A single product decision has to be narrated to wildly different people: the engineer who will build it, the executive who will fund it, the designer who will shape it, the salesperson who will pitch it, the customer who will live in it. The common mistake — and strong communicators make it too — is to build one narrative and deliver it identically to all of them, on the theory that the truth is the truth and consistency is a virtue. But telling the same story the same way to different audiences is not consistency. It is a failure to perform the last and most demanding act of the craft: customizing the telling to the specific mind receiving it. This piece is about adapting a narrative across audiences without distorting it — holding the truth fixed while varying nearly everything about how you convey it.
The distinction that keeps customization honest: message versus telling
Before any technique, a principle, because customization done wrong slides quickly into telling different people incompatible things. The safeguard is a hard line between your message and your telling. The message is the underlying truth — what is actually happening, what you actually believe, what you are actually asking for. It must remain invariant across every audience. The moment it changes from room to room you are no longer customizing, you are lying, and in an organization the contradictory versions eventually meet — someone forwards the finance deck to engineering — and the room concludes, correctly, that you cannot be trusted. The telling is everything else: the framing, the emphasis, the altitude, the vocabulary, the evidence, the emotional register, the entry point. Customizing it well is not manipulation but respect.
Model the peers and the powers first
You cannot customize for an audience you have not mapped, and the map that matters most for a PM is not the org chart. It is the peers-and-power map: who evaluates this on what criteria, who can say yes, who can only say no, who is not in the room but can quietly reverse it afterward. Every function runs its own frame, and a frame is not a preference you can charm someone out of — it is the machinery that constitutes their competence. Finance thinks in unit economics, payback periods, and risk. Engineering thinks in complexity, maintainability, and technical debt. Sales thinks in deals, objections, and quota. Design thinks in the user’s experience and the coherence of the whole.
The useful mental model for your own posture across all these frames is not mini-CEO. It is editor. A great editor does not write the book and does not have authority over the author, but they understand what the book is trying to be, they help each contributor get there in that contributor’s own voice, and they hold the whole thing coherent.
The dimensions you vary
The entry point
The most important customization is where you begin. Every audience arrives with an existing preoccupation — a question they are already asking, a fear they already hold, a goal they are already chasing — and the effective narrative enters through that door rather than knocking on a door of your own choosing. For an executive worried about competitive position, enter through the competitive threat. For an engineering team proud of its craft, enter through the technical problem worth solving. Same product, same underlying message, five different first sentences.
The altitude
Different audiences need the story at different heights. An executive audience usually needs it high — the strategic shape, the bet, the outcome — and is actively harmed by implementation detail. An engineering audience often needs it low — the constraints, the mechanism, the real trade — and finds a purely strategic telling both vague and faintly insulting. Matching altitude to audience is not dumbing down or nerding out. It is delivering the resolution of detail at which that particular audience can actually engage.
The frame and the vocabulary
Customizing the frame means projecting the same message onto the terms the audience uses to decide what is true. The identical reliability investment is “reducing churn risk in our enterprise accounts” to a revenue leader, “paying down the technical debt that’s slowing every release” to engineering, and “we will not lose your data” to a customer. None of those is false. Each is the same truth projected onto a different frame, and choosing the right projection is what makes the message legible to that mind instead of bouncing off it.
The evidence
Audiences are convinced by different kinds of evidence. Some are moved by quantitative rigor — the model, the number, the confidence interval. Others are moved by the vivid particular — the single customer story that makes an abstraction real. Part of customizing is selecting, from the full body of evidence you already hold, the specific proof that will land here, and leading with it.
The register
Calibrate the emotional register — how much urgency, confidence, vulnerability, or excitement the telling carries. The same news delivered to a nervous team and to a complacent executive can require opposite framings: the anxious team needs calm and command to act well, while the too-comfortable leader may need a sharpened sense of the stakes to act at all.
One message through five rooms
Take a single invariant message: we should stop building net-new features this quarter and spend it hardening reliability. That is the truth, the ask, the through-line, and it does not change no matter who I am talking to. Now watch it get five different tellings.
To the executive I enter through the competitive threat and lead with a number: “We’re losing two enterprise renewals a quarter to outages, and that’s the single fastest-growing reason in our churn data — one hardening quarter protects roughly the revenue of a whole feature year.” To the engineering team I enter through the craft: “Every release is getting slower because we keep building on a foundation we haven’t paid down — this quarter we fix the foundation.” To design I enter through the user’s experience: “The product feels unreliable, and no amount of polish survives a page that doesn’t load.” To sales I enter through the deal: “You keep losing the security-conscious accounts on the reliability question — after this quarter, that objection is one you can answer.” To the customer: “The thing that made you lose an afternoon last month — we’re spending this quarter making sure it can’t happen again.”
Five entry points, five altitudes, five frames, five kinds of evidence, one message. Nobody heard a different truth. Everybody heard it in their own language. That is the whole discipline in a single decision.
The trap of the universal deck
A closing warning about a specific, seductive, extremely common failure. Organizations love the universal deck — the single set of slides that goes to everyone. It is efficient, it feels consistent, and for anything that actually matters it is almost always a mistake, because a deck built to serve every audience serves none of them well. The efficiency is an illusion: you save the hours of building multiple tellings and repay them many times over in the meetings that go nowhere because the story never connected to anyone in particular. Reuse is fine for reference material and record-keeping. But for the communications where alignment actually has to happen, resist the false economy. Build the telling for the audience in front of you. The message stays the same; the telling is theirs.