How to filter out unwanted replies based on a growing keyword list (without using lots of operations)?

Hey everyone,

I’m using an API from a cold email software to pull in untracked replies — mostly automated emails like DMARC reports, out-of-office replies, warmup replies, and other junk.

In very rare cases, some important human replies end up in this bucket. So I’m building an automation that sends only the potentially relevant replies to Slack.

What I’m trying to achieve:

I want to filter these replies based on their text content — specifically, I want to ignore anything that contains certain keywords like:

DMARC

domain report

out of office

postmaster

(this list will keep growing…)

Right now, I’m using a basic filter in Make (see screenshot), but:

  • I don’t want to manually add a filter rule for every new keyword
  • I want to avoid iterating over every reply and every keyword because of the high operation cost
  • Ideally, I want to store the keywords in Airtable or Google Sheets and check if a reply does not contain any of those

And do that in one single filter condition if possible (no looping!)

Is there a way to:

  • Pull the keyword list as a single array (or joined string)?
  • Use a single contains() or match() expression to check if a reply’s text contains any of the keywords?
  • Apply that in a filter step without needing a separate operation per keyword or per check?

This would help me scale this automation without burning through operations, and keep the keyword list easily manageable.

Appreciate any guidance or examples.

See this screenshot for a representation of what I hope to achieve.

Absolutely.

You can use a “Text operators: Does not match pattern (case-insensitive)” operator, with the following pattern:

\b(DMARC|domain report|more|keywords|goes|here)\b

(copy-paste the above into the field, or type it exactly as shown)

Hope this helps! Let me know if there are any further questions or issues.

@samliew