Question about "Data is confidential" setting and replaying failed executions

[koto]

What is your goal?

I am building a webhook scenario that receives sensitive data (personal information etc.) from an external form service.
For privacy and compliance reasons, I want to enable the “Data is confidential” option in Scenario settings so that Make does not store any processed data.

What is the problem & what have you tried?

The problem is:
If I turn on “Data is confidential”, will I still be able to replay or re-execute failed scenario runs (from History or Incomplete Executions)?
If replay is possible, for how many days is the data retained?
I have read the official Scenario settings documentation, which only states:
“If enabled, there are very limited options to solve errors that occur in a scenario execution.”
However, it does not clearly say whether replay/re-execution becomes completely impossible.
I have not tested it yet with real data because this is a production scenario handling sensitive information, and I want to confirm the exact behavior first.

[Henk-Operative]

Hi @koto,

If you enable data is confidential, Make will not store data.

But Make does store the execution, so you can see when the scenario ran and (in case of an error) see what module caused the error. However, you will not be able to see any processed data.

Replay is possible but you will see: ‘The data contained in the scenario is marked as confidential and therefore cannot be displayed.’

Cheers,
Henk

[koto]

Hi Henk,

Thank you very much for your quick reply and the clear explanation!

I tested the “Data is confidential” setting in a test scenario, and I noticed something that doesn’t fully match the documentation or your description:

• In some failed executions, Replay is still possible (I see the message “The data contained in the scenario is marked as confidential and therefore cannot be displayed.”).
• In other cases, Replay is completely impossible (the Replay button doesn’t appear at all).

This left me with three specific questions before I enable it in production:

1. What are the exact conditions that determine whether Replay is possible or not?
   (For example: History vs. Incomplete Executions, type of error, “Store incomplete executions” setting, time since execution, or something else?)

2. When Replay is possible, does this mean the original data bundles are still retained somewhere (even if hidden)?

3. If any data is retained, what is the exact retention period?

I haven’t enabled this in any live scenarios yet, as we handle sensitive personal data via webhook and need to be 100% sure about the behavior.

I would really appreciate your clarification on these points!

Best regards,

[Henk-Operative]

Have you tested the same scenario, using the same payload? And the same ‘data is confidential’ setting at the time of first execution?

If data is stored in the execution logs, the retention period is based on your tier: