startups

As of last night Guestlist has officially completed our first month of business. To mark this occasion I thought I’d share a funny story with you.

In addition to cleaning up workflows and building great new features, we’ve also been adding more analytics to give us better insight into how people are actually using Guestlist. As I was working on this last night I stumbled on a mistake we’d made in rolling out the payment system.

When we started charging on July 12, we grandfathered all published events that our beta users had previously created but not completed. However, we also have this great copy feature to help customers who run repeat events. I think you can see where this is going. Copied events were inheriting the grandfather status and hence were not being charged commission!

As it turns out, a lot more customers were using the event copy feature than I had thought. In fact, this little oversight cost us about 20% of our first month’s revenue! Well, better learning this now than twelve months from now.

I only noticed this because I was investing time in understanding how people were actually using our app. Before this I had incorrectly assumed that not many people were copying events and as a result did not have that in mind when implementing our payment system.

So do yourself a favor and get to know your application usage data a little better. Not only will it help you build a better product, you might just learn something that surprises you.

In mid-July, after a month-long private beta among friends, we began discussing a release date for Guestlist. Figuring this out was easier said than done.

Not until we’re finished

At this point, now 7 months into development, we were dawdling. The team was spinning their wheels, building one “must have” feature after another in an elusive hunt for doneness. You can’t ship until you’re done, right?

A typical conversation around this time went like this:

“We can’t go live without hot swappable redundant servers.”
“How can we release without the ability to comp tickets?”
“We can’t run Guestlist without a way to bill our customers.”

This was just the tip of the iceberg. The list of things preventing us from finishing, it seemed, were endless.

Redefining the problem

Since we couldn’t decide on what constituted a “finished” product, we decided to release an unfinished product instead. We scrapped our billing system, and opted to go free until we had one. We removed references to unrealized features in our website copy, added some beta stickers, and kicked Guestlist out the door.

Even without a single user, we’d happily reached our first major milestone - shipping a software product.

Will it blend?

Our biggest fear with releasing Guestlist as beta? That no one would use it to host their one-time events. This isn’t the same as playing around with bug tracking software; your event happens once, and it’s got to run without hiccups.

Still, that didn’t stop users from taking a chance on Guestlist. I’m sure it’s been said before, but early adopters are a forgiving bunch. They’ve worked with us to help improve the app, whether by tipping us off to a bug or glitch, suggesting a new workflow or feature improvement, or just offering a vote of confidence. And we are grateful for it.

You don’t know jack

Real users means real problems - not the phony ones we’d invented for ourselves. Of the dozen features we’d labeled as “must have”, only a handful carry that label today, and much of those have been implemented since.

So, if there’s a lesson in all of this, it’s to stop spinning your tires and get your software out the door. Things might start off slow, but you’ll waste less time and build a better product for it. We like to think it worked out for us.