m above 18 years old to register for a pottery class. You don’t need my exact date of birth.)
It’s not just about data minimization, though. It’s also about question order. Asking questions in a form in the wrong order causes people to drop out of the funnel even when you’re not asking questions that are too detailed. Every product and growth marketer I’ve worked with in the last 5 years has been obsessed with the Noom sign-up flow. Beyond looking great, and having great holding pages, it doesn’t ask you your weight up front. Smart.
We can think about these questions in terms of intimacy. Which questions do people feel comfortable telling strangers? Which ones do they keep to themselves? I’ve been in countless user research sessions about designing forms where this comes up. If you’ve ever worked in a call center (I have, I was very bad at selling phone subscriptions), you’d know the order roughly coincides with a foot-in-the-door sales script. It’s also a version of progressive disclosure, a design pattern that’s almost old enough to buy alcohol in the state of California. All of which is to say that the thinking behind this is well-documented, but getting it right can still be tricky.
You can finetune your question order with user research if you want to. It’s especially useful to do this if you’re working in new-to-you cultural contexts. For example, people in the US seem much more comfortable adding their phone number to a web form than people in the UK are. Plus, just on a practical level, it’s an easy piece of user research to set up and run. Who doesn’t love a card-sorting exercise?
If you are a bit uncertain and don’t have the time or budget to do a lot of research up front, think through a conversation order:
- What’s your name?
- What’s your email address?
- What do you want to achieve?
- What do you do/like?
- [Plus questions specific to your service]
- …and then the big ask
Does any of it feel weird if you say it out loud? Change the order. If you don’t have time/budget/permission to do a card sort, ask people in the office or who you know. If you have time, do this with real users. If you work in a regulated industry, always check with legal. So that you’re not wasting lawyerly and/or data protection office time, know which, if any, questions are regulated as personal or sensitive data in the jurisdiction you’re working in. For instance, even though California is known to have strict data protection rules, asking about political or trade union membership is less strictly regulated than in the UK.
For this startup, I changed the order based on experience and we had quick analytics to show our results.
By doing this, I managed to increase the relative sign-up starts for this company by 157% (a 12.4% absolute increase). It doubled the conversion rate, too (2.6% to 5.2%).
Yes, that meant that there was still a drop-off further down the funnel. That’s often the case. People get distracted, realise it’s not for them, plus there’s always room for fine-tuning how and when you ask questions. Starting off on the right foot, though, can help you make your funnel big enough that it offsets these drop-offs with just one simple trick.
For public sector design, this matters because we’re designing services for people who (often) have to be there. When we ask questions in the wrong order, or in the wrong way, it’s not just a time tax. It’s also a trust tax. Our users may fill out the portal, but they do so resentfully and unhappily, with the sense that the machine and the man don’t care. And really, it’s an easy fix.
* Forgive me, friends, for the click-baity title: it struck me as funny at the time.