Customers Can Ruin Your Vision, But They Don't Have To (Part 1)

Hi entrepreneur, I guess this one's for you.

When you take a brand new idea and decide to make it into a reality, just about anything is possible. You have an unadulterated image in your mind of the ideal customer who will love exactly the thing you want to build. Of course your opinions will be seen as visionary and spot-on. How could you not know what you're talking about? (Side note: founder hubris is one of the reasons a Product Validation Tour is so important)

It's important to have a strong vision. Here's hoping
Paul Bettany can pull it off (too obscure a reference?).
Then one day you finally get a real customer (for Instructure it was BYU-Hawaii, and we drank mango nectar and wore leis to celebrate). They're your first customer ever, so you take a picture of their check and you listen very closely to what they have to say. They are well aware of their position of influence and push you hard to implement their pet features because holy cow why wouldn't you?

This is your first big test of vision (second if you count the product validation tour, but if you did that right it wasn't much of a test, more a collaboration). For us, we told our #1 that we were aiming high, and while we were grateful for their business, we needed to make sure we were implementing things that worked for at least 80% of our customers. They cheekily responded that they were 100% of our customers :-P, so what was the big deal? BYU-Hawaii is not the bad guy in this story, by the way. Customers aren't bad guys.

This was the Test of The Special, and it's a tricky one to handle. The Special is the extra feature that doesn't really fit into your vision, but which at least some of your customers are telling you they need in order to do business. If you want to build a consulting gig then embrace the Special. If you want to build scalable software then the Special is very dangerous. Not evil, just dangerous. Especially early-on.

When your product is young it should have a very well-defined trajectory. That whole founder hubris thing, where you know exactly who you're serving and what you're giving them and what features you're going to add when, that's actually kind of important. You make darn sure it's valid and that there are enough early adopters to get some critical mass (ahem, PVT), but then you stick to it. If at this point you're already spinning off into side-projects and Specials, then you've got a vision problem, and you're going to end up wherever someone else pushes you. There's a verse in the Bible that covers this nicely, actually:

"For he that wavereth is like a wave of the sea driven with the wind and tossed." -James 1:6
Adding extra appendages can be dangerous.

Know where you're going. If you build a third arm that you're not excited about but that you think your customer might need to be happy, remember that you're going to be maintaining that arm for the rest of forever. If it glitches, you have to stop working on your laser-guided vision processing chip and go rewire that stupid arm. Maybe you can't get out of building it, because maybe your vision could use some adjusting, and if that's the case then be very clear with everyone (especially internally) that the change you're making is a reshuffling of priorities, and that other things are going to slip but that this is important enough to do it anyway. In the future you'll get big and changing directions will be hard enough on its own, but for now you should introduce a little artificial friction toward change because it will force everyone to stop and say, is this really the right move for us?

If you decide to build it you'd better figure out a way to make it an awesome part of your vision. Because if you're not excited about it, your team won't be either, and it'll be a second-class citizen in your system and your customers will notice and resent you for it and you'll be worse off than if you'd never built it in the first place.

For us, the thing BYU-Hawaii was asking for was a video discussion board. Some other edtech tools would let you create either a text or video/audio board. BYU-Hawaii needed video discussions as part of whatever LMS they were using -- and for good reason, their distance ed program was serving students on different islands, regions, and continents with all sorts of bandwidth constraints and time zones. But we couldn't afford to maintain two different discussion tools, and frankly it felt gross and weird to have multiple tools that served almost the same purpose.

So we had the aforementioned chat with BYU-Hawaii. Sorry, we're not building two discussion systems. We didn't want to blow them off completely, so we talked to them about the plans we had for discussions and also grading. We outlined our vision with them and said, "there's got to be a way we can fold this into our vision that is as exciting for us as it is for you." This was a critical piece of their educational process, so it wasn't something we could lay away for future work, which meant it needed to be something we could get stoked enough about right now that we were willing to delay something else to make it happen.

We batted some ideas around and in the end what we proposed to them was a audio/video widget that could be used to drop media into text-based discussions as well as other parts of the system. After talking to BYU-Hawaii we looked for how we could fold their need into our vision since it wasn't going away, and it turned out we were actually really excited about the prospect of "video-everywhere" to improve the learning experience, it had just been a little farther down the path. Implementing it was going to take longer than they wanted, but they now saw the vision and agreed with it, and there were some feature-parity elements they were willing to let slide to help make it happen, so it worked out nicely for everyone.

If we had just built what BYU-Hawaii was asking for we would have been taking a very large step down the path that we resented other edtech companies for taking. Subpar one-off features is pretty standard fare in edtech, and the whole reason we started Instructure was to not be That Guy. It wouldn't have ruined us as a company I don't think, multiple discussion types would have been a bother and not particularly innovative (video-everywhere, it turns out, is now a huge value proposition for people thinking about switching to Canvas, because it's so tightly integrated and easy to leverage), but we would have survived. Just not with the same vision.

Sometimes the solution is 12 moves away, but it's there.
The point is, you HAVE to start with a strong vision, and think hard about any time you're asked to stray from that vision. It may be ok, but it may not. And you have to get creative. At the end of the day a strong product visionary will look at customer demands and trace out in their mind a variation that could fit in with the existing vision. Maybe it's right there ready to go, maybe it's a few steps down the path. But somebody at your company should bet getting a rush out of trying to find that overlap, and then helping customers see how their needs can serve the "greater good" as you head toward that awesome thing of the future you started out building in the first place. And most of the time you should still probably say "no". Figure out where it might fit, then decide if it's worth it. BYU-Hawaii is one of the examples where we adjusted course, but those were the rare exceptions.


Unknown said…
I thank BYU Hawaii (and you guys) every time I use Canvas to teach a class!

Popular Posts