Theory Of Everything
Junior Member
- Apr 18, 2020
- 150
- 158
A few months ago I shipped a compliance SaaS for a small UK trades vertical. £2k MRR last month, 60 paying customers, tiers from £19 to £99. Built the whole thing on Replit and Claude Code. No ad spend.
Posting this because I see a lot of "I made $50k MRR in 90 days" Reddit threads and not many at the slightly-awkward £2k-and-climbing stage, which is where most of us actually live if you do SaaS.
The vertical is unglamorous trades compliance. Contractors who have to issue certificates to stay legal. I won't name it because the niche is small enough that anyone Googling specifics would find me, but the point is: it's the kind of work people pay for because they have to, not because they want to. That's the bit I'd recommend looking for if you're picking a niche, boring legal/regulatory obligations beat shiny consumer wants by a long way.
Avoid "nice to haves" sort of industries, if they don't need it, they won't pay for it. It's just the way it is.
I'm vibe coding the whole thing. Claude Code and Replit for the app, Supabase for the database and auth, Stripe for billing and Resend for transactional emails I send across.
Every time a customer says "this is annoying" or "can it do X" I write it in a notes file. Most days I pull two or three items off the file and ship them the same evening after I finish my dayshift. Nobody's used to that pace in compliance software, where the incumbents push updates quarterly if you're lucky. The first time a customer asked for something on a Tuesday and saw it live on Wednesday they emailed asking if I was a real person lol.
That speed is most of the moat. The features are catchable. The willingness to ship the next thing the same week isn't.
Distribution-wise: Reddit and Facebook groups. That's the whole story.
Reddit was the messier of the two. Some subreddits removed posts on sight , the kind that auto-nuke anything with a link or anything that smells of self-promotion. I worked around it by posting without links, leading with the problem rather than the product, and letting people DM me. A few subs I'd expected to do well were dead zones. Others I almost didn't bother with quietly turned into the best ones.
There's no way to know in advance, you just have to keep posting and see what survives.
Facebook groups have been more forgiving. The trick there is you have to actually be in the conversation for weeks before you mention what you've built. Drive-by self-promotion gets you banned. Showing up, answering questions, occasionally referencing "the thing I built for this" gets you DMs.
And some FB groups despite having 10k+ people in the group have trash engagement, you can tell by a quick glance by checking likes of recent posts, they usually have 0. And people spamming the page too.
The £19 to £99 monthly tiers are the visible pricing. What's actually moving the needle on cash flow is lifetime deals.
I run LTDs selectively, discount for the customer, a chunk of cash to me upfront, no recurring revenue from that customer ever again. On paper it looks like you're trading future MRR for cash today, which is bad if you believe the customer would have stayed for years.
In practice, in this niche, B2B customers churn for reasons that have nothing to do with you (they sell up, they retire, they change accounting software and forget you exist). Taking the lump sum today is often better than the expected value of monthly billing for those customer profiles.
LTDs also do something quieter, they convert hesitant people. Someone on the fence about £39/month will pay a few hundred once because their brain treats it as a purchase rather than an ongoing commitment. Even if the lifetime value maths comes out the same.
Cold email and cold calling are in the mix too. Both have produced customers. Neither was as efficient per hour as community posting. If I was starting again I'd do less email outreach earlier and more community posting earlier — people in the groups have already self-selected as the type of contractor with the problem, and warm intros from "the guy in the group" beat cold intros from "this stranger emailing you" almost every time.
£2k MRR is not life-changing money. It pays for the stack, the tools, and a small wage's worth of beer. The reason I'm writing it up is because the trajectory matters more than the number — feature velocity is high, churn is low so far, and the customer base is concentrated enough that I can talk to most of them by name. That's a much better position at month four than "I have one massive customer paying £2k" would be.
I'll do a follow-up at month eight with what changed. Happy to answer questions on the build, niche selection, or the LTD maths.
Posting this because I see a lot of "I made $50k MRR in 90 days" Reddit threads and not many at the slightly-awkward £2k-and-climbing stage, which is where most of us actually live if you do SaaS.
The vertical is unglamorous trades compliance. Contractors who have to issue certificates to stay legal. I won't name it because the niche is small enough that anyone Googling specifics would find me, but the point is: it's the kind of work people pay for because they have to, not because they want to. That's the bit I'd recommend looking for if you're picking a niche, boring legal/regulatory obligations beat shiny consumer wants by a long way.
Avoid "nice to haves" sort of industries, if they don't need it, they won't pay for it. It's just the way it is.
I'm vibe coding the whole thing. Claude Code and Replit for the app, Supabase for the database and auth, Stripe for billing and Resend for transactional emails I send across.
Every time a customer says "this is annoying" or "can it do X" I write it in a notes file. Most days I pull two or three items off the file and ship them the same evening after I finish my dayshift. Nobody's used to that pace in compliance software, where the incumbents push updates quarterly if you're lucky. The first time a customer asked for something on a Tuesday and saw it live on Wednesday they emailed asking if I was a real person lol.
That speed is most of the moat. The features are catchable. The willingness to ship the next thing the same week isn't.
Distribution-wise: Reddit and Facebook groups. That's the whole story.
Reddit was the messier of the two. Some subreddits removed posts on sight , the kind that auto-nuke anything with a link or anything that smells of self-promotion. I worked around it by posting without links, leading with the problem rather than the product, and letting people DM me. A few subs I'd expected to do well were dead zones. Others I almost didn't bother with quietly turned into the best ones.
There's no way to know in advance, you just have to keep posting and see what survives.
Facebook groups have been more forgiving. The trick there is you have to actually be in the conversation for weeks before you mention what you've built. Drive-by self-promotion gets you banned. Showing up, answering questions, occasionally referencing "the thing I built for this" gets you DMs.
And some FB groups despite having 10k+ people in the group have trash engagement, you can tell by a quick glance by checking likes of recent posts, they usually have 0. And people spamming the page too.
The £19 to £99 monthly tiers are the visible pricing. What's actually moving the needle on cash flow is lifetime deals.
I run LTDs selectively, discount for the customer, a chunk of cash to me upfront, no recurring revenue from that customer ever again. On paper it looks like you're trading future MRR for cash today, which is bad if you believe the customer would have stayed for years.
In practice, in this niche, B2B customers churn for reasons that have nothing to do with you (they sell up, they retire, they change accounting software and forget you exist). Taking the lump sum today is often better than the expected value of monthly billing for those customer profiles.
LTDs also do something quieter, they convert hesitant people. Someone on the fence about £39/month will pay a few hundred once because their brain treats it as a purchase rather than an ongoing commitment. Even if the lifetime value maths comes out the same.
Cold email and cold calling are in the mix too. Both have produced customers. Neither was as efficient per hour as community posting. If I was starting again I'd do less email outreach earlier and more community posting earlier — people in the groups have already self-selected as the type of contractor with the problem, and warm intros from "the guy in the group" beat cold intros from "this stranger emailing you" almost every time.
£2k MRR is not life-changing money. It pays for the stack, the tools, and a small wage's worth of beer. The reason I'm writing it up is because the trajectory matters more than the number — feature velocity is high, churn is low so far, and the customer base is concentrated enough that I can talk to most of them by name. That's a much better position at month four than "I have one massive customer paying £2k" would be.
I'll do a follow-up at month eight with what changed. Happy to answer questions on the build, niche selection, or the LTD maths.
