/INVITEROBOT • 3 min read

48h to build a Slack paywall

Edit: This challenge has now grown into a real product, and is used by live clients. Read the journal on how we launched InviteRobot, or check out inviterobot.com if you need a payment form for your paid Slack community!

“Let’s do this. Let’s make it a 48h challenge! We start Monday and try to finish it by Wednesday night!”

In hindsight, 48 hours would have never been enough1.
But ignorants we were, and so we started, and made it ;)

The idea

Two days ago, I pitched my friends Bastien and Romain on an idea which — I believed 😅 — we could deliver in 48 hours.

The idea?

Provide a quick and easy solution to take payments and send invitations to Slack teams.


Discussion went a bit like this:

  • We 💗 Slack, and we’re not alone: more and more communities create private Slack groups2, which become their main communication tool.
  • Some of them already switched to a paid membership. We believe this trend of VIP/Premium communities will continue, mainly because:
    1. they can be niche-focused and provide enough value that people are willing to pay,
    2. this is one of the easiest idea to execute for a solopreneur. Which is particularly appealing to digital nomads — which is a trend in itself.
  • A quick google search revealed that people are already hacking things around to automate Slack invitations and take payments.
  • We so want to 💙 Stripe, but never had the occasion to implement it. Bingo, purrfect pretext!

So that’s easy: both Slack and Stripe offers APIs, nicely documented. It’s just a matter of plugging them together. What could go wrong?

48 hours later

We not only had fun, but we have something to show the Internet!

Our payment widget for Slack

It’s quite barebone, and is more a working proof of concept than a complete solution. We do process payments and send Slack invitations but big, important things are missing: SSL to process LIVE payments, oAuth tokens storage, exceptions handling, and many many other small details.


We need to practice more.

To be clear: we spent most of the last 48 hours stretching our skills. Reading over Sinatra documentation, setting up Heroku, playing with Stripe and Slack’s APIs, etc.

We did more, and we did it faster than what we could have done just a few years ago, by far. But I really think 48 hours should be enough for a basic, A-to-Z implementation.

The good news is, now that we practiced a bit, we should breeze through these stages for our next project!

We don’t like tests anyway :)

We took shortcuts wherever we could to fit within our 48-hour window. We definitely could be even more selective.

Momentum is key.

Setting a time-limit forced us to move our asses and get going. The deadline was also the purrfect argument to end lengthy discussions. Just do it, see what happens.

We’re pretty pumped up 💪 about this first step. We think we can have a working product (= used by customers) in a few days.

Setting us in motion was the hardest part. Now, here we go!

Subscribe to Til' You Make It!
Get the latest posts delivered right to your inbox.
No Spam, No Bullshit. Unsubscribe Anytime.

Want to be smart and loved? Share this article!