Command

After about ~17 months of development and testing with a small group of users from my mentorship business, Command went live to the public. In its current form, Command is a tool designed to help small SaaS businesses manage:

• Planning their roadmap and sharing it with customers.
• Their support inbox.
• Monitoring revenue data.
• Performing behavior tracking.
• Host help articles and documentation.
• Customer data in a simple CRM.

I had intended to ship much faster (my initial goal was October 2019), but throughout the development experience, I ran into quite a few roadblocks. There was a lot that I didn't know and had to figure out on the way.

One of the trickiest parts being importing data from Stripe and figuring out how to calculate revenue data for customer's products. While I understood Stripe's API, it turned out that the data I needed for calculation had to be derived in a lot of cases. This meant writing custom calculations as well as scaling those calculations.

The first iteration of the revenue import I built was done directly on the same server as the app. This worked, but imports were taking several hours to complete. Next, I tried out cloud functions. These worked even better due to their flexible resources (and got the time down considerably), but limits on execution time meant that it would work perfectly for some customers and not at all for other customers.

Finally, I landed on setting up a separate job server and queue. This was a completely new experience to me and took quite a bit of time to wrap my head around. I ended up using the Node.js library Bull to help me manage queues which is backed by a Redis database instance.

Ultimately, now that I've launched, I've learned that my positioning of the product wasn't the best—it's not terribly clear who the product is for and I've gotten a lot of feedback around it looking "full-featured" but not really saying "hey, I want that!"

I did a write up (here: https://oncommand.io/blog/taking-too-long-to-ship) on what I've learned from the experience and am taking the steps to iterate quickly. My goal is to have the next major iteration (which includes reducing and consolidating features and adjusting my pricing) done by the end of the month.

I'm more confident in this direction because I'm doing what I didn't do for the first version: building out the marketing as I build out the features. This time around, the marketing is far clearer. It has an audience and it speaks directly to multiple wants.

Most importantly for me as a solo founder? This new iteration has less features to manage which means I can do what's suggested in one of my favorite quotes:

"The less you do the more of it you can do. Don’t invite more of what you don’t want."

• Scott Hanselman

I'm excited to dig into the details as opposed to constantly bolting on "something else."

I think the lesson learned here is to not get too caught up in the "brilliance" of your own features, but also, to not let that get you stuck in a cycle of navel-gazing. I had realized that my direction wasn't the best a few months before launch, but I resolved to just get it done and shipped instead of hitting the brakes and letting the app rot on my hard drive.

I'm glad I did that as being out in the open has given me a boost of energy and drive to iterate quickly to try and find some customers.

Onward :)