Full disclosure: some links below are affiliate links. I only recommend tools I’ve paid for and actually use.
Short version: Make (formerly Integromat) is a solid choice if you want to how to build AI workflows with no coding and don’t mind paying a bit more for advanced features. I’ve used it for six months to automate content pipelines, and it works most of the time.
When I first looked for a no‑code way to chain AI models together, I wanted something that felt less clunky than Zapier’s UI and offered more control over data transformations. Make.com promised a visual builder with fine‑grained error handling, so I gave it a try.
Is Make.com worth $29/mo?
That’s the question I kept asking myself after the free tier ran out. The starter plan gives you 10,000 operations per month, which sounds generous until you realize each AI call, each filter, and each router step counts as an operation. In practice, a simple workflow that pulls a tweet, runs it through GPT‑4, and posts the result to LinkedIn can eat 15‑20 operations per run. If you run it four times a day, you’re already at 2,400 ops a month—still under the limit, but any extra testing or branching quickly eats into the buffer.
I think $29/mo is fair for the number of operations you get, especially when you factor in the built‑in AI connectors that let you call GPT‑4, Claude, or Llama 2 without writing a single line of code. If you need more than 10,000 ops, the next tier jumps to $99/mo for 100,000 ops, which feels steep unless you’re running high‑volume automations.
— and good luck finding docs for this — the operation counter is hidden in the settings menu, not on the dashboard, so you have to dig to see how close you are to the limit.
What sucks about Make.com
The biggest gripe I have is the vague error messages. When a module fails, the tooltip often says “Something went wrong” with no clue which input caused the problem. I once spent an hour troubleshooting a JSON parsing error only to discover that a trailing comma in the payload was the culprit—something the UI never highlighted.
Another annoyance is the limited version history. You can only roll back to the last 30 versions, and there’s no way to label or tag a specific version as “stable”. If you experiment with a new branch and want to revert to a known good state, you have to manually copy modules or rely on external backups.
Finally, the mobile app is essentially a read‑only viewer. You can’t edit workflows on your phone, which feels like a missed opportunity for quick tweaks on the go.
What’s actually worth paying for
I love the built‑in AI model connectors. Adding a GPT‑4 module is as simple as dragging it onto the canvas, selecting your API key, and mapping the input and output fields. No need to mess with HTTP requests or authentication headers—Make.com handles the OAuth flow for you.
The data transformer tools are also a standout. The array aggregator, text parser, and JSON mapper let you reshape complex payloads without writing code. I’ve used the XML to JSON converter to ingest RSS feeds and turn them into clean data for my a newsletter platform like Beehiiv automation.
Another feature that saved me time is the error handling routing. You can attach a “break” handler to any module and direct failures to a Slack notification or a fallback workflow. This means my pipelines keep running even when an external API hiccups.