Editorial review: 3 November 2026. This guide describes a proposed workflow; HD Media has not conducted hands-on tests for this article.
Make and Zapier can both move information between the apps a small agency uses. The useful question is what happens when a client project is approved, a folder must be created, and somebody needs to know what to do next.
Start with one workflow and test it in both products. A simple trigger and one action may fit a basic Zap. A workflow with several decisions and updates needs a plan that supports them. Make's and Zapier's usage units are different, so a headline comparison of “credits versus tasks” does not tell you which will cost less. This guide uses current documentation; HD Media has not run both scenarios.
The common test: a new client is approved
Imagine a three-person agency with a spreadsheet of signed engagements. Each new row has project_id, client_name, service, owner_email, and approved_to_start. When approval is Yes, the workflow should:
- Check that the project ID and owner exist.
- Find or create the client's project folder.
- Write the folder link to the correct project row.
- Notify the owner of the next human action.
- Record an exception if a connection or notification fails.
This is a proposed onboarding design; confirm the trigger, retry behavior and permissions before relying on it. Use a fictional client and internal addresses for the trial. Never place a client password in the trigger row.
What the published plans say
| Question | Make | Zapier |
|---|---|---|
| Free-plan allowance shown on 3 Nov 2026 | 1,000 credits/month | 100 tasks/month |
| Free scheduled polling shown | 15-minute minimum interval | 15-minute polling interval |
| Free workflow scope | Visual scenarios, filters and routers; account limits apply | Two-step Zaps: one trigger and one action |
| Paid entry shown | Core displayed at $12/month for a selected 10,000-credit allocation | Professional displayed from $19.99/month |
| Usage unit | Many module actions consume credits; some advanced features vary | Successful Classic Zap actions count as tasks; other products have different rules |
Sources: Make pricing, Make credits, Zapier pricing, Zapier Free plan, and Zapier task usage. The displayed entry prices depend on the selected allowance and billing interval; check checkout before buying.
Do not divide 1,000 by 100 and call Make ten times cheaper. A Make scenario can spend credits checking for new records and executing several modules. Zapier's classic task rules count successful actions differently. In both products, the exact workflow and plan matter more than the nominal unit count.
Build the test in Make
The proposed modules are Google Sheets Watch New Rows, a filter for approved rows, Drive Search for Files/Folders, Drive Create a Folder when needed, Google Sheets Update a Row, and a notification. Make documents these modules in its Sheets and Drive catalogs.
Keep a stable project ID. A new-row trigger may process a row before a later approval edit; decide how approved records enter the workflow. Searching and then creating a folder is not an atomic duplicate lock, so design and test idempotency and concurrent runs. Search for an existing folder before creating one. If the notification fails after folder creation, a retry should use the same folder rather than creating a duplicate. Make's scenario settings describe incomplete executions and optional ordered processing; choose the behavior deliberately. Scenario settings.
Build the same test in Zapier
Start with the same approved project record and attempt the same folder, row-update, and notification actions. Check whether the needed apps and multi-step workflow are included in the plan you would buy. Zapier's current Free documentation limits a Zap to a trigger and one action, so this complete process requires a different arrangement or a paid plan. Zapier Free-plan features.
Do not treat a version that merely creates the folder as equivalent to a version that also updates the source record and handles failure. Match outcomes first, then compare cost.
Count actual usage over a representative week
Run sample events at a realistic frequency, then record the product's own usage meter. Include days with no new clients if polling consumes allowance. Include duplicate IDs and a failed connection. Compare the two results against the plans you might purchase.
| Measure | Why it matters |
|---|---|
| New approved projects | How many true starts occurred |
| Trigger checks | Scheduled checks may consume allowance |
| Actions per project | Search, create, update, and notification are different steps |
| Error retries | Recoveries may consume allowance or require manual work |
| Time to repair | A cheap subscription can still be costly to maintain |
The point of this test is to discover your own bill and operating effort. The example project above is a design, not measured performance or a recommendation derived from running the products.
Which should a small agency shortlist?
If you only need a trigger and one action, try the free scope of both and see which connection is easier for your team to maintain. If you need branching and several outcomes, compare the relevant paid plans through the full scenario. If your team already has functioning automation, account for migration work before changing it.
Start with the onboarding checklist so you automate an agreed process. A tool should create the folder and hand off the task; a person still confirms scope, approves dates, and checks sensitive access.