Understanding how data flows through your boards requires tracing the lifecycle of individual columns across multiple Zapier workflows. Because all Zaps are stored as JSON files within subfolders in the GitHub repository, you can globally search this codebase to pinpoint exactly which automation reads, writes, overwrites, or clears a specific data field during an order's lifecycle.
This guide provides the methodology for tracking down column changes and troubleshooting data inconsistencies.
Tracing a Column's Lifecycle
To determine what affects a specific column (e.g., "Column X") or data point ("Data Y"), you must search the raw Zap configurations.
- 1
Identify the Column ID
First, find the underlying monday.com column ID (e.g.,
text_mky5xz73). Relying on the human-readable column name is often unreliable, as names can change, but the underlying ID remains static in the Zap's JSON payload. - 2
Search the GitHub Repository
Use GitHub's global search or your local IDE to search the repository for the exact column ID. This will return every Zap (JSON file) that interacts with this column.
- 3
Determine the Operation Type
Open the matching JSON files or their corresponding documentation and look at the action steps to determine how the Zap interacts with the column:
Read: Look for
Get Column Valuessteps. This means the Zap uses the column as an input (e.g., fetching raw address text).Write/Overwrite: Look for
Change Multiple Column ValuesorChange Column Valuesteps. This means the Zap is mutating the data.Clear: Look for write steps where the value is explicitly set to empty or a reset character (like
-).
If a column is acting as both an input and an output within the same Zap (e.g., reading a raw address, parsing it, and overwriting the same column with the cleaned version), it will appear in both the Get and Change steps of that workflow.
Visualizing the Data Flow
When you identify a Zap that mutates your target column, map out its pipeline to understand the transformations the data undergoes.
For example, if you are tracing the text_mky5xz73 (Address) column, searching the repo might lead you to the NOE Address Fixer Zap. Its lifecycle looks like this:
flowchart TD
A["Webhook Trigger (Catch Hook)"] --> B["Get Column Values (monday.com)"]
B -->|Reads raw address| C["Python Code Step (Format Address)"]
C -->|Cleans & splits data| D["Google Geocoding API"]
D -->|Returns Lat/Long| E["Change Multiple Column Values"]
E -->|Overwrites address, sets Location| F[("monday.com Orders Board")]By mapping the Zap, you can see that if an address is formatted incorrectly, the issue likely originates in the Python Code Step, not the initial data entry.
Troubleshooting Common Data Issues
When an order is experiencing data anomalies, use the repository search method above to diagnose the root cause.
Data is being overwritten unexpectedly
Cause: Multiple Zaps are triggering at different stages of the order lifecycle and writing to the same column ID.
Action: Search the repo for the column ID. Identify all Zaps with a Change Multiple Column Values step targeting it. Check the triggers for those Zaps to see if a downstream automation is firing out of order and overwriting manual corrections.
A column is suddenly blank or cleared
Cause: A Zap is explicitly wiping the column to prepare for new data, or a mapping error is passing a null value.
Action: Search for the column ID and look for Zaps that update the column with an empty string or a reset character (e.g., "-"). For example, the Address Fixer Zap explicitly clears several text/color columns to wipe out prior manual entries.
Formatting or casing is incorrect
Cause: The data is being passed through a Formatter or Code step that is applying incorrect logic.
Action: Trace the column back to the Zap that writes to it. Inspect any intermediate steps (like Python, JavaScript, or Zapier Formatter steps). For instance, if unit numbers like "none" or "n/a" are appearing, check if the Python parser's exclusion list is failing to filter those placeholder values.
Coordinates or Location data is missing
Cause: The external API responsible for enriching the data failed, or the API key is invalid/rate-limited.
Action: Locate the Zap handling the enrichment (e.g., Google Geocoding API). Check if the step relies on hardcoded credentials or if the input data (the concatenated address string) was malformed before being sent to the API.
Documenting Known Transformations
To speed up future troubleshooting, it is highly recommended to maintain a mapping table of critical columns and the specific Zaps that own their transformations.
| Column Name | Column ID | Primary Read Zap | Primary Write/Mutate Zap | Known Transformations |
|---|---|---|---|---|
| Raw Address | text_mky5xz73 | NOE Address Fixer | NOE Address Fixer | Title-casing, abbreviation expansion, unit extraction. |
| Location | location_x89 | N/A | NOE Address Fixer | Populated with Google Geocoding Lat/Long. |
| Status | status_c21 | Order Router | NOE Address Fixer | Reset to blank (-) after address parsing. |
Always verify the active status of a Zap directly in the Zapier dashboard. A Zap may exist in the GitHub repository as a JSON backup but be paused or disabled in production.