Save to Memory Block
Save to Memory writes a value or template result to a memory key so a later run can read it.
At a glance
What it does
Save to Memory stores state outside the current run. It is useful for checkpoints, last-seen values, and deduplication across runs.
Use it after a successful step so memory only changes when the relevant work completed.
Use it when
- Remember the last processed item ID.
- Store the latest known price or status.
- Save a checkpoint for the next scheduled run.
Do not use it when
- The value should only exist in the current run.
- You need changed/unchanged routing; use Track Change.
- You are saving sensitive data that should not persist.
Settings
Save to Memory needs a key, value source, optional group/scope, and optional skip-empty behavior.
Memory key
Required: Yes- Variables
- Yes
- Description
- Key to write in persistent memory.
Value mode
Required: Yes- Variables
- No
- Description
- Choose variable or template input.
Value
Required: Yes- Variables
- Yes
- Description
- Variable name or template to save.
Skip empty
Required: No- Variables
- No
- Description
- Avoids overwriting memory with an empty value.
Outputs
Save to Memory reports the saved key and value metadata.
savedbooleanWhether memory was written.
keystringMemory key that was written.
Example Flow
Save memory after a successful extraction or export when the next run should remember progress.
- Extract or compute the value.
- Run the success path.
- Save the value to memory at the end.
Errors and fixes
Start with symptoms visible in the editor or Inspector, then check the earliest Block that produced the wrong page state, variable, or output.
The old memory value remains.
- Check
- Check whether Skip empty prevented the write.
- Fix
- Provide a non-empty value or disable Skip empty when overwriting is intended.
Read from Memory cannot find the value later.
- Check
- Check key, group, and scope on both Blocks.
- Fix
- Use the same memory address in Save and Read.
Local and cloud runs
Save to Memory is intended for persistent project or schedule state. Verify schedule scope when using scheduled Cloud Runs.