By Mo Hussein

2 min read

Start with the cheap thing

If two of your apps both have a Zapier or Make connector, and you just want a record to move from one to the other, you probably don't need us. Someone handy in the office can set that up in an afternoon, and it'll run quietly for years.

I say that on discovery calls a lot. It means less work for us, but nobody wants to pay for a bespoke build that a monthly subscription would have covered.

The signs you've outgrown it

No-code tools are brilliant right up until they aren't. These are the moments we tend to hear about:

  • Someone fixes a broken automation most weeks, and it's become part of their job without anyone deciding it should be.
  • You need a two-way sync, where people edit the same record in both systems. That's where duplicates and overwritten changes creep in.
  • Your bill keeps climbing because pricing is per task, and you're busier than when you set it up.
  • The step you need isn't in the connector. The app can do it, but only through its API.
  • Customer, payroll or health data is passing through a service nobody has checked properly.
  • The person who built it has left, and nobody knows what it does or why.

What custom actually means here

It's rarely a whole new app. Usually it's a small piece of software that sits between your systems, talks to each of them directly, and does exactly the job you need. A good one keeps a log of every run, retries when a service has a wobble, and tells a real person when something genuinely needs attention.

You also own it. It doesn't vanish if a connector is withdrawn or a price plan changes.

An honest way to compare the cost

Don't just compare the subscription with the build price. Add up the hours people spend nursing the automation, re-entering what it missed and fixing the knock-on errors. Then set that against the build, plus hosting and a bit of maintenance each year.

Sometimes no-code still wins comfortably. Sometimes the hidden hours are the biggest cost in the whole process, and nobody had added them up.

You don't have to choose all or nothing

The most common answer we give is a mix. Keep the automations that work and replace the one that keeps breaking. If you're not sure which one that is, bring a list of your automations to a discovery call and we'll go through them with you.