From 120-150 How to stall your growth on purpose like an idiot
Usually, by the time I’m writing a post on my business, I’m already up 10+ recurring paying users compared to what my post says. This is not the case with this post and it’s 100% my fault…
I promised to share good and bad and today it’s mostly bad.
Growth for me is getting more recurring paying users than losing them but this month, I lost almost as many as I gained, meaning my growth virtually stopped. But let’s start by the beginning to understand what happened there.
After my last post, I had a big influx of new accounts and I started to have too many of them (I know, a good problem to have) so performance issues started to appear. It was clear that the “old” architecture of the software would not scale to 1,000, let alone 300. I knew that quite well but didn’t think it would matter just yet.
At that time, I could have thrown more power at the problem (by increasing the cloud ability to process more web requests) but like any good engineer, I preferred to solve the problem properly.
So I revisited how things were done and identified a simple area where I could solve 80% of the problem in one sweep. Live updates from the server were putting a heavy burden on the database due to people with lots of pending follow ups. This could have been solved easily with a few changes and a simple service like Pusher. They happened to be a client of mine and I really liked them so I happily signed up for their service (and don’t regret it for one sec, good job guys!).
So far, so good, but here comes the rooky mistake:
Rather than fixing just that, I experimented the new technology in a single page application (no need to navigate pages on a site, so all changes happen very fast) and like it so much that I started expending the single page application to have all the features that the old interface had (rather than fixing the old interface).
I figured that since I was acquiring users almost on auto-pilot at the moment, I could work a bit on polishing the product. My thoughts were that this would slow down my sales efforts for 2 weeks but that should be manageable.
The move to fix the performance problem and move to a single page application at the same time was however just plain stupid and as a technical team leader I would never have allowed my directs to do that, yet I did it and not surprisingly, it proved to be a bad idea.
For a start, I still had support to do and this meant I could not always dedicate all the time I wanted to work on the product. The product development dragged and turned out into a massive rewrite of the web pages as well as how browser communicates with the server. Meanwhile performance could not be addressed on the old interface.
The effort took me 6 weeks in total (and I still didn’t nuke the old interface just yet as I write these lines because I need to move one last feature to the new interface). It did end up being what I didn’t want: a complete rewrite of the application layer.
A LOT of things can happen in 6 weeks (QuickMail is just 6 months old), and many happened.
One thing was that my predictable revenue outbound email campaign ran dry. My own QuickMail cold email campaign didn’t have enough prospects anymore and I needed to re-populate my group. But since my interface was always supposed to be finished next week, I neglected my revenue generating channel foolishly. Not surprisingly, sales started to slow down.
During that month, I also happened to experience a higher than usual churn (people cancelling), hard to know if it was due to the interface, most mentioned they changed direction and didn’t need cold emailing any more. But it took its toll.
So, in the end, this is the result I got:
– A freaking awesome app, super scalable offering visual feedback on each action and super fast to operate.
– Less support request (as most common support request on the old interface have been factored into the design of the new interface), and easier to reach out to me when you need help.
– A stalled business growth, 4 weeks of missing opportunities, higher churn, lost potential customer (as wizard for on-boarding has not been migrated to the new interface yet), lesser focus on users (I so wanted to finish this interface as fast as possible), and an empty sales pipe.
Pretty lame results if you ask me…
Sure, this brings me enormous technical advantages (I should be able to scale easily now, and I have less support requests… although old users needs to find their way back as I moved quite a few things around) but that was just not the right time to do it. I knew it but failed to see that I could separate the Pusher initiative from the front end rework until I was quite deep into the changes. The move to single page application (with AngularJS) to speed up user interactions should have been done as a separated task later down the road.
Technical note: small note on AngularJS. If I had to redo it, I would probably not use AngularJS again. I found the framework awesome to work with but they broke all trust of a technical long term good choice with the announcement of 2.0, breaking significantly from 1.x. My deepest condolences to anyone who bet on learning intensively AngularJS for their career future and I would not recommend going with a framework capable of saying F*Off to their whole user base with a big smile on their face.
All in all, I’m happy it’s done now, because I can now concentrate on resuming my marketing + sales effort as well as spending more time helping my users be successful!
Can’t wait to see what the new post will contain…