Make vs Nrth Star: a canvas you draw, or a plan you read

Make is a canvas. You lay modules out, draw the connections, and end up with a diagram of your process that you own completely. It is more capable than most people ever need, and for anything with real logic in it, it is better than what we do.

Nrth Star has no canvas. An automation here is a page of English sentences you read and approve. Nothing is dragged, nothing is connected, and there is no diagram, because the thing we are optimising for is not how quickly you can build it but whether you can tell what it will do before it does it.

That is a genuine disagreement about how an automation should be represented, and which side you want depends on facts about your business rather than on which tool is better.

What Make is genuinely good at

Worth saying properly rather than as a courtesy, because it decides whether the rest of this page applies to you.

Real logic. Routers, iterators, aggregators, error handlers. If your process needs to fan out over forty records, do something different depending on what each one contains, then collect the results back into one summary, Make does that properly and it is the reason people move to it from Zapier.

Control. Every module is yours to inspect, reorder and change, immediately, without asking anybody. For somebody who enjoys this, that is not a chore, it is the appeal.

Coverage. Thousands of apps, so the obscure tool one client insists on is probably in there.

Cost at volume. Charging per operation is finer-grained than charging per task, and for a business running high volume it usually works out cheaper than the alternatives.

If you have somebody who wants that and will keep wanting it in eight months, stop here. Make is the better product for you and this comparison is not close.

The canvas, and what it costs

The canvas is the best thing about Make and it is worth being precise about what it is good for, because it is not one thing.

For building, a canvas is excellent. Complex logic is genuinely easier to lay out spatially than to write down. You can see the branches. You can see where a thing splits and where it comes back together. Nobody would design a fifty-step process as a list of sentences.

For reading it back, a canvas is a different story. A diagram of ten modules is clear. A diagram of fifty is a picture of a city, and the person opening it eight months later to work out why a client got the wrong email is frequently not the person who drew it. They have to trace it. And tracing a diagram tells you the shape of the process rather than what it will actually do to a specific record.

That gap is the thing this product is built around. An automation here renders as sentences of the form when this happens, wait this long, unless this is true, send this, and somebody who has never seen the tool before can read it and say that is wrong. Not because sentences are a nicer format, but because the person who most needs to catch a mistake is usually the least technical person in the room, and they will not trace a diagram.

The price we pay for that is real and it is the concession above: a process with heavy branching cannot be rendered as a readable page of English, so the schema refuses to express one. Nested conditions are kept deliberately weak and iterating over a list is not supported yet. That is a limitation, not a phase.

Somebody has to know what to build

Make opens on an empty canvas, which quietly assumes you have already done the hardest part.

Most people, faced with an empty canvas, build the process they happened to think of first. That is almost never the one costing the most, because the expensive ones are usually invisible: spread across four people, twenty minutes at a time, never discussed because everyone assumes it is just how things are.

Nrth Star starts at the other end. The audit asks how the business actually runs, including the people who are not the founder, and returns a ranked list of where the hours go with a number against each. You pick from that list. You can run that exercise yourself with a spreadsheet, and it is worth doing whichever tool you end up using.

What happens in month eight

Building an automation is a day. Keeping fifteen of them working is a job, and it is the part nobody prices in.

Systems change underneath an automation without telling anybody. A field is renamed, a permission is tightened, a pipeline stage is added. In Make the scenario errors and stops, you get told, and fixing it is yours. That is a clean failure and it is the good case.

The bad case is the one no tool reports: the automation keeps working and starts doing the wrong thing. A filter that was right for the pipeline you had in March is wrong for the one you have in June. Every run succeeds. Nothing errors, so nothing alerts, and you find out from a client.

That failure is what the shadow run exists for. Before an automation is allowed to do anything, it is replayed across your last 30 days with every send and write switched off, and you read what it would have done against real clients and real dates. It is the question worth asking any vendor, including this one.

The two pricing shapes

Not the numbers, which move, but the shapes, which do not.

Make charges per operation, and an operation is one module execution rather than one run. A scenario with eight modules that fans out over thirty records does not cost eight operations. This is finer-grained and usually cheaper than per-task pricing, and it is genuinely harder to forecast, because the cost of a run depends on how much data came back that day.

Nrth Star charges per automation you have running, whatever it does and however often. A follow-up that fires twice a month and one that fires four hundred times cost the same. That is worse value if you are running very high volume and better if you want to know the number before the month starts.

Which one you want

Use Make if somebody at the business wants to own this, if your processes have genuine branching and iteration in them, if you are running high volume, or if you want to be able to change a module at four in the afternoon without asking anyone.

Use Nrth Star if nobody wants to own an automation platform, if you are not sure which process is worth automating in the first place, or if the thing that worries you is not building it but what it will quietly do to real client records once it is running.

Plenty of businesses should use Make. If you are still deciding between several of these, the wider roundup covers the rest including the automation you may already be paying for and have not switched on.