Building my first scenario in Make and stuck trying to test the full flow

Beginner Questions
, , , ,
1

Scenario goal: Automatically parse a Nubank (Bank app) CSV transaction file and create one row per transaction in a Notion database.

Modules:

  1. Google Drive – Watch Files in a Folder (trigger, polling every 15 min)
  2. Google Drive – Download a File (maps File ID from module 1)
  3. CSV – Parse CSV (uses toString(2.data) to convert binary, headers: Data, Valor, Identificador, Descrição)
  4. Flow Control – Iterator (array from module 3)
  5. Notion – Create a Data Source Item (maps fields from iterator)

The problem:
When I click “Run once”, only module 1 executes (with a green check). Modules 2–5 never run. The History log shows the trigger completes successfully with 0 data size, meaning it checks the folder but finds no new files to process — even though I’ve uploaded CSV files to the watched folder.

make
make1844×978 94.7 KB

4

  1. Right-click module 1 → Choose where to start

  2. Select All files or a specific date before your CSV uploads

  3. Click Run once

  4. Upload a brand-new CSV to the watched folder after Run once is waiting, or use an older start point

  5. Confirm module 1 outputs at least 1 bundle; only then modules 2–5 will run

Also check these common blockers:

  • The watched folder is the exact folder, not a parent folder.

  • Files are real .csv files, not Google Sheets.

  • If using Shared Drive, the connection/module supports Shared Drives and the correct drive is selected.

  • The scenario has not already “seen” those files in a previous test.

  • File type/filter in module 1 is not excluding CSVs.

For testing, the fastest fix is: delete/re-upload one CSV with a new filename after clicking Run once. Then module 1 should emit the file bundle and module 2 will download it.

5

also see https://skills.make.com/ if you are using Claude or ChatGPT/Codex these skill scan help you debug or solve.