Active Listening as a PM Superpower: Hearing What's Not Said

Active listening means treating hesitation, omission, and tone as data, not noise. The stakeholder who says "sure, that could work" while avoiding the roadmap slide is telling you more than the one who objects out loud. The skill isn't hearing words faster — it's noticing what got left out, then asking the question that surfaces it.

Quick answer: Active listening for PMs means decoding hedges, silence, and tone as signal, then testing what you heard with a reflective question before acting on it. The stakeholders who never object out loud are usually the ones with the most left unsaid.

Active listening is one specific muscle inside the broader discipline of PM communication and influence — the one most guides skip because it's harder to teach than a slide template. It's also the muscle that determines whether everything else you do — the deck, the doc, the roadmap review — lands on the actual problem or on a polite fiction everyone agreed to nod along with.

What "Hearing What's Not Said" Actually Means

Hearing what's not said means noticing the gap between a stakeholder's stated position and their actual interest — the reason behind the position they're willing to say out loud. Carl Rogers and Richard Farson, who coined the term active listening in 1957, described it as listening for the feeling behind the words, not just the words.

"Focus on interests, not positions." — Roger Fisher and William Ury, Getting to Yes

Most PM advice about listening stops at "don't interrupt" and "take notes." That's table stakes, not the skill. The real leverage is in the gap between what a stakeholder is willing to state as their position and what's actually driving it — the interest underneath it.

Fisher and Ury, who ran the Harvard Negotiation Project, named this distinction in Getting to Yes: a position is the stance someone states ("we need this by Q3"); an interest is the underlying need or fear that produces it — their own roadmap depends on it, or they've been burned by a late launch before. Negotiators who argue with the position instead of the interest end up winning an argument nobody actually had.

PMs do the same thing every time they respond to "we need this by Q3" with a project-plan defense of the timeline, instead of asking what happens to the stakeholder if Q3 slips.

Positions vs. Interests: The Gap Where the Signal Hides

In PM conversations, this gap shows up constantly — a stakeholder states a position that's really a proxy for something they haven't said aloud, often because saying it directly feels risky, political, or premature. The table below maps five phrases nearly every PM has heard to the position they express and the interest that's more likely underneath, with a question that tests the read instead of assuming it:

None of these interpretations are guaranteed — that's the point of the question in the last column, which turns your guess into a checked hypothesis instead of a private assumption you act on unchallenged. Guessing the interest and never checking it is just projection wearing the costume of empathy. Ask, then listen to how they answer, not just what they answer.

Where the Unsaid Hides: Six Signals Most PMs Miss

The unsaid usually hides in six places: hedging language, timing of a response, who a concern gets routed to, disproportionate brevity, channel-switching, and the gap between tone and words. Each is a low-cost way for a stakeholder to register a concern without paying the social cost of stating it directly.

• Hedging language — "should be fine," "probably okay," "I think this works" instead of a clean yes. Hedges let someone agree now while preserving deniability later.
• Response timing — the stakeholder who usually replies within the hour takes two days on this one thread. Delay is often processing conflict, not just being busy.
• Concern routing — feedback that shows up in a private DM or hallway comment instead of the group thread or the doc. Where a concern goes tells you how safe the person feels raising it publicly.
• Disproportionate brevity — someone who normally writes three paragraphs sends "Looks fine." A one-word answer from a normally verbose person is rarely actual indifference.
• Channel-switching — moving from written comments to "can we hop on a call about this," often because the concern is sensitive enough that they don't want it in a searchable, attributable record.
• Tone-word mismatch — the words say yes; the tone (flat, fast, clipped) says something else. This is the one place 7-38-55 gets invoked loosely, and it's worth pausing on why.

Albert Mehrabian's 7-38-55 rule gets misquoted constantly as "93% of communication is nonverbal" — it isn't, and treating it that way is bad advice. Mehrabian's 1967 studies measured something narrower: when tone and words disagree about feelings and attitudes specifically, listeners trust tone over the words. It doesn't apply to communicating facts, plans, or instructions — only to the emotional read when words and delivery contradict each other.

That narrower version is still useful. If a stakeholder's words say "I'm on board" but the tone is flat and the response lands a beat too slow, trust the tone — not as gospel, but as a reason to ask a follow-up question.

Mehrabian's rule describes a conflict between tone and words about feelings — it was never a general ratio for how communication works.

The same six signals show up outside stakeholder meetings, in customer research:

• In a journey-map interview, "it's fine, I just switch tools when I need to" often masks a real frustration the customer has stopped expecting anyone to fix — precisely the kind of dip a well-built customer journey map is designed to catch when you plot the emotion curve stage by stage instead of trusting the summary sentence.
• In a JTBD interview, the anxiety and habit forces central to Jobs-to-be-Done are almost never stated as a clean sentence. Customers rarely say "I'm anxious this new tool will embarrass me in front of my boss" — they say "let me think about it," and the anxiety force is the interest hiding underneath that position.

A Practical Toolkit for Surfacing What Stakeholders Won't Say Outright

Four techniques reliably surface what a stakeholder is withholding: reflective paraphrasing, labeling the emotion instead of the content, asking calibrated what and how questions instead of yes/no ones, and deliberately holding silence after you ask. Each trades a small amount of awkwardness for a much more honest answer.

1. Reflective paraphrasing — repeat back what you heard in your own words before responding to it: "What I'm hearing is you're fine with the direction but not the timeline — is that right?" This is the mechanic Rogers and Farson described as the core of active listening, and it forces a stakeholder to confirm, correct, or add to your read instead of letting an ambiguous "sounds good" stand unchecked.
2. Labeling — name the emotion you're picking up on, tentatively, without insisting on it: "It sounds like you're worried this timeline puts your team at risk." Chris Voss, the FBI's former lead international kidnapping negotiator, built this into a formal technique in Never Split the Difference — labeling defuses a feeling by naming it out loud, because most people relax once they feel accurately seen, even if the label isn't perfectly right.
3. Calibrated questions — replace yes/no questions with what and how questions that can't be answered in one word: not "are you OK with this?" but "what would need to be true for this to work for your team?" A calibrated question hands the stakeholder the pen instead of asking them to sign what you've already written.
4. Deliberate silence — after you ask a real question, stop talking. Most people are uncomfortable with a silence longer than two or three seconds and will fill it, often with the more honest, less rehearsed version of their answer. Interrupting your own question with a follow-up before they've answered is the single most common way PMs talk themselves out of the answer they were trying to get.

The "That's Right" Test

The that's right test comes from Chris Voss: when a stakeholder responds to your summary with that's right instead of the more common you're right, you've actually landed on their real position — you're right is what people say to end a conversation, not to confirm one.

Voss uses this distinction to tell a genuine breakthrough from polite dismissal in a negotiation, and it maps directly onto PM stakeholder conversations. If you summarize what you heard and get a fast "you're right, sounds good," that's often a stakeholder ending an uncomfortable conversation, not agreeing with it. If instead they pause, add a correction, and land on "yeah, that's right, that's exactly it," you've actually converged, and you can act on it.

"It sounds like you're worried this timeline puts your team at risk. Is that fair, or am I off?"

That single sentence does the work of labeling and reflective paraphrasing at once, and it costs you nothing if you're wrong — a stakeholder correcting a slightly-off label is itself useful information.

Small prompt changes produce very different quality answers. The table below swaps five common PM questions for calibrated versions that make it harder to give a reflexive, uninformative response:

Every "better" question in the right column does the same thing: it removes the easy exit of a one-word answer and asks the stakeholder to do a small amount of work to respond. That friction is exactly what filters a reflexive answer from a considered one.

Why Unheard Silence Costs More at Scale

Silence in a group setting is not consensus — it's usually the least safe response available, and it scales badly because the same authority gradient that keeps one person quiet in a meeting keeps a dozen people quiet in an org. The cost shows up weeks later as scope creep, a late blocker, or a decision that quietly gets undone.

Amy Edmondson's research on psychological safety at Harvard Business School — built originally from studying hospital teams and later extended across industries — found that authority gradients, the perceived status gap between you and whoever's speaking, predict who stays quiet far better than how good their idea is. A junior engineer with a real concern about your spec often says nothing in a room with a VP in it, not because the concern is weak, but because the room makes voicing it expensive.

Authority gradients show up in predictable patterns once you know to look for them:

• Someone who agrees fast and stops asking questions the moment their manager joins the call.
• A concern that surfaces only after the most senior person has left the room.
• A team that nods along live, then quietly reopens the decision over Slack an hour later.

This tracks with Gallup's long-running employee engagement research, which has repeatedly found that only a minority of employees — historically well under half — strongly agree their opinions seem to count at work. If stakeholders don't believe raising a concern changes anything, the rational move is to stay quiet and let the decision fail on its own, which is a worse outcome for you than an uncomfortable meeting would have been.

Quiet is cheap to give and expensive to misread.

This is exactly why leading with the conclusion in an update — the discipline behind the executive communication pyramid — has to be paired with deliberately built-in space for pushback, not just a tighter opening line. A crisp top-down structure that never invites the quiet disagreement in the room just gets you to the wrong decision faster.

Reading what a stakeholder needs is itself a listening act, not just a delivery choice. Some stakeholders check out the moment they see a chart and need the narrative behind the numbers before the data will land; others check out during a story and are waiting for the table. Which one you're facing is something you learn by listening to how they engaged with your last three updates, not by defaulting to your own preferred format.

Make the Read Durable: Turning a Decoded Signal Into a Record That Holds Up

Decoding a stakeholder's real position is wasted work if it isn't written down the same day, dated, and specific enough that it means the same thing to you in six months as it does right now. "We aligned" is not a record. "Eng agreed to the Q3 date but flagged migration risk they didn't raise in the room" is.

Most of what active listening surfaces has a short shelf life. The tone you caught, the hesitation before the "sure," the topic that got routed to a side channel instead of the group thread — all of it is legible to you for maybe 24 hours before your own memory smooths it into a cleaner, wronger version of what happened.

"Sarah approved the pricing change." The version most meeting notes discard: "...while saying, 'I guess we don't have a choice at this point.'"

Write the second sentence, not just the first. The hedge is the part of the story that explains why the pricing change underperforms six weeks later, and it only survives if someone wrote it down at the time.

This is a writing problem as much as a listening one — the read only survives if the sentence you write down is precise enough to carry it. The same discipline behind clear PM writing — specific over vague, attributed over anonymous, dated over timeless — is what makes a decoded signal durable instead of decorative.

Three habits keep the record honest:

• Attribute it. Name who said it and who read the subtext — "Sarah said X; I read Y underneath it" — so a future reader can tell your inference from their statement.
• Date it. A read that isn't dated can't be checked against what actually happened later, which is the only way you get better at making them.
• Separate the fact from the inference. Write the quote, then your interpretation, as two different sentences — never blend them into one that reads like a fact.

The Prodinja Angle

Catching the unsaid is the harder half of this skill. Keeping it — accurately, dated, attributable — past the meeting it happened in is the half that actually determines whether it changes anything.

When that read eventually hardens into a written decision inside a spec, Spec Studio's PR-style diffs keep every version of that decision visible, so the moment your reading of a stakeholder's hedge turned into a documented tradeoff is traceable too — not just the final wording, but how it got there.

Key Takeaways

• Positions aren't interests. What a stakeholder states is rarely the full reason behind it — treat the stated position as a clue, not the answer.
• Silence is a data point, not consensus. In group settings, quiet is usually the safest response available, not agreement, especially across an authority gradient.
• Calibrated what/how questions outperform yes/no questions. They remove the easy, reflexive answer and ask for a specific one instead.
• That's right beats you're right. A fast you're right often ends an uncomfortable conversation; a considered that's right usually confirms real alignment.
• The 7-38-55 rule is narrower than the internet thinks. It applies to reading feelings when tone and words conflict, not to communication in general.
• A decoded signal is only as useful as the record you make of it. Write the quote and your read as two separate sentences, dated and attributed, before memory smooths them into one.