Make vs Power Automate for UK Businesses

The shortest useful answer is this: choose Power Automate when the workflow is deeply tied to Microsoft 365, Dynamics, Power Platform or Windows desktop software. Choose Make when the workflow crosses a mixed set of cloud applications and a visual, platform-neutral builder is more valuable.

That rule will not settle every case, but it removes most of the noise. The tool is only one part of the system. Ownership, error handling, permissions and the way usage is billed decide whether an automation remains useful after launch.

Make vs Power Automate at a glance

Do not select from this table alone. Build one representative workflow and price its real execution pattern.

What is Make?

Make is a visual automation platform for connecting cloud applications and APIs. A scenario is built from modules. Those modules can retrieve data, transform it, branch on conditions and send the result somewhere else.

The canvas is particularly helpful when a workflow crosses several unrelated products. A team can see the route and inspect where data changes shape.

Make’s current pricing page describes its usage unit as a credit and says a module action, such as adding a row or fetching data, generally consumes one credit: Make pricing and credit model. That means one business event may consume several credits as it travels through several modules.

What is Power Automate?

Power Automate is Microsoft’s automation platform. It covers cloud flows and desktop automation, and sits alongside Microsoft 365, Dynamics 365, Dataverse and the wider Power Platform.

Its advantage is not simply that it has connectors. It can fit the identity, administration and data model a Microsoft organisation already uses. It can also automate desktop interfaces where an old system has no suitable API.

Licensing needs careful reading. Microsoft’s current documentation separates user and capacity licensing and explains that some Microsoft 365, Dynamics and Power Platform licences include limited Power Automate capabilities: Power Automate licence types.

Choose by workflow shape

Choose Make when the workflow is cloud-first and mixed

Make is a strong candidate when the path looks like:

• website form;
• independent CRM;
• enrichment or document service;
• spreadsheet or database;
• team notification.

Its visual routes are useful for branching and data transformation. The trade-off is that a long scenario can consume many credits and become difficult to reason about if nobody owns it.

Choose Power Automate when Microsoft is the operating system of the business

Power Automate is a strong candidate when the path lives inside:

• Outlook and Teams;
• SharePoint and OneDrive;
• Dynamics 365;
• Dataverse;
• Excel-based processes;
• Windows desktop applications.

The deciding factor may be governance rather than the builder. Existing identity, environment and administration controls can reduce the number of separate systems a team has to manage.

Choose Power Automate for desktop RPA

If a task requires clicking through a legacy Windows application, Power Automate Desktop belongs on the shortlist. But treat interface automation as fragile. A changed label, screen or timing assumption can break it. Prefer a reliable API when one exists.

Compare cost without using a headline price

Headline plan prices are not a fair comparison because the billing units differ.

For Make, map:

1. business events per month;
2. modules run for each event;
3. retries and error-handling routes;
4. polling or scheduled checks;
5. peak volume.

For Power Automate, map:

1. who runs the flow;
2. which connectors and premium features it uses;
3. whether it runs attended or unattended desktop automation;
4. whether a process or capacity licence is needed;
5. which rights are already included in existing subscriptions.

Then confirm the result against the vendors’ current licensing pages. Both products can change their plans and usage rules.

Data protection and security

Neither tool is “GDPR compliant” in the abstract. Compliance depends on your contract, configuration, data, purpose and controls.

Before connecting live systems, document:

• the personal data each step receives;
• the lawful purpose and minimum fields required;
• storage and processing locations for the chosen plan and environment;
• credentials and service accounts;
• retention in execution histories;
• who can edit, run and inspect the automation;
• the response when a connection is revoked or a person leaves.

Avoid copying entire records when the next step needs one field. Reducing the data surface is usually safer and easier to debug.

Reliability matters more than the demo

A successful test run proves the happy path. Production needs more:

• idempotency, so a retry does not create duplicate work;
• explicit handling for rate limits and unavailable services;
• a queue or review route for ambiguous records;
• alerts that name the failed step and affected item;
• an owner who can repair the workflow;
• a manual fallback for critical processes.

Also keep business rules separate from accidental tool behaviour. If a workflow decides whether to contact, refund or reject someone, write the rule down and test it.

What are the downsides of Make and Power Automate?

Make can become difficult to govern when visual scenarios grow into long, branching systems. Credit consumption can also be surprising if one business event triggers many module actions, polling checks or retries.

Power Automate licensing and environment design can be harder to understand than the first flow suggests. Premium connectors, process licensing, desktop automation and the wider Power Platform administration model need an owner who understands the organisation’s Microsoft setup.

Both platforms can hide operational complexity behind a friendly builder. Documentation, alerts, version discipline and a named maintainer are still required.

A five-day evaluation

Use the same small process in both tools.

Day 1: define the outcome

Write the trigger, required fields, allowed actions and success condition.

Day 2: build the happy path

Use test accounts and non-sensitive data.

Day 3: force failures

Remove a required field, expire a connection, create a duplicate and make the destination unavailable.

Day 4: hand it to the owner

Ask the person who will maintain it to diagnose a failed run without the builder’s help.

Day 5: compare total fit

Score build time, execution cost, permissions, error recovery, auditability and maintenance. The best tool is the one that survives this day, not the one with the prettiest first demo.

The verdict

Power Automate usually has the clearer case inside a Microsoft-centred organisation or where desktop automation is essential. Make usually has the clearer case for visual workflows across a mixed cloud stack.

If the workflow is critical, high-volume or heavy on custom logic, normal application code may be the simpler long-term choice. No-code is not a requirement.

For a wider comparison, read n8n vs Zapier vs Make and common AI automation mistakes.

If you want help choosing around a real workflow, Get unstuck.