MillerKnoll AI
← All case studies
Surveys / Insights · HR· Jul 2026 · 3 min read

Reading every survey comment — with the PII stripped out

Open-text comments are the richest and least-read part of an engagement survey. MK Comment Signal turns thousands of them into ranked themes backed by real verbatim evidence, with names, emails, and IDs redacted automatically.

The opportunity

Reading and coding thousands of open-text engagement comments by hand isn’t realistic for a People team — so the richest feedback in the survey mostly goes unread, and what does get read can’t safely circulate because it names individuals.

What AISE built

Upload the Perceptyx comment export and MK Comment Signal extracts the text for each survey question, identifies four to five recurring themes ranked by how many comments support each, and attaches at least a dozen verbatim quotes per theme as evidence — while automatically redacting names, emails, and employee IDs. An optional flow exports a formatted Word document to OneDrive.

The benefit to the business

What would take days of reading becomes a ranked, evidence-backed theme set a leader can act on — and because PII is stripped in the pipeline, sensitive feedback can be analyzed and shared without exposing anyone. It’s been in Production since January 20, 2026.