SaaS App Retrospective - 12 Lessons I Learned

An assessment of the journey creating my SaaS app - highs/lows and things I learned through the process.

In October 2020 I decided to build a SaaS app using NoCode, with the aim of getting multiple monthly paying customers.

I did it.

thank you for attending my TED talk.

If you want some more details, read on :)

Why NoCode

For a year leading up to choosing a Nocode approach, I had built a shopping list app with code. I used the Ember.js front end framework, and it was going great. I had a working app, with multi user OAuth login, talking through AWS API Gateway, being handled by Lambda using Node, persisting data in a DynamoDB database. Great.

The trouble was I was only really 60% done, and it had taken 12 months, part time, squeezed into the gaps around my day job. I hadn’t built the invitations logic, nor made the UI look nice. I had come to a choice in the road.

Keep yak shaving with code, or pivot?

I knew that even though I liked code, and problem solving, that the actual fighting with syntax and dealing with all of this and this and this wasn’t actually getting me to the endpoint very fast.

Lesson 1 : Just because you like to code, doesn’t make it the best choice for solving problems quickly.

After some research I found Bubble.io and I gave it a go. Within a short amount of time I had rebuilt all the work I had done over the last year, and knew that this was a good option for me… a faster option. My Chore Roster automation app was live.

The next steps I’ve already documented on this twitter thread, so pop over there if you want to know what came next, but don’t forget to come back :)

Marty Lindsay @MartyLindsay_NZ

SaaS Build Log MegaThread As I built @RosterBuddyApp I wrote down all the milestones from the starting point, to the first paying customer Details Below 👇 #BuildInPublic #NoCode #Bubble

July 5th 2021

1 Retweet6 Likes

December 2021 - review time

OK, having made it through to December I had a review of progress to date. I had previously built public metrics into the app, so it was easy to see the status.

And with Stripe reporting I knew who had upgraded to become a paying customer, which was 5 paying customers and 1 trialling.

So, not terrible, but not great either. I was confident that since I had multiple recurring paying customers that if I spent enough of the right effort marketing in the right places that this would grow over time, and some profit would come my way.

I wasn’t exactly enthusiastic about the effort required, because the topic of ‘household chores’ isn’t the most exciting area of research.

Lesson 2: make sure you find the topic interesting, so you want to invest time into it

But there was a bigger problem. User Activity.

I basically had very little activity from the user base. Quite a few sign ups, and some initial experimenting. But I never quite got to the point where the same users would come back week after week. This shows that product market fit was still a while off. I suspect that a lot of the signups were from tech/twitter types who were somewhat curious about the product but no so much about the problem it was solving:

Lesson 3: Just because you have an audience in one area, doesn’t mean they are the audience that matches your product.

Even the paying customers weren’t that interested in coming back, which was a surprise to me.

Lesson 4: Even if you get multiple paying customers, it doesn’t mean your app will be a success.

Other Problems

There were some other issues two which I suspect impacted the uptake:

Invitations

I made a decision that a team member couldn’t be allocated chores by a team admin, unless that team member had actively accepted the invitation to the service. I still maintain this was the right choice, but it did have an impact in that there were many incomplete teams, with pending invitations. This meant that work couldn’t be allocated evenly, thereby reducing the value the customer received.

Lesson 5: make onboarding simple and easy, and deliver value early on.

Lumpy Traffic

My lack of enough ‘interest’ in the topic manifested in lack of content on my site [ i.e no blog], which affected by ability to SEO optimize, which meant very little organic traffic. The traffic I did get was very lumpy, and tightly coupled to sporadic marketing efforts.

Lesson 6: Choose a topic you can write a lot about, and often, to get that sweet organic traffic.

Target Market lacked clarity

Was I targeting families… or flatmates.. or dorms.. or flat-shares… or all of them? I’m not entirely sure. I do know I could have benefited from narrowing the scope to help with marketing efforts.

Lesson 7: Start with a small well defined audience and solve a simple problem well

Not all problems are Tech problems

I have always known this, but it was a good reminder. I came to realise that if someone in a household was getting frustrated with their messy housemates not cleaning up, then having an app to manage that wasn’t necessarily going to change their housemates into caring and thoughtful contributors to the household. Some peeps just plain lazy.

Lesson 8: Only choose a technology solution for a problem that needs a technology solution.

Some good news

I should clear up that I never actually expected for my first SaaS app to get multiple paying customers, so on that basis I’m really pleased with how it turned out. My main goal of the process was to learn how Bubble.io worked, and to also have that learning process as the fuel for my writing, which is this newsletter.

Lesson 9: it’s great to solve a problem of your own, as a means to getting better at building

Lesson 10: You can in fact build a SaaS which generates revenue via multiple paying monthly subscribers, using NoCode tools, as a solo founder, while still keeping your day job

Experiment

I did a lot of experiments and changes. I’m glad I did. I edited pricing, plans, features, layouts, onboarding, invitation logic. I did marketing experiments. I recorded lots of metrics.

Lesson 11: be comfortable changing things, and recording the results.

A Warning

Just a note here. Don’t get lead down the path by path dependency. Don’t get hoodwinked by the sunk cost fallacy.

By which I mean, set yourself a reminder to review progress, and don’t just keep going forever flogging a struggling app. Have an honest assessment of your costs, growth rate, interest, challenges etc. Some people even have their domain renewal as the reminder.. i.e to not renew a domain if that app isn’t yet profitable.

Lesson 12 : Be well informed and don’t get emotionally attached to your projects

Summary

So, I reviewed progress in December 2021, and even though I had multiple paying customers, I decided to decommission RosterBuddy.app. It was nearly making a profit, but I didn’t think the rate of growth was going to justify the time I’d need to invest in it. Time to move on.

What’s next

Plenty. I’m digging into the new Responsive Engine that Bubble released. I’m planning to collaborate on some new build projects using Bubble. I’m going to try the new Atomic Fusion system that Neil is releasing soon. I’m reading Arvid’s book about Audience Driven business, to help me align with a better topic area. I’m cooking up my next business idea.

And I’ll be sure to write a lot more in 2022 about what I learn :)

Thanks so much for reading.. and ps if you have found these articles useful please consider subscribing to help power my writing with coffee. Also a subscription gives you access to every article, not just the free ones.