Guide

Why your team keeps re-arguing decisions it already made

By Craig Muller · Last updated
Settled decisions come back two ways: someone re-argues them, or another team quietly decides the opposite. How to stop both, with or without a tool.

I spent seven and a half years running production operations on global releases at Prime Video and Amazon Studios, often with 20–25 people across production, post, engineering, marketing and outside partners. The decisions that hurt were rarely the bad ones. They were the good ones a different team never heard about.

That is the part most advice on re-argued decisions misses. A settled decision can come back in two ways, and the usual fixes only catch one of them.

Two ways a settled decision comes back

The loud way. Someone reopens it. A meeting that was meant to be about the next step spends twenty minutes on whether the last step was right. It is frustrating and it costs time, but at least everyone can see it happening.

The quiet way. Nobody reopens it. Instead a different team, or the same team a few weeks later, makes a new decision that cannot stand alongside the old one. No one says “we’re reversing that”, because no one realises they are. Both plans go ahead until they collide.

Here is the quiet kind in miniature. On our example board, a team chatting about its CRM contract settles on moving to a cheaper vendor in Q4. Two days later the vendor review meeting settles on renewing for a year and revisiting in January. Each decision is sensible on its own. Nobody in the meeting mentioned the earlier one. If both stand, someone books a migration that the renewal has just ruled out.

Why decisions get re-argued

The loud kind usually has one of four causes.

  • Only the outcome was written down. The tracker says what was decided; the reasoning stayed in the room. When circumstances or people change, nobody can tell whether the original reason still holds, so it gets argued again from scratch.
  • Nobody owns it. If no one person is answerable for a decision, anyone can reopen it and no one can close it.
  • Nobody said what would change it. Without a stated condition for revisiting it, any new doubt is reason enough.
  • The people affected were not there. A decision made without the people who carry it out is a decision they will question the first time it costs them something.

Four habits that stop the loud kind

These are the common advice, and they work.

  • Write the reason next to the decision, in the words used at the time. Quote the sentence from the meeting notes rather than summarising it later. Summaries drift towards what people remember wanting, and an exact quote is something anyone can check.
  • Name one owner. A request to reopen goes to them, and they can say what it would take.
  • Add a “reopen if”. “We revisit this if the export test fails” turns a vague doubt into a question with an answer.
  • Give it a review date. People let go of a decision more easily when they know it has a date to be looked at again.

Why the quiet kind slips through anyway

All four habits make a decision easier to defend when someone challenges it. None of them helps when nobody challenges it, because the new decision is made somewhere the old one is not.

  • A decision log is only checked when someone remembers to check it. When a team makes a new decision, it rarely opens the old list first.
  • Teams keep separate records. Marketing’s decisions live in marketing’s notes, engineering’s in engineering’s. A decision log nobody else reads cannot warn anyone else.
  • Conflicting decisions rarely use the same words. “Move vendors in Q4” and “renew for a year” share no keyword. Searching the log for the new decision’s words will not find the old one.

How to catch contradictions by hand

Without any tool, these four steps catch most of the quiet kind.

  • Before you record a decision, ask what you already decided about the same thing. Search by the subject (the vendor, the launch, the budget line), not by the new decision’s wording.
  • Keep one decision log across teams, or failing that, a short weekly list where each team writes down what it decided, and one person reads it looking for decisions that cannot both be true.
  • When you find a conflict, record which decision stands and why. Mark the other as replaced, rather than deleting it, so the history still explains itself.
  • Link every decision to where it was made. The meeting notes, the chat, the document. A conflict is only settled well when both sides can see what was actually said.

Where doing it by hand breaks down

This works for one team, or a few teams with someone disciplined reading across them. It gets harder as the number of teams grows and as more decisions are made in conversation rather than written up. The weekly read becomes a job nobody wants, and it is the first thing dropped in a busy week, which is exactly when conflicting decisions are most likely.

What EchoLogic does about it

EchoLogic is built for the quiet kind. You write notes, talk things through with Echo, its assistant, or upload meeting transcripts and documents. Echo picks out the decisions, open questions, risks and assumptions, and keeps each one with the exact words it came from.

Each new decision is checked against what has already been decided on every board you can see, so a conflict is caught across teams as well as within one, and a pair that cannot both hold is flagged with the reason. You record which decision stands, ask whoever made the other one, or mark it as not a conflict. There are no integrations yet: you bring things in by writing, uploading or pasting.


See it on a real board

The example board is the one every EchoLogic trial starts on: the decisions, risks and questions Echo picked out of a chat, a meeting and a note, each with the words it came from, and two decisions that disagree. No signup.

EchoLogic
Continuity for how your team thinks, decides, and remembers why.
© 2026 EchoLogic. All rights reserved.