Derek Brown
All case studies

How will employees react before we hit send?

A message-testing method I built and use at a regional utility. It runs a draft all-employee message past a panel of employee roles and comes back with a send decision and specific fixes.

  • Change communications
  • Message testing
  • Manager readiness
  • AI-assisted workflow
ToAll employees
SubjectThank you for a long weekend

Saturday’s storm brought down lines across our service territory, and our people answered.

Line workers, field supervisors and contractor crews worked through the weekend to get customers back on safely. Leadership is grateful for your skill, your stamina and your commitment to safety.

Thank you for everything you do.

Send as writtenSend with fixesDon’t send
  • Field supervisor Pass 1
    My crew will ask why the call center isn’t in here.
  • Corporate support Pass 1
    Reads warm. Leadership sounds grateful.
  • Call center rep Pass 1
    We took the calls all weekend. We’re not in it.
  • Must fixName the call center. A message about the weekend leaves out the people who took the calls all weekend.
  • Should fixGive supervisors one sentence they can say in the yard when the question comes.
  • ConsiderSay when make-up overtime will be scheduled, or when people will know.

Illustrative run on a sample message. Panelists are roles with years of service, never real people. Six roles sit on each panel; three are shown. The output is a behavioral projection, not employee feedback.

The challenge

Most of what we learn about an all-employee message arrives after it’s sent: in the hallway, in the question a supervisor can’t answer, in the concern nobody raises out loud. By then it’s on the record.

What I built

A library of 20 workforce roles and a two-pass review: sealed first reactions, then a room where the roles hear each other. Every run ends in a send decision with ranked fixes, written at executive, standard or deep-dive length.

What happened

Its read informed a live send decision reviewed by a communications director. I version it and recalibrate it after real events. It’s a projection, not employee feedback, and its accuracy isn’t yet measured.

The method stays fixed; the organization is a layer. Swap in a company’s workforce roles, recent history and channels, and the same review reads messages anywhere.

  • Regional utility, in use
  • Your organization