Make.com vs Power Automate: canvas against tenant
A visual scenario builder priced by operation against a per-user platform built into Microsoft 365. Where each is genuinely stronger, and what the two cost at the same workload.
Rebecca Pearson10 min read
On this page
These two are less alike than a feature comparison suggests. Make is a visual automation product you buy on its own merits and price by the operation. Power Automate is a component of Microsoft 365, licensed per person, and its greatest advantages come from being inside the tenant rather than from anything on its feature list.
So the question is rarely which builds automations better. It is whether your work lives inside Microsoft, and whether you are buying capacity or seats.
Prices were checked against each vendor's pricing page in September 2026 and change often. Those pages are the authority.
What we'll cover
What each actually is
Make is a visual scenario builder. You lay modules on a canvas and draw the paths between them, including branches, routers, iterators, and error handlers. The structure of the logic is visible as a shape, which is the product's defining characteristic and the reason people who like it like it a great deal.
Power Automate is Microsoft's automation layer across Microsoft 365 and the Power Platform. It builds cloud flows through a designer, and it also carries desktop flows for interface automation, process mining, and integration with Dataverse and the rest of the Power Platform. It is a platform component more than a standalone product.
That difference explains most of what follows. Make is chosen; Power Automate is frequently already present.
Pricing, and the shape it rewards
Make publishes simple tiers: Free at $0, Core at $9, Pro at $16, and Teams at $29 a month. The billing unit is the operation, which is roughly one module doing one thing. The counting is relatively generous, particularly where a single module processes a batch rather than one record at a time.
Power Automate Premium is $15.00 per user per month, paid yearly, with higher tiers at $150 and $215 for hosted and unattended process automation. The unit is the person.
At small scale Make is remarkably cheap. Core at $9 covers a great deal of genuine work, and the free tier is enough to evaluate the product properly rather than merely look at it.
The crossover depends on headcount. One or two people running heavy automation is cheaper on Make almost regardless of volume, because $9 to $16 buys a lot of operations. Twenty people each automating their own work is cheaper on Power Automate, because Make's cost would be driven by the combined volume while Power Automate simply counts the twenty.
The subtlety on Make is that operations count per module per run, so a scenario with thirty modules consumes thirty operations each time it fires. Elaborate scenarios are where the bill grows, and it grows with sophistication rather than only with volume.
The same workload, priced both ways
Three shapes, using the published rates.
One operations person, a 12-module scenario running 500 times a month. Make: 6,000 operations, which sits inside Core at $9. Power Automate: one Premium licence at $15. Close enough that either is fine, with Make marginally ahead and considerably more pleasant for a scenario of that size.
A four-person team, several scenarios, 40,000 operations a month between them. Make: Pro at $16 or Teams at $29 depending on the seat and feature needs. Power Automate: four licences at $15, so $60. Make is meaningfully cheaper, and this is its sweet spot — a few builders producing automation for everyone else.
Twenty-five staff each running two or three personal flows. Power Automate: 25 licences at $15, so $375 a month, and nobody thinks about volume again. Make: the combined operation count across twenty-five people's automations is hard to forecast and the seat model on Teams starts mattering. Power Automate is the more predictable answer, and predictability is what you are buying at that headcount.
The rule underneath: Make prices your throughput, Power Automate prices your org chart, and neither vendor will tell you which of those is growing faster in your business. A few builders serving many colleagues is Make's shape. Many colleagues each serving themselves is Power Automate's.
The building experience
Make's canvas is the better environment for complex logic, and this is not a close call. Branching is drawn rather than described, routers split paths visibly, iterators and aggregators handle lists properly, and the data mapping between modules is genuinely strong. For a process with real conditional complexity, being able to see its shape is worth a lot.
The cost arrives with size. A scenario of forty modules across six branches requires scrolling and zooming to comprehend, and the visual tidiness starts to matter in a way unrelated to the work. Teams frequently find that the person who built a large scenario has become the only person who can safely change it.
Power Automate's designer is more conventional, a vertical list of steps with conditions and loops nested inside. It is less pleasant for complex branching and more familiar to anyone who has used similar tools. Its expression language is capable and is where the learning curve bites: people comfortable with Excel formulas adapt quickly, and people who are not find it unwelcoming.
Both are configuration. You hold the process in your head and translate it into the tool's vocabulary, which is where the time goes and where detail gets lost. That translation is what a conversational builder removes: on CodeWords you describe the process in plain language and Cody, the automation builder, builds it, connects it to the tools you already use, and deploys it. Automations connect to more than 3,000 integrations. The free plan covers light use, with Pro at $39 per month and Business at $100 per month as usage grows; details are on the pricing page.
Inside Microsoft, and outside it
If your process touches SharePoint, Teams, Outlook, Dataverse, or Excel in OneDrive, Power Automate reaches them as first-party systems. It sees events and properties that third-party connectors do not surface, authentication is the tenant's own, and permissions follow the user. Approvals in Teams, document workflows in SharePoint, and anything involving Dataverse are meaningfully easier here.
Make connects to Microsoft services too, through the same public APIs available to everyone. That works well for ordinary operations and stops short of the deeper integration points.
Away from Microsoft the advantage reverses. Make's catalog across the wider SaaS world is broad and its connectors are generally pleasant to work with, particularly for marketing, e-commerce, and content tools. Power Automate covers a lot of non-Microsoft software and puts some of it behind the premium licence.
The governance point is worth stating separately, because it decides some organizations outright. Power Automate flows sit inside the Microsoft 365 tenant, so data loss prevention policies, conditional access, audit logging, and the compliance tooling already in place apply without anything additional. For a regulated business already on Microsoft, adopting Make means having a conversation with security that adopting Power Automate does not.
Error handling and reliability
Make gives you explicit error handling as part of the canvas: error handler routes attached to modules, with directives for how to respond. It also keeps detailed execution history showing exactly what data passed through each module, which makes debugging genuinely straightforward. Incomplete executions can be stored and retried.
Power Automate offers configure-run-after settings on each action, scopes for grouping steps with shared error handling, and retry policies. Its run history shows inputs and outputs per step. It is capable and rather more scattered: the controls live in several places rather than being part of one visible structure.
Make is the more pleasant of the two to debug, largely because the visual execution history matches the visual build. Power Automate is the more capable when you need governance around failures rather than just recovery from them.
Which to choose
Power Automate if your organization runs on Microsoft 365, if you need desktop automation for systems with no API, if you have many light users rather than a few heavy ones, or if security and compliance need this inside the existing tenant.
Make if you want the best visual builder in the category, if your stack is mostly outside Microsoft, if a small number of people build on behalf of everyone else, or if you want to start for $9 rather than negotiate licences.
Look wider if the constraint is that the people who understand the processes cannot build in either tool. Both require learning a builder and translating the process into it, and that translation is the bottleneck far more often than any missing feature.
Frequently asked questions
Which is cheaper?
Make, usually, for a small team doing a lot; Power Automate for many people each doing a little. Make starts at $9 a month against $15 per user, so a two-person team is dramatically cheaper on Make and a thirty-person rollout is cheaper on Power Automate. Count people and monthly runs before comparing headline prices.
Does Make work with SharePoint and Teams?
Yes, through Microsoft's public APIs, which covers the common operations well. It does not reach the deeper tenant-level events and properties that Power Automate sees natively, so anything depending on those is easier on Power Automate.
Can Make do desktop automation?
No. Make connects systems through APIs. Interface automation of software with no API is Power Automate's desktop flows or a dedicated RPA product.
What is an operation in Make?
Roughly one module performing one action. A scenario with ten modules that runs once consumes about ten operations. This is why elaborate scenarios raise cost with sophistication rather than only with volume, and why consolidating modules is a genuine optimization.
Which handles large lists better?
Make, comfortably. Its iterators and aggregators are built for processing collections and its data mapping between modules handles nested structures well. Power Automate can loop over collections and is more awkward about it, particularly where the items need reshaping.
Is Power Automate included in my Microsoft 365 licence?
Some plans include limited capability for flows using standard connectors, which is genuinely useful for simple work and runs out quickly. Premium connectors, desktop flows, and process mining need the $15 per user per month Premium licence.
Can I move scenarios from one to the other?
Not directly. Each expresses logic in its own format, so a migration is a rebuild. What transfers is the description of what the process does, which is a good argument for keeping that written down independently of whichever tool currently runs it.
Which is better for approvals?
Power Automate, if the approvals happen in Teams or Outlook, because approval is a first-class action there with a proper interface and a record of who decided what. Make can build an approval loop through messages and webhooks, and it is more work and less tidy.
Does Make have a free tier worth using?
Yes, unusually so. Make's free plan is enough to build and run a real scenario rather than merely look at the interface, which makes it genuinely useful for evaluating the product. Power Automate offers a 30-day trial plus whatever limited capability your Microsoft 365 plan includes.
What happens to Make scenarios if I stop paying?
They stop running, and the definitions remain in the account. Power Automate flows likewise stop when a licence lapses, with the additional wrinkle that flows owned by someone who leaves the organization can stop when their account is disabled. Reassigning ownership before someone departs avoids an unpleasant surprise.
Which is better for high-volume data processing?
Make, generally. Its iterators and aggregators are built for working through collections, and operations are counted in a way that rewards batching. Power Automate handles volume and applies per-user request limits that a heavy data job can reach.