Long time lurker, first real post. It's a long one so bear with me.
I've been building custom CRM / ERP type systems for small businesses since late 2022. Mostly companies that have outgrown their spreadsheets but are nowhere near the size where a Salesforce partner would even pick up the phone.
The client I still think about
There's one client I still think about a lot. A wholesale book distributor. Honestly the most excited person I've ever built anything for. He wanted everything in one place, sales, customers, prospects, inventory, deliveries, an online store, all of it, and he wanted it to work the way his team already worked and not the other way round. So that's what we built. Deployed it, it ran fine, he was happy, I was happy, project done.
A few months later they needed some new features and they didn't call us. They went to a local agency.
I am still not very sure why. My guess is they wanted it quick and cheap and we probably weren't that, which, fine. But the agency kept bolting things on and bit by bit the whole thing got worse. Slower, buggier, more confusing to use. And when something broke nobody on his team could even tell where the problem was, so they'd wait for the agency, pay, wait again. He told me later it just became too much. Nobody inside the company could own it, and honestly why would they, they sell products, they're not a software shop.
So they gave up and moved to Shopify. Which as far as I can tell wasn't because it fit them better. It was just the only option where nobody had to be on call for it.
It's not just him
That one bugged me for a long time. And since then I've seen more or less the same thing happen a few more times. Custom software fits perfectly on day one and then the fit slowly rots, because keeping it alive costs more than anyone planned for.
The part that surprised me though is that the clients who gave up and went to some big SaaS aren't happy either. When I catch up with them it's the same complaints every time. The app feels bloated, half the screens are stuff they'll never touch. It doesn't quite do what they need so they've got workarounds all over the place. The software calls things one name and their team calls them another, so there's this constant translating going on in people's heads. And their data is actually messier than before, because when the tool doesn't have a proper place for something people just shove it wherever it fits.
So you end up choosing between two bad options I guess. Custom fits great but it's expensive to keep alive, SaaS is cheap to keep alive but never really fits. The Odoo / Salesforce partner route is supposed to be the middle, but for a 10 person company that's a lot of money and months of onboarding before you see anything at all.
What I've been trying instead
I've been chewing on this for about a year and a half now. The question I kept coming back to was, what if the part that breaks and needs maintaining was shared across everyone, and the bit that's actually specific to your business was really small. Like, tiny.
That's more or less what I've been building. One core engine that I maintain, and then each business gets a thin layer on top that speaks their language, their workflow, their terms, nothing they didn't ask for. I've moved four of my old clients onto it so far. A hardware store that does retail and wholesale, an electrical goods wholesaler (wires, switches, fans, that sort of thing), a company that installs solar for homes and offices, and a hostel with a canteen across a few locations. Pretty different businesses, and that was kind of deliberate, I wanted to see if the same core could bend to all of them. So far it's going better than I expected. But these are four people who already trusted me, so I know it doesn't really prove anything.
What I actually want to understand is whether people even want software to work like this.
The way I've started thinking about it, and I'm not sure this is the right way to put it, most business software is sold a bit like a supplement. Here's 200 features, here's what's good to have, here's the "right" way to run your pipeline, and you sort of bend yourself around it. What I'm trying to do feels closer to medicine, if that makes sense. You come in with a specific pain, the thing that jams up your week, and the system is built around getting rid of that one thing and then staying out of your way. Less "here's how you should work", more "what's stuck, let's unstick it". I don't know. I keep going back and forth on whether that's a real distinction or just something that sounds nice.
So, honestly asking you guys
And maybe in reality everyone wants the 200 features anyway. So:
- What's the one bottleneck in your day that your current software makes worse instead of better? Not a missing feature, I mean the actual thing that slows you down.
- What's the workaround you've done so many times you stopped noticing it's a workaround?
- If a tool just removed that one thing and did nothing else fancy, would that be enough for you? Or would you need the whole feature list before you'd trust it?
- And has anyone ever tried to get a vendor to fix a real bottleneck instead of adding another feature? How did that go?
I'm not trying to sell anything here, I'm genuinely still working out whether I'm right about this. Happy to be told I'm not.
And if your situation is messier than a comment can hold, my DMs are open. No pitch, just curious what's getting in your way.
TL;DR: built custom CRMs for small businesses for 4 years, kept watching them get abandoned because nobody could maintain them, and the ones who fled to big SaaS aren't happy either. Now building something in between and want to know what your actual bottleneck is, not what features you want.
Login
