Sit in on four conversations at the same company in the same week.
Marketing describes a platform that gives teams visibility. Sales describes a tool that replaces a manual process. The founder, on a podcast, describes a fundamental shift in how an entire function will work. Support, answering a ticket, describes something narrower and far more useful than any of the other three.
Everyone is telling the truth. None of them is telling the same truth.
Buyers rarely say "your messaging is inconsistent." What they say is that they want to see it again, or that they need to check with someone, or nothing at all. The cost shows up as caution, and caution shows up on the pipeline report as time.
Drift is a versioning problem
The temptation is to treat this as a discipline failure. Someone went off-script, sales needs enablement, marketing needs to get closer to the product.
Look at when each artefact was written and a different explanation appears.
The homepage was written eighteen months ago, for a product that did less and a buyer who was easier to find. The sales deck was rebuilt after a big deal, so it is shaped around that one buyer's objections. The talk track lives in the head of whoever closed most recently. The founder's version updates continuously, because they are actually thinking about it, and it never propagates anywhere.
Nobody drifted. Each asset is a snapshot, frozen at the moment it was made, and nothing updates one when another changes. A company that ships product weekly is running its story on annual releases.
What buyers actually experience
Every company is a circle, and no customer ever sees the circle. They see one arc at a time.
Someone finds a comparison page. Someone else gets a cold email. A third person sits through the demo. A fourth reads the pricing page at eleven at night, alone, deciding whether to bring this to their boss. None of them see the whole company, and each treats the arc they touched as the whole company, because it is the only evidence they have.
That is why inconsistency is expensive in a way that is hard to measure. The buying committee assembles one impression out of four disconnected fragments, and if those fragments do not resolve into the same shape, the conclusion is not that the company communicates poorly. The conclusion is that this company has not worked out what it is yet. That is a reasonable inference, and a dangerous one when a safer-looking vendor is also in the running.
Alignment meetings do not fix this
The standard response is a workshop. Everyone gets in a room, agrees the messaging, leaves feeling good, and updates their own material over the following month.
Six weeks later the drift is back, because the workshop produced agreement rather than infrastructure. Agreement decays. The product shipped something new, a competitor repositioned, a big prospect used a phrase that worked, and each person integrated that update into their own copy at their own pace. Nothing about versioning changed. You re-synced once.
What holds is a source, not a summit. One maintained reference that holds the narrative, the buyer, the category frame, and the win condition, and from which the assets are written rather than alongside which they sit. When the source changes, the derived assets get flagged as stale. That is a system. A shared doc nobody owns is a file.
The dinner party test
Ask five people on your team, separately, what the company does. Give them one sentence and no warning.
Do this at the end of a Tuesday, not in a meeting where they can see it coming, because the version that matters is the one that comes out unprepared. That is the version a buyer gets at a conference, on a first call, in a Slack message forwarded to a decision maker.
Then compare the five sentences for two things: whether they name the same buyer, and whether they name the same alternative. Tone can vary and should. If the buyer and the alternative change from person to person, the problem is not messaging discipline in the marketing department, it is positioning that marketing has been asked to paper over.
Building the thing that holds
Write the source of truth as something operational rather than aspirational. Short enough to read in ten minutes. Specific enough that two people writing from it independently produce copy that agrees. Owned by a named person, with a scheduled review, because a market that moves faster than your document will quietly make it fiction.
Then write everything from it, and mean everything. Homepage, core pages, deck frames, battlecards, onboarding email, the answer to "so what do you do." Each of those is an arc, and the buyer is only ever standing on one of them.
The goal is not that everyone says identical words, which sounds rehearsed and buyers notice that too. The goal is that five different phrasings all trace back to the same centre, so a buyer who hears three of them gets three angles on one company rather than three companies.