Reputation Math

Every business has two reputations: the one it believes it has, and the one a stranger can look up in eleven seconds. The second one has mechanics — measurable, improvable, and far less mysterious than the word “brand” makes them sound. Reputation math is the practice of treating being known and trusted as a system with inputs, instead of weather that happens to you.

Start with the stars, because the stars are arithmetic. A local search for any service returns a short list, and the searcher's eye does the same two things every time: filters by rating, then sorts by believability. The gap between 4.2 and 4.6 stars — less than half a star — is routinely the gap between getting the call and getting scrolled past. And that gap is made of individually tiny events: whether the ask happened in the happy moment or three days later by email, whether the QR code was in the truck, whether anyone replied to the review once it landed. Replies deserve special respect: prospects read a company's answers to its worst reviews as a preview of how their own bad day would be handled. A calm, specific reply to an unfair one-star review is marketing no budget can buy.

Then there's the channel nobody's dashboard can see: word of mouth. It gets treated as luck, but the research says otherwise — in a well-known study of bank customers, referred customers proved about 16% more valuable over their lifetime and measurably less likely to leave. A referral arrives pre-sold, expectations set by someone they trust, which makes referred customers not just cheaper to acquire but better to keep. And referral volume responds to design: a story worth repeating (the problem fixed without being asked, the invoice under the estimate), an ask specific enough to picture (“we're taking on two new projects this fall”), and a thank-you loud enough to prove the introduction mattered. Businesses that do those three things get recommended constantly and call it luck out of modesty.

Holding the whole system together is one division problem: marketing spend ÷ new customers it produced. Cost per customer is the number that grades everything else — the ads, the sponsorships, the website, the referral habit. Impressions and “engagement” exist mostly to keep reports cheerful when that number is bad. The hard part is almost always the denominator: it requires asking every new customer how they found you and writing it down, a distinct phone number or landing page per channel, a month of unglamorous honesty. But once the number exists per channel, marketing arguments dissolve into arithmetic — the channel worth doubling and the one worth retiring both become obvious on sight.

None of this requires a marketing department. It requires noticing that reputation compounds like interest: every review, every reply, every referred customer makes the next one slightly more likely. The pieces below each work one gear of the machine — the strange economics of asking for a review in person, where word of mouth actually comes from, and the one division problem that keeps the whole system honest.