I had asked a friend to try our signup process and he came back to tell me that he couldn’t make heads or tails out of what was going on? We had recorded the entire experience and the video was, as I would call such videos, “disturbing”.
The strange thing was that we’d had never been criticized on our UX. Existing users who had been using the system were happy with it. Some wanted cosmetic changes, a link here & there, but they weren’t facing problems in getting things done. In fact, one time when we had silently slipped in a tagging feature, it was picked up almost instantly. Discovery was never a problem for existing users. They knew what to do and how to do it.
However, most of these existing users had been given an initial training by us. So we really hadn’t tested our self-service getting started model. This exercise was all about the self service getting started model.
Nevertheless, here were some users complaining that they weren’t making heads-or-tails about what our application does. I should also point out that the user group facing these problems was largely new startups who hadn’t really used a CRM before.
Should we even be doing self-service getting started?
If you are ever asking yourself this question, you should read this post by Joel York. Self Service works very well for low complexity, low price point applications. And as an industry matures, the complexity levels are automatically brought down.
Self service activation is a very very important factor in scaling your SaaS application. Even if you are not there yet, you should continuously keep working on making things self-servicable.
What did not work?
We always had self-service as part of our sales model so it’s not like we hadn’t anticipated this problem. We had a running deck of clickable tips for our first time user. Our new user would see a banner with an arrow, explaining how to move around the page.
Remarkably, all of these users closed this banner without reading it. There was a big NEXT button and a small X on top right hand corner and yet, the user clicked the small X on top. Clearly, the tips had failed for us. To be honest, I still think it is more of a design issue than anything. We would need to work on the design of those tips to make them work.
Let’s ask the customer?
To figure out our solution (in true customer development style), we went ahead and started asking the user.
How can we help you get started on Yavvy?
We got a plethora of ideas, but they were largely around creating videos and manuals. There were some asking for a wizard, but the steps of the wizard were more around their own personal problems and solutions.
Going by the diversity, we’d have to build a wizard to help you select which wizard to use.
How was competition solving it?
They weren’t. We went back to our users and asked them to sign up on other cloud CRMs out there. Some of them were an even bigger disaster. Some (fairly successful CRMs) asked you for mandatory upfront money (setup costs) and assigned you a “specialist” who’ll help you get started (No Self Service at all)
Simpler CRMs were doing a better job on-boarding but did not have the necessary features to keep them interested. “Can it auto assign a new lead?” “No” “Well, then I’ll never use it”. It was as-if everybody was telling us that self-service does not work in feature-rich CRM-ERP applications.
How was the world solving it?
I went about searching applications which were considered easy-to-use. I started referring to my notes where users had said “I use this app called xxxx and I love it”.
Most of these apps were point solutions, doing just one thing! Moqups, a wire-framing tool, simply shows a “Wireframe” with most elements on it. There was text on it but I never read them. You could drag and drop stuff and get started.
Gliffy, which had more options than moqups, was a two-step process. They would ask which diagram and then you can drag and drop stuff and get started.
Basecamp, the project management app provides you with a dummy, prebuilt project that you can play with to see how things work. Trello, another awesomely simple application had three-list board and a clear link saying “Add Card”. No wizards in either case; the user was brought directly into the same screen that he’ll be using later.
Mailchimp, which had a multistep process, had a wizard. So did Shopify.
Unfortunately, I could only find out what these guys were doing, I really did not have any validation on whether it was working for them or not. So I called up a friend who had no idea what some of the apps did and asked them to try those apps out. The best of the best were failing. One thing was absolutely clear …
You can’t get a person started on your application if the user does not have a reason or purpose (to use your application)
And if you extend and extrapolate this idea, the goal of activation is to bring your user to a place where his/her purpose is met. The easiest to use applications are the ones that get you there quickly.
So what should we do to make our self-service signups work
Give your users the simplest, most-attractive UX that gets his work done (in the least no of steps). The solution to getting started is “Instant Gratification”. Can you by the end of the signup process, get your customer’s work done & also give him/her enough reasons to come back to you?