In March 2013, we started building Ansa because we didn’t feel we had a safe place to share privately with friends. We designed it for ourselves, to fill the void in the current offerings of social networks and messaging apps. In September we launched Ansa on stage at TechCrunch Disrupt (video) and received our first 10,000 users from the proverbial “TechCrunch bump”. Even though our user base continued to expand, we knew we had to accelerate growth in order to succeed in this highly competitive space. We decided to embed ourselves on a college campus, where every day we continued to design, code, and talk directly to users. Here is what we learned:
Nobody’s perfect. Facebook’s mobile app still crashes when I use it, and they have a market cap of $100B+. If the big companies are still fixing bugs, it’s only reasonable a new startup would be too. By all means, aim for perfection, but don’t sit around waiting for it to happen, because it probably won’t. It was hard, but we stopped hoping people would discover and fall in love with the next version of Ansa, and decided to start talking about what we have now.
We knew we wanted to talk to college students, but we weren’t sure how to do it. We emailed, we surveyed, and we skyped, but nothing compared to the results we got when we left the office and started talking to students in person. Our first day on campus felt much like the first day of college. We were lost. Where should we talk to people? Who should we target? How do we approach them? What do we say? Is this even legal? We sat down with our notebook and wrote down all the ideas we had. One by one, we tested them, pivoted and improved until we found a routine that worked. It turns out a slice of pizza, a cookie, or a cup of coffee is a fairly reliable way to get someone to sit down and chat with you. The more conversations you have, the more likely you are to run into the people who really need and love the product you’re creating.
So you’ve found people who love your product. That’s awesome. If you haven’t and you are spending your income on Facebook ads to buy a “like” from people, stop. Those likes don’t mean much. Go back and repeat the first lesson.
Talk to the people who really like what you do and get them to test your product. Many entrepreneurs say “we do A/B testing with Optomizely and talk to our users with Olark.” Those are both great tools, but they don’t tell you what a person is thinking, or let them explain why they make the choices they do. It’s much better to get to know your best users, and earn their respect. The conversations you have with these people will help you discover which of your messages resonate best, and which details people really care about. Unlike chatting on Olark, or A/B testing, talking in person allows you to watch for tacit signals like eye rolling, laughing, smirking, etc and really understand what someone is thinking.
Oh, and when you do earn your early users’ respect, they will be proud to help and contribute to your product... and guess what, just like that, you get free representatives. Some of the users we met on campus gave us more than feedback. They came back, offered to to volunteer, and even asked to join our team! Others introduced Ansa to their own organizations, like student coops and sororities, and advocated better than we ever could.
Students are brutally honest, and I love them for it. They don’t have time to be diplomatic like we all try to be in Silicon Valley. The students we spoke with told us when they didn’t like our features (“that drawing was shit…you should make it better or I’m not going to use it”), and helped us realize some steps were unnecessary (phone verification for registering).
Before we built Ansa we planned each little detail, as if we were designing blueprints to our dream home. We thought our UX, information architecture, and flows were well designed, but as we spoke to people, we realized they were flawed. For example, sometimes we asked people to show us how they would share a photo. When they responded, we would ask them “how confident are you that is how you do it?” Their answers helped us realize what confused users, and uncover mental models and UX issues that needed improvement. It was amazing to see how many times users got stuck on the same problems we had missed. We repeated this over and over again. As we spoke to more people, we saw patterns in the responses, and used the data to tweak the product.
We found the things that people loved and even needed about our app, and began removing the parts that didn’t work (bonus: fewer bugs to fix now!) Lesson: Don’t spend a lot of time on wireframes and visual comps… they are probably wrong. It’s better to have a rough, blurry vision. Let your users point you in the right direction.
Ansa was founded by women, and designed with their personal experiences in mind. We assumed men would like Ansa for the same reasons we do, but once again we were wrong. As we spoke to students we realized how important it is to understand different users’ perspectives.
We stopped by fraternities and sororities to get feedback on our app. Men would often say, “I know a few people who really need this”. They instantly associated disappearing messages with scandalous action, and were greatly intrigued by the ability to take things back. When we spoke we got “oohs” and “aahs”. Women responded differently, and were keenly interested in the ability to go “off the record” and share more with friends. They would often look at each other, laugh, and say “we need this”. The reactions were both positive, but for very different reasons. When we figured this out, we dug deeper and spent more time learning about how each group communicates now, what problems they have with their current options (too fragmented), and how Ansa could help solve them. These talks taught us not to assume everyone thinks similarly.
Before we launched at TechCrunch Disrupt, we kept Ansa to ourselves. We didn’t think the app was ready for outsiders to test, so we kept it within the team and planned for a grand unveiling on stage. What we presented was nowhere near perfect, but we survived and ended up with tons of data to help us improve. I only wish we had user feedback sooner.
If I’ve learned one thing, it’s to start introducing your product to people sooner. If you’re planning to reenact Steve Jobs’ dramatic uncloaking of the perfect product, and hoping each person in the audience gasps in amazement, you are probably wasting your time. Even if it does happen, it’s not a winning strategy. The winning strategy is moving fast, getting exposure, and adapting quickly. The faster you can get feedback and iterate, the better.
We’re taking everything we’ve learned, and relaunching Ansa at college campuses. If students are in any way salient to your product, you should also do it there, and hope to get the same level of honest feedback we did.
Got it. Thanks! We'll get back to you as soon as we can. :)