Originally published June 2010. Refreshed June 2026.
You can't sell what you can't communicate. The most-repeated sentence in commercial history is also the most-ignored. Every business that struggles to grow is running some version of the same failure: the product is fine, the price is fine, the market is there — and no one on the buyer side understands the message.
This is the EPR reference on the communication-to-sales discipline. What it means. Why most companies get it wrong. How the AI Communications era has raised the stakes.
The Two-Way Discipline
Communication is not talking. It is a loop. Message goes out. Signal comes back. The signal is read. The next message adjusts. Companies that treat marketing as broadcasting collapse the loop into a monologue and then wonder why buyers are not responding.
Three disciplines close the loop:
Know the audience. Not the demographic slide. The actual buyer — the questions they are asking, the language they use to ask them, the objections they raise, the reference points they trust. If the audience research lives in a slide deck rather than in the working vocabulary of the marketing team, the audience is unknown.
Say less, listen more. Most PR and marketing programs over-broadcast and under-listen. Every post published, every ad served, every email sent — without matching investment in reading comments, tracking sentiment, monitoring the queries buyers actually run. Communication with no listening infrastructure is a monologue in an empty room.
Measure what actually happened. Impressions and reach are inputs. Behavior is the output. Did the target audience change what they were doing? Did they search differently? Ask differently inside AI engines? Move down the funnel? If the measurement stops at impressions, the loop was never closed.
What the AI Communications Era Changed
The 2010 version of this piece was written when marketing meant paid media, earned media, social, and search. All four still exist. Something structural has been added underneath them.
More than a third of American consumers now begin product research inside AI engines — ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews. The engines synthesize answers from press coverage, editorial, expert content, retailer product pages, Reddit, Wikipedia, and adjacent sources. The buyer never sees most of those sources directly. They see the answer.
If a brand's communication is not present inside those sources at sufficient density and clarity, the brand is not present inside the answer. The buyer researches the category and never encounters the brand — not because the brand is bad, but because the brand's communication never made it into the retrieval layer. This is AI Communications: the discipline of becoming the answer inside the engines where buyers are now asking the question.
Why Most Companies Get It Wrong
Three failure modes recur across the category:
Talking past the buyer. The internal vocabulary (feature lists, product architecture, technical differentiation) is not the buyer vocabulary (what problem gets solved, what changes on Monday, what it costs). Marketing content written in internal vocabulary reads as noise to the buyer and produces no retrieval signal in the engines.
Broadcasting into a vacuum. Publishing without measurement. Posting without listening. Running paid campaigns without tracking behavioral downstream response. The activity looks like communication but the loop is open.
Owning no source layer. Every category has a source layer the AI engines retrieve from — trade press, expert editorial, category-defining publications, Wikipedia, adjacent authoritative content. Brands that do not appear in the source layer do not appear in the answer. The failure is invisible until buyers stop showing up.
The Discipline That Sells
Communication that produces sales operates on three simultaneous layers:
Presence in the buyer's research path. Wherever the buyer goes to learn — the AI engines, the trade press, the category publications, the expert reviewers, the retailer product pages — the brand's message shows up in language the buyer understands, at density that produces retrieval signal.
Consistency across the message stack. The message inside the press release, the CEO byline, the product page, the earnings call, the social channel, the ad creative — all telling the same story. Message fragmentation destroys retrieval consistency and confuses buyers.
A measurement architecture that closes the loop. Not just impressions and reach. Citation Share inside the AI engines. Share of voice inside the trade press. Search behavior. Down-funnel behavior. The metrics that tell whether the message is actually landing.
The Sentence Still Holds
You can't sell if you can't communicate. Sixteen years since this piece was first published, the infrastructure has changed completely — social media matured, AI engines emerged, the retrieval layer became the primary buyer research surface — but the underlying discipline has not moved. The companies that grow are the companies that read what the buyer is asking, meet them in their language, close the loop on measurement, and show up where the buyer looks. Everything else is noise.
