Some tech decisions look completely fine on the day they’re made. The problems show up two years later, at 3pm on a Tuesday, when everything stops working and the person who built it has vanished into the ether like they never existed.
I had to tell a client I couldn’t fix it. Here’s the full story.
The brief (aka: this is going to be fun, I can feel it)
A business owner came to me wanting to set up a membership site. Established, results-proven, with a genuine reputation for going above and beyond for their clients. Cards in the post, little gifts, the kind of coach who really sweated the client experience. Their brand was warm, fun, personality-led. Now they wanted to take their content and turn it into a membership.
They weren’t technical and didn’t want to be. Budget was sensible rather than lavish. So I looked at what was available, weighed the options, and came back with a recommendation: a well-established membership platform with a long track record, solid documentation, and a wide pool of people who knew how to work with it.
Boring? A little. Built to last? Absolutely.
How I actually make these decisions (it’s not vibes, I promise)
My background is ten years in corporate IT, so I think about two things: can this keep running, and when it breaks, can you actually get it fixed?
I also have what I call the Service Pack 2 rule, which is less official but equally useful. Don’t deploy the newest, shiniest tech. Wait until it’s been around long enough for everyone else to find the problems first. Service Pack 1 fixes the bugs. Service Pack 2 fixes the bugs that Service Pack 1 caused.
The platform I recommended was well past that point. Stable, supported, and boring in the best possible way.
Where it went a bit sideways
They said no.
They had ideas. Fun, creative ideas that made complete sense for their brand but that the platform couldn’t accommodate without a custom build. I explained what that actually meant in practice:
Higher cost
Significantly harder to support
A much smaller pool of people to call on when something inevitably went wrong
Custom solutions only make sense when you have someone full-time to maintain them. This business wasn’t at that scale, and honestly, most aren’t.
What mattered, practically, was a membership platform that stayed up, that they could keep promoting without constant firefighting, and that generated as few support problems as possible. They were excited about the fun side of the build. I was quietly thinking about 3pm on a Tuesday in two years’ time.
They found a developer who said yes anyway.
Fair enough. They launched, it went well, and for a couple of years, things were fine.
Then their site fell over.
What no disaster recovery plan actually looks like (spoiler: it’s grim)
When I asked about backups, there weren’t any usable ones. The developer who’d built the custom plugin had disappeared. No forwarding address, no documentation, nothing to hand over. There was no disaster recovery plan in place, which was something I’d raised during our original conversations.
I had to tell them I couldn’t fix it.
I gave them some referrals and was honest that I wasn’t holding out much hope. When a custom system breaks with no documentation and no backups, options are limited. Their site needed to be rebuilt from scratch.
Not the email anyone wants to receive.
The actual happy ending (yes, there is one)
They went through a rebrand, built a new site, refreshed everything. When they were ready to try the membership again, they came back. This time we set it up properly: an established platform, on a separate site, the way I’d suggested originally.
It launched without drama. They could use it without needing a manual. When they had a new idea I thought was risky, they listened to the explanation and we found a better way forward together.
That is what working with the right tech person is supposed to feel like.
The moral of the story (and it’s not just “listen to Jenn”)
There are people in tech who will say yes to almost anything, take the work, and build exactly what you asked for.
They might not build you what you actually need.
A membership site with every bell and whistle sounds great in theory. Something simpler, maybe a little boring, that runs predictably and stays out of your way will save you more energy than any clever custom feature ever could.
If you love the fun and the colour and the personality, brilliant. Put that into your copy. Put it in your social media. Post the pretty stuff on Instagram. That is what those tools are for.
Your backend systems are not the place for self-expression.
In my world, solid and predictable wins over pretty and fun. Every time.
Want someone who’ll tell you the truth before your site falls over?
If you’d rather hear the uncomfortable truth now than deal with a rebuild later, drop me an email. Tell me what you’re building. We can make sure it’s solid enough for you to sleep well.
