2026-11-06
November 6, 2026
Nodding heads are not commitment. A scale you answer at the same moment as everyone else is the cheapest way to find out who is actually on board.
Every program lead has run this meeting. The proposal goes up, someone asks "any concerns?", four seconds of silence follow, and the decision is recorded as agreed. Three weeks later, one of the people in the room is quietly routing around it, and another is telling their team the decision was made without them. Nobody lied. The meeting simply had no instrument for measuring what silence meant.
Silence is ambiguous on purpose. It can mean "I agree," "I don't understand this well enough to object," "this doesn't affect me," "I disagree but the VP just spoke," or "I'll deal with it offline." A yes/no question collapses all of those into one signal, and the signal is always yes. If your decision meetings end in unanimous agreement every time, you are not running a high-trust team. You are running a meeting that cannot hear dissent.
The fix is not more discussion. It is a better question. Instead of asking whether people agree, ask each of them to place themselves on a scale that has honest options between "love it" and "block it," and ask everyone at once, so that the first answer in the room does not set the answer for everyone else. That is the whole idea, and the rest of this issue is how to make it work for senior programs where the stakes and the hierarchy both distort the signal.
The scale I am recommending comes from Sam Kaner's The Facilitator's Guide to Participatory Decision-Making, as summarized by sources such as Lucid Meetings. It is called the Gradients of Agreement, and its core move is to stop treating agreement as binary. In the version Lucid Meetings describes, there are eight positions, running from whole-hearted endorsement ("I really like it") through "I can live with it," abstention, "I don't understand the issues well enough yet," "I don't like it but will support it," serious disagreement, and finally veto. What matters is not the exact labels. What matters is that the scale gives people a legitimate way to say "I'll go along, but I have reservations," and a legitimate way to say "I'm not ready," both of which a yes/no question forces them to hide.
Consider what changes for a decision owner. Under a binary question, you learn almost nothing. Under a scale, you learn the shape of the room. A decision where eleven people say they can live with it and one says they have a serious disagreement is a completely different decision than one where six people are enthusiastic and six are lukewarm. The first has a single, specific objection you can go and understand. The second has a motivation problem that will show up in execution. Both look identical as "no objections."
The second mechanism is simultaneity. The reason the scale works is not only its granularity but the fact that everyone commits to a position before hearing anyone else's. If the senior person speaks first, the rest of the room recalibrates around them, a pattern anyone who has watched a staff meeting knows well. Reveal positions together, whether with fingers held up, a poll, or a shared doc where answers stay hidden until a set time, and you get independent readings instead of a cascade.
The third mechanism is where the live time goes. Once positions are visible, you do not need to discuss the people at the top of the scale. Spend the meeting on the people at the bottom, and on the people who said they need more information. Their objections are the only information the meeting can still change. This is the point at which gradients of agreement connects directly to async practice. The poll itself is a perfect async step: circulate the written proposal, collect positions by a deadline, and look at the distribution before the call. If the distribution is clean, cancel the meeting. If it is not, the meeting has a precise agenda: the specific concerns of the specific people who marked themselves low.
This also changes what a decision owner is allowed to do with the result. The scale does not make the decision a vote. If you use a DACI or RAPID structure, as we have covered in earlier issues, the approver still decides. What the scale gives the approver is calibrated information about the cost of deciding now. A senior TPM can say, honestly, "I have the call, three people are at 'don't like it but will support it,' and here is what I am doing to make sure that support survives contact with the first problem." That sentence is worth more than a hundred "alignment" slides.
There are traps. The first is false precision: do not average the numbers. A mean of 3.4 hides exactly the split you are trying to see, so read the distribution, not the average. The second is poll theater. If you collect positions and then decide the opposite of what the room told you without explanation, people will stop answering honestly, and the instrument is dead. The third is rank. If the poll is not anonymous and the room contains someone who punishes disagreement, you will get the same polite silence in a new format. Where that is a risk, collect anonymously or let people answer in writing to you directly.
For tech leaders, there is a quiet organizational benefit as well. Teams that practice this develop a shared vocabulary for partial support, and that vocabulary lowers the social cost of dissent. "I'm at a five" is a much easier sentence to say to your director than "I think this is a mistake." Over a few quarters, the gap between what people say in the room and what they say in the hallway narrows, which is the actual goal.
Try this week. In the next decision meeting you run, write the proposal down, then ask everyone to post a number from your scale in the chat or a poll before anyone speaks, and spend the first discussion minutes only on the lowest positions. Note afterwards whether the distribution matched what you would have guessed from the room's silence.
What it is. A facilitation scale that replaces yes/no agreement with a spectrum of positions, so a group can see how strongly each person supports a proposal and what it will take to move forward.
When to use it. When a decision needs real commitment from people who must carry it out, when a meeting ends in suspiciously fast agreement, or when you suspect that people are privately holding objections they will not voice in the room.
How to run it:
When NOT to use it. Do not use it for trivial or reversible decisions where the polling overhead exceeds the cost of being wrong.
Example: A migration cutover date gets eleven "can live with it" and one "serious disagreement." The objector turns out to own the only on-call rotation that the date collides with, which no one else knew.
Standup Meetings Are Dead (Honeycomb) — Ben Darfler argues that standups, including the Slack-based async versions, often fail to engage teams, and describes Honeycomb's replacement, a daily team sync with social time and agenda items; worth reading as a case study in redesigning a ritual instead of just moving it async.
Fast, Informed, Impactful: How to Master Decision-Making (LeadDev) — Katrin Freihofner's LDX3 London 2025 talk, based on interviews with over 50 tech professionals, argues that RFC processes meant to speed decisions often become bottlenecks, and points to DACI plus clear timelines as the fix.
What are agentic meetings in 2026 (Tana) — A vendor post, so read it as positioning, but its distinction between notetakers that summarize and agents that act during the call is a useful lens, and it concedes that most meeting AI in 2026 is still notetaking.
"I can live with it."
— One of the positions on Sam Kaner's Gradients of Agreement scale, as summarized by Lucid Meetings
Don't miss what's next. Subscribe to Critical Path: