No-Code & Automation

No-code covers two related promises: build software without writing code, and connect software you already use without writing code. The second is where most of the durable value has landed.

Automation tools — the ones that watch for an event in one system and do something in another — solve a problem almost every organisation has. The tedious middle of most jobs is moving information between systems that do not talk to each other, and a workflow that does it reliably saves real hours every week. The category is mature, the reliability is generally good, and the pricing model is usually per-task or per-run, which means a workflow that fires constantly can get expensive quietly.

App builders are the more ambitious half. They have improved enormously and they genuinely work for internal tools, admin panels, forms, dashboards and simple customer-facing products. The honest limitation is that the ceiling arrives suddenly. Everything is easy until one requirement sits outside what the builder anticipated, and then you are either abandoning the platform or writing code inside a constrained environment that is harder than just writing the code would have been.

The question worth asking before you commit is what the exit looks like. Some platforms export a real codebase you could host yourself. Some export data but not logic. Some export nothing meaningful, which means the tool is not a shortcut but a permanent dependency. None of these is wrong, but they carry very different risk, and the answer is rarely on the pricing page.

Cost is the other thing that surprises people. Per-seat pricing is predictable. Per-run and per-task pricing is not, and the bill for a workflow triggered by every row in a growing database can climb sharply. Model the cost at ten times your current volume before you build something load-bearing on it.

The tools we find most interesting here are the narrow ones: a single automation done extremely well, or a builder scoped tightly to one kind of application. Broad platforms compete on feature count and tend to be mediocre everywhere. A tool that only builds internal admin panels, and does it perfectly, is usually the better purchase.

For makers, show the thing being built in real time. This category sells on the demonstration more than any other, and a sixty-second screen recording of a workflow going from nothing to working does the entire job.