Make is a visual automation platform. You build scenarios by connecting modules on a canvas, so that an event in one app can trigger reads, writes and transformations across others. In 2026 the company has been adding AI features on top of that base, including an assistant inside the scenario builder, AI agents, and tools for seeing how automations connect.
This review covers what Make is, how its credit-based billing works, what its release notes show for 2026, what we could and could not confirm about privacy, and who might prefer something else. We lean on Make’s own help centre and blog, and where we could not open an official page we say so.
How we researched this: we did not run hands-on tests for this article. It is based on the vendor’s own documentation and public sources, checked on 11 November 2026. Features and prices change, so confirm details on the vendor’s site before you buy.
Key takeaways
- Make moved its billing unit from operations to credits in August 2025. Existing operations converted one to one, and most actions still use 1 credit.
- Make’s built-in AI Provider can use more than one credit per action because billing tracks actual AI usage costs, according to its help centre.
- The official pricing page lists Free, Core, Pro, Teams and Enterprise plans, and the Free plan includes 1,000 credits a month. We could not read paid prices from the page text, so we give none.
- The 2026 release notes show a steady stream of changes, from If-else and Merge in March to event subscriptions in October, so the product is moving quickly.
- Unused credits expire at the end of the term, which matters if your usage is spiky.
What it is
Make is a no-code and low-code automation service built around scenarios. A scenario is a chain of modules, and each module performs one step such as reading a spreadsheet, creating a record or transforming data. Make’s pricing page says that reading, searching, creating, updating, deleting, transforming, aggregating and iterating data each use credits, which gives a sense of the kinds of steps a scenario contains.
The company has been positioning itself around AI agents. At its Waves ’26 event on 20 October 2026, Make announced MakeOS and described a “glass box” approach in which every agent, process and action is meant to be visible, permissioned and reversible. MakeOS is described as a new product for running company work through AI agents and is open to waitlist sign-ups, so it is not something you can buy today on the evidence we saw.
Who it is for
The plan names suggest the intended audiences. Free is for trying the product. Core, Pro and Teams scale up through individuals, small groups and collaborating teams. Enterprise is for larger organisations, and the pricing page also names a Start-up Plan that gives companies under 20 employees a discount. The 2026 release notes lean toward teams and organisations: custom roles, private spaces, audit logs and single sign-on routing all appear.
Make suits people who prefer to see a workflow as a diagram and who want to inspect data as it passes between modules. If you are comfortable thinking about branching, iteration and data shapes, that style may feel natural. If you want the simplest possible two-step connection, a simpler tool might do the job with less to learn.
Key features (documented)
The table lists features and releases that Make’s own release notes and blog describe in 2026. Dates are those shown in the release notes.
| Date (2026) | What Make documents |
|---|---|
| 10 March | If-else and Merge, for splitting a scenario flow and rejoining it |
| 24 March | Make Grid update: a visual map of AI and automation objects, with edit layout and real-time collaboration, available to all paid plans |
| 17 April | Make CLI launched for all users, for accessing Make resources from a terminal |
| 25 May | Make Skills, which help AI assistants perform tasks in Make |
| 17 July | Private spaces, a personal workspace for scenarios and connections |
| 5 August | Custom organisation and team roles, for Enterprise users |
| 12 August | Maia, an AI assistant in the Scenario Builder that creates and troubleshoots scenarios from prompts |
| 8 October | Event subscriptions: signed real-time platform notifications and delivery logs |
| 9 October | AI Agent fallback connections and prompt caching, SSO login routing, and MCP tools for audit logs |
What was announced but not yet generally available
The Waves ’26 post separates what is available from what is coming soon. According to that post, agent orchestration, Maia’s Plan mode and Memory, and Knowledge are available, while human-in-the-loop approvals, data pinning, Functions, Endpoints and MCP toolboxes are coming soon. Agent evals and AI spend tracking are also listed as coming soon, and Make Process Intelligence (task mining and process mining, built on Celonis technology) is waitlist-only. If any of those are the reason you are considering Make, wait for general availability or ask the vendor for dates.
Seeing how automations connect
Make Grid is described as a visual map of a team’s automation and AI landscape. The March 2026 update says some connected objects, such as Google Sheets, Google Docs and JotForm, can be viewed and edited from the grid, while others such as Google Forms and Typeform are view-only. This is useful mostly for teams that have accumulated many scenarios and need to understand dependencies.
Pricing model
Make bills in credits. Its help centre says the change from operations to credits took effect in August 2025, that existing operations converted at one to one (so 1,000 operations became 1,000 credits), and that current plans and pricing stayed the same at the time. The pricing page repeats the main rules, and we summarise them below as shown when we checked on 11 November 2026.
| Topic | What the official pages say |
|---|---|
| Plans named | Free, Core, Pro, Teams, Enterprise, plus a Start-up Plan for companies under 20 employees |
| Free plan | 1,000 credits a month, 1 MB of data storage, 1 data store |
| Normal action | Most actions use 1 credit |
| AI features | Make’s built-in AI Provider may use more credits; third-party AI apps still use 1 credit per operation |
| No credits used | Router module and the error handler modules Rollback, Break, Resume, Commit and Ignore |
| Run size | A single scenario run can use from two credits to thousands |
The pricing page also lists allowances per 10,000 credits a month: 5 GB of data transfer, 10 MB of data storage, 10 MB of incomplete executions storage (up to 2 GB maximum) and a webhook queue of 667 (up to 10,000). Those are useful numbers if you handle large files or heavy webhook traffic.
Expiry, extra credits and downgrades
Unused credits expire at the end of the term. Extra credits are sold in bundles of 1,000 or 10,000 and expire after one month on monthly subscriptions, or at the end of the year on annual Pro and Teams subscriptions. Core, Pro and Teams can enable auto-purchase of 10,000 credits. Paying annually in advance costs less than monthly. If you downgrade, active scenarios that use higher-plan features stop, and unused credits transfer to the new plan. Cancelling moves the organisation to the Free plan at the end of the billing cycle or annual term.
We did not manage to read plan prices from the official pricing page, because the text we could access did not list them. Third-party articles quote figures, but we have not verified them against Make, so we leave them out. Open the pricing page directly and read it alongside our guide to reading an AI tool’s pricing page.
Strengths
- Transparent credit rules. The pricing page lists which modules use credits and which do not, including the Router and error handlers.
- Visual debugging style. The canvas model lets you follow data module by module, which many people find easier to reason about for branching logic.
- Active development. The 2026 release notes show frequent additions, from flow control to a CLI and AI assistance.
- Governance direction. Custom roles, private spaces, event subscriptions, audit logs and SSO routing suggest attention to organisational needs.
- Overview tools. Make Grid helps teams see dependencies between automations.
Limits and cautions
- Credits can vary a lot per run. The pricing page says a single run can use anywhere from two credits to thousands, particularly with iteration over large data sets, so estimating needs takes care.
- AI credit use is less predictable. The built-in AI Provider may use more than one credit per action.
- Expiry. Unused credits do not carry over, so heavy and light months are not averaged.
- Some announced features are not live. MakeOS, Process Intelligence and several agent controls were waitlist-only or coming soon in the October 2026 post.
- Release pace cuts both ways. Frequent change means documentation and interfaces move, so check that tutorials match what you see.
- Some features sit on higher plans. The 5 August 2026 release notes say custom roles are for Enterprise users, and Make Grid updates are for paid plans.
Privacy and data
We could not open Make’s security, privacy or data-processing pages during this research, so we cannot report certifications, data location or AI training terms. The pages we did read mention governance tooling: the Waves ’26 post says agents act with their owner’s role, and the release notes list audit logs and MCP tools for audit logs. Those describe controls, not legal commitments.
If you plan to pass personal or confidential data through scenarios, ask Make for its data processing agreement, hosting regions and subprocessor list, and confirm whether any content is used to improve AI features. Our guide on protecting company data when using AI tools gives a checklist for this conversation.
| Privacy question | Status in our research |
|---|---|
| Security certifications | We could not confirm |
| Data hosting regions | We could not confirm |
| Use of content for AI training | We could not confirm |
| Audit logs for agents and MCP | Listed as available in the October 2026 post and release notes |
Alternatives
Zapier and n8n are the usual comparison points. Zapier bills by tasks, meaning successful action steps, and has a broad app catalogue. n8n offers a free self-hosted Community Edition under its Sustainable Use License as well as paid cloud plans. We publish a decision guide for Zapier, Make and n8n that explains how to choose using your own workflow, and separate reviews of Zapier and n8n. For a wider decision process, see how to choose an AI tool.
Who should and should not use it
Best for these people, if these conditions apply
- You like to build and inspect workflows visually, with branching and data transformation.
- Your team wants governance features such as roles, audit logs and an overview map, and is on a plan that includes them.
- You can estimate credit use per run and your workload is steady enough that expiry is not a problem.
Look elsewhere, or test carefully, if these apply
- You need self-hosting, which we did not find documented for Make.
- You need confirmed certifications or data residency now and cannot get them in writing.
- You are buying mainly for features listed as coming soon or waitlist-only.
Questions to ask before buying
- How many credits will my most complex scenario use per run, including iteration over large lists?
- Will I use the built-in AI Provider, and how many credits does a typical call use?
- Which plan includes the governance features I need, such as custom roles?
- What are the hosting region, data processing terms and AI training policy?
- Which of the features I want are available now, and which are coming soon?
- How will unused credits at month end affect my cost?
You can test many of these answers cheaply. Build one real scenario on the Free plan and watch credit use, following the approach in how to test an AI tool before you pay for it.
Frequently asked questions
Is Make free?
There is a Free plan. The pricing page says it includes 1,000 credits a month, 1 MB of data storage and 1 data store.
What is a credit in Make?
A credit is Make’s billing unit. For non-AI features and third-party AI apps, one operation uses one credit. Make’s built-in AI Provider may use more, according to its help centre.
When did Make switch from operations to credits?
Make’s help centre announced the switch in August 2025, with existing operations converted one to one.
Do unused credits roll over?
The pricing page says unused credits expire at the end of the term, and extra credits expire after one month on monthly plans or at the end of the year on annual Pro and Teams plans.
What is MakeOS?
Make describes MakeOS as a new product for running company work through AI agents. In its 20 October 2026 post it was open to waitlist sign-ups.
Can I self-host Make?
We found no self-hosting option in the pages we read. Confirm with Make if this matters to you.
Conclusion
Make’s documented strengths are a visual builder, transparent credit rules and a fast release cadence with growing governance features. Its cautions are variable credit use per run, expiring credits, and a number of headline AI features that were still coming soon or waitlist-only in October 2026. We could not confirm its privacy terms from official pages, so ask for them.
Before you decide, build a real scenario on the Free plan, read the pricing page for your own numbers, and compare against alternatives using how to compare AI tools side by side fairly.
Sources
- Make pricing page, 11 November 2026
- Make Help Centre: Introducing credits, 11 November 2026
- Make Help Centre: 2026 release notes, 11 November 2026
- Make blog: Waves ’26, the glass box for the agentic era (20 October 2026), 11 November 2026
- Make blog: Make Grid update (24 March 2026), 11 November 2026
