Make pricing and credits: estimate your workflow before choosing a plan

A workflow that looks like one job can consume several credits. Fetching a record, updating another app and sending a notification are separate pieces of work. That is why the useful Make pricing question is not “How much is one automation?” It is “What does our actual scenario consume in a normal and busy month?”

Make, also searched as Make.com, lists plans by feature tier and credit allowance. This guide uses current official pricing and credit documentation checked on 6 October 2026. The examples are calculations, not scenarios we executed.

What Make Free includes

The pricing page lists 1,000 credits per month, a visual builder, routers and filters, and a 15-minute minimum interval between scheduled runs. Confirm whether your design uses scheduled polling or another trigger; do not apply the scheduling interval indiscriminately to every type of event.

A workflow may fit the credit allowance while missing a required feature or timing requirement. Check both. A free pilot is useful when it measures the actual process, not when it merely produces a successful demonstration once.

Credits versus operations

Make’s documentation says a non-AI app operation uses one credit. Credits describe usage cost; operations describe executed activity. Some AI features use different credit rules, and a connection to a third-party AI provider can create a separate provider bill.

That makes “one module equals one credit” too broad as a planning rule. Record the specific module and current charging rule, especially for built-in AI features. An app’s presence in the directory does not prove that a particular action is included on the plan you selected.

A worked capacity example

Suppose a scenario receives a record and then runs three downstream non-AI operations per record. For illustration, assume each of those operations costs one credit and that 300 records reach all three operations.

300 × 3 = 900 credits for those downstream operations. That leaves only 100 of the free 1,000-credit allowance before accounting for the trigger’s own usage, repeated processing or any additional path. We deliberately do not call that “300 complete free jobs,” because the whole scenario has not yet been measured.

Now add a second branch reached by half the records, with two extra chargeable operations per record. Under the same illustrative assumptions, add 150 × 2 = 300 credits. The downstream total becomes 1,200, already above the free allowance before other usage. A branch that looks small can therefore change the plan.

The final row is a deliberately limited estimate. Add the actual trigger, polling and other metered activity before choosing a plan. The repeated-work row assumes those three operations really execute again; it is not a claim that every error costs three credits.

Use a measured rate for your own estimate

Run safe representative records and inspect the usage log. Separate the common path from exceptions and repeated records. Then estimate:

monthly credits ≈ normal-path usage + additional branch usage + expected repeated processing + other measured activity.

Do not add a universal retry charge without checking how the actual operation is counted. Measure the behavior you intend to use. Keep the module configuration with the estimate so someone can reproduce it after the workflow changes.

Paid Make pricing: record the selection, not just the headline

The page we checked displayed Core at $12, Pro at $21 and Teams at $38 per month for the selected 10,000-credit allowance. Those are the displayed values at retrieval; this guide does not establish a checkout quote for every billing toggle or country. Save the selected credit tier, billing period, total charge and renewal terms before comparing them.

A separate browser observation on 6 October 2026, with Pay monthly visibly selected at 10,000 credits, showed Core $10.59, Pro $18.82 and Teams $34.12. These differ from the retrieved text above. We have not established the reason or a checkout quote, so neither display is presented as a universal confirmed charge. Save the selected currency, billing period, allowance, total invoice and renewal terms before comparing. Include tax and separately billed services where relevant.

Core lists more scheduling control and unlimited active scenarios. Pro and Teams add further capabilities. Buy a higher tier because a required feature or workload needs it, rather than because its name sounds safer.

What to record from a trial run

Keep a small worksheet with the input ID, path taken, credits shown, result and any second attempt. Include a new record, an existing record and a repeated event. Use test destinations so the exercise cannot create duplicate customer messages.

For illustration, if a complete normal path is measured at four credits per record and an additional path at two credits for 100 records, then 300 normal records plus those branches require 300 × 4 + 100 × 2 = 1,400 credits, before additional activity. These are assumed rates, not a Make benchmark. Replace them with your logged values.

Separate the completed job count from the executed operation count. A successful job may contain multiple operations; a failed attempt may still have completed earlier actions. When usage differs from your estimate, investigate the path rather than simply buying a bigger allowance.

AI can introduce a second cost

For connected third-party AI apps, inspect both Make usage and the provider’s token billing. For Make-provided AI features, use that module’s credit documentation. Keep generation size and expected calls in the estimate. A credits-only spreadsheet can understate the bill if it ignores a separately billed service.

What happens near the limit?

Make documents notifications as purchased-credit usage approaches the allowance and options for obtaining additional credits. It also describes scenarios stopping when credits run out. Confirm the current plan behavior and any automatic-purchase setting in your account. Agree on who receives the warning and who decides whether to buy more capacity.

Do not automate a time-sensitive handoff without a recovery plan. Record how to identify unprocessed records and resume safely without duplicating downstream actions.

Compare Make with the same job elsewhere

Make credits and Zapier tasks are different units. Run the same safe inputs through both designs before comparing measured usage. Read Make vs Zapier for the buying decision or free Zapier alternatives for a broader shortlist. The Make profile links the official plans, and the workflow trial worksheet helps record the results.