A workflow's step count decides whether Make or n8n costs less to run

A dev.to comparison reduces the cost question to one multiplication. On Make, multiply billable steps by runs per month. On n8n cloud, take the runs alone ^[7]. Runs appear in both bills and steps appear in only one. Depth decides which tool is cheaper, and volume decides how large the gap is in money. The post's own summary is that "deep workflows favour n8n, shallow high-frequency ones favour Make" ^[8].

Per-unit prices show where the lines cross. At the Core floor, a Make credit costs $0.0009 ^[1]. An n8n cloud execution costs 0.005 EUR on Pro and 0.008 EUR on Starter ^[2]. A three-step run on Make Core costs $0.0027. A twenty-step run costs $0.018 ^[3]. Compared with one n8n Pro execution, the shallow run costs about half as much and the deep run 3.6 times as much, in nominal terms with the currencies left unconverted ^[4]. Counted in runs, Make Core and n8n Starter cover the same number at four billable steps ^[7]. On Make's Pro and Teams tiers a credit costs $0.0016 and $0.0029, so n8n wins at a lower depth there ^[6].

Those figures hold only under narrow conditions. The post says it took them from each vendor's own pricing page, and that several other comparisons quote tiers that no longer exist ^[9]. Make's credits scale on a slider from 10,000 a month to more than 8 million ^[10]. The post prices only the 10,000 floor, so the per-credit number above is valid there and nowhere else. The n8n prices assume annual billing ^[4]. Make's unit is a module execution, and the trigger is free ^[1]. What you count is modules executed per run.

Frequency still matters on n8n, through concurrency. Starter allows 5 simultaneous executions and Pro allows 20 ^[4]. A shallow flow that fires constantly has to fit under that ceiling. Self-hosting removes the meter. The community edition has no licence fee and no execution limit, and the post puts the server at $5-50 a month on a VPS ^[6].

When workflows are short, the post says, the numbers land close together and the decision comes down to team and compliance ^[15]. Its case for Make is that a non-developer can open a scenario built six months ago and understand it ^[13]. Make is also cloud only. The post says data passing through Make's infrastructure ends the conversation for some compliance requirements ^[14]. I think that is the right split for most teams. With a developer on staff and pipelines past five steps, I would run n8n. I would leave three-step flows maintained by someone who does not write code on Make.