Make vs win.sh: Two Operating Models
Make centers a visual no-code automation canvas. win.sh runs a 24/7 business loop inside an authority matrix.
Founder, Task Machine
A Make scenario does exactly what you drew: when an order arrives, add a row, send a message, update the CRM. An agent loop decides what to do based on what it sees. Make and win.sh represent those two ways of running repeated work, and the right one depends on how much judgment the work needs.
Make is a visual automation platform. You build scenarios on a canvas from triggers, modules, routers, and data mappings across 3,000+ pre-built apps, and you add AI modules or Make AI Agents where you want judgment. win.sh has no canvas. Its 24/7 loop watches the accounts you connect, proposes the next move, and acts inside an authority matrix you set.
How do Make and win.sh compare?
| Dimension | Make | win.sh |
|---|---|---|
| Core abstraction | A scenario of triggers, modules, routers, and data mappings | A 24/7 loop that monitors connected business accounts and acts |
| Human role | Draw every route and inspect each scenario run | Set rules per type of work, then review decisions and receipts |
| Execution | Hosted visual automation across 3,000+ apps, with AI agents you add to scenarios | Cloud sandboxes connected to accounts the customer owns |
| Strongest fit | Predictable data flow between many apps | Founders who want the business watched with little daily involvement |
| Recurring work | Scenarios run on schedules or app events exactly as drawn | Runs continuously, with recurring workflows and a morning brief |
| Control | Every path is visible on the canvas before it runs | Approval gates, authority levels, a hard budget cap, and receipts |
| Main cost | You build and maintain each path, including review steps and exceptions | The loop acts before the brief, so much of the review happens afterwards |
| Commercial model | A free plan with no time limit, and paid plans priced by usage | Self-serve monthly budget from $50 to $10,000, no revenue share |
Prices and features were checked against each vendor's own pages on 6 October 2026.
Who does Make fit?
Make fits operators who know the steps and want them to run the same way every time. Copying new Shopify orders into a spreadsheet, routing form leads by country, or posting a Slack alert when an invoice is overdue all suit a scenario, because each step can be drawn and tested in advance. The canvas shows the whole data flow, which makes problems easy to trace.
The canvas also shows the limit. Someone has to draw every branch, exception, and approval, and when a supplier email needs judgment, the scenario only knows the routes it was given. AI steps help inside a route, and keeping the routes current as the business changes stays your job.
Who does win.sh fit?
win.sh fits founders who want the loop to work out the steps. It watches connected accounts such as Stripe or Plausible, decides which change is worth acting on, and proposes or takes the next step without a drawn route. A Telegram brief and the Decisions tab report each morning.
Trust is set per type of work, and your approvals, edits, and rejections become rules that widen or narrow what the loop may do. A hard monthly cap limits spend, receipts show the dollars behind each action, and win.sh takes no revenue share. You trade Make's predictability for judgment, and you review much of that judgment after it has acted.
Where does each product put human control?
Make puts control in the design. Nothing happens that is not on the canvas, so a reviewer can read a scenario and know every possible outcome before it runs.
win.sh replaces drawn routes with standing policy. Policy sets how far each type of work may go alone, and approval gates still hold spending, outreach, publishing, and sensitive changes. To compare them on one job, take an angry customer email: Make follows whatever route you drew for it, and win.sh decides on a response inside its authority. Ask which failure you could live with.
Where does Task Machine fit?
Task Machine keeps the drawn structure and adds judgment where you want it. A workflow is an explicit graph with branch conditions, like a scenario, but its steps can be agents that read, draft, and decide, followed by approval, question, or verifier steps before anything moves on. Whoever approves sees each decision in one Inbox, next to the run it belongs to.
Make stays the simpler tool for pure data movement across its larger app catalog. Task Machine fits the jobs in between, such as answering supplier emails or preparing a weekly report, where an agent should do the thinking and a person should approve the result.
Which should you choose?
Choose Make if the work is a fixed sequence of app steps that should run identically every time. Choose win.sh if you want the business watched around the clock and are comfortable with a loop deciding the next step inside rules you set. Choose Task Machine if the work needs an agent's judgment inside a structure you can read, with approvals before anything goes out.