Sit beside the phone at a local business for an hour and you will hear the same four questions, in the same order, with the same hesitations. Do you take my insurance. How long does it take. Do you come out this far. How much is it, roughly. Whoever answers has said all of it a thousand times. And almost none of it is written down anywhere a customer could find it.
That gap is the cheapest opportunity in content marketing for a local business, and hardly anyone works it. Most owners I talk to think content marketing means having something to post. It doesn't. It means writing down the answers you already give out loud, and putting them where the person asking can find them at eleven at night without calling you.
Start with a transcript, not a keyword tool
The exercise takes two weeks and no budget. Put a notepad at every point where a stranger reaches you — the phone, the front desk, the shared inbox, the social messages, the quote follow-up call. Every time someone asks something, write the question down in their words, not yours. If it has been asked before, add a tally mark.
Two rules make this work. First, the customer's phrasing stays. "Is this covered?" is not the same question as "What is your coverage eligibility policy?", and only one of them is what a real person types. Second, capture the questions that come after the sale and the ones that come from people who never bought. Post-purchase questions are where your reputation is actually made. Pre-purchase objections tell you what a competitor has already told them.
You will end up with a list much longer than you expected and a short cluster at the top that accounts for most of the volume. In my experience the meaningful set is around twenty questions. Not two hundred keywords. Twenty questions that a human being asked a human being, ranked by how often.
Sort by what the question is actually asking
Before writing anything, put each question in one of four buckets, because the bucket determines the shape of a good answer.
- Cost — wants a range and the two or three things that move it up or down.
- Fit or eligibility — wants a yes or a no in the first line, then the exception.
- Risk — wants to know what happens when it goes wrong, and who pays.
- Logistics — wants one specific fact: hours, travel radius, lead time, what to bring.
Most bad business writing happens because the writer answered a cost question with a fit answer, or a risk question with reassurance instead of a process. A person asking "what if it fails after you leave?" is not comforted by the sentence "we stand behind our work." They want to know who they call, how fast someone comes, and whether there is a charge.
One question, one page, answer in the first two sentences
The page format barely varies. The headline is the question in the customer's words. The first two sentences answer it outright. Then a short paragraph on why the answer is what it is. Then the exception — the case where the answer changes. Then the single next step for someone who is now ready to act.
No preamble. No "when it comes to choosing a provider, there are many factors to consider." If a reader takes the first two sentences and leaves, the page did its job. That is not a lost conversion. That is a person who now trusts you slightly more than they did a minute ago, at no cost to you, and who will come back when the need is urgent.
Length follows the answer. Some of these pages are a hundred and fifty words and should stay that way. Padding a short answer to look substantial is the single most common way this exercise gets ruined, and it is obvious to every reader.
This format also happens to be exactly what answer engines and AI assistants can lift and quote, because a clearly-stated question with a direct answer underneath it needs no interpretation. I wrote about the mechanics of that separately in a working checklist for being found by AI. The structure that serves a nervous customer at eleven at night turns out to serve the machines too. That is not a coincidence; both are trying to extract an answer quickly.
Why this beats most local advertising on cost per customer
Local advertising rents attention. This builds something you own. The page you write this month answers the same question next year without another invoice, and the ad stops the day the card declines. Over any horizon longer than a quarter, that difference compounds in a way that is hard to argue with.
But the stronger argument is about who calls you afterwards. Someone who read your price range, read the eligibility answer, read what happens if the job runs long, and then picked up the phone is a different prospect from someone who clicked a banner. They have pre-qualified themselves. Your staff spend less time on the call, the quote converts more often, and the people who would have wasted an hour never dial. Advertising buys volume at the top. Answering questions improves the mix.
There is also a competitive asymmetry worth noticing. Your ad competes against every other advertiser in the market bidding for the same attention. Your question page competes against silence, because your competitors have not answered either. I run BrandAmplifi, which works in online reputation and search visibility, and the thing I see most often is not businesses losing a content arms race — it is entire local categories where nobody has published a plain answer to the most-asked question in the trade.
I will give the honest caveats. This is slow, it costs staff time rather than cash, and it does almost nothing for genuine impulse purchases. If you need bookings this Saturday, write a page and buy an ad. When I think about where a Florida business should put its next dollar, the question-page work is rarely the most urgent line item and is almost always the one with the best return three years out.
The three ways this gets ruined
First, writing for search instead of for the asker. The moment the page is longer than the answer requires, or the headline becomes an "ultimate guide," you have started serving a machine instead of a person, and it reads that way.
Second, refusing to answer the money question. "Every project is unique — call for a quote" is not caution, it is a toll booth. Everyone who leaves your site does so because you asked them to phone a stranger to learn something a competitor published in a sentence. Publish a range. Publish what pushes a job to the top of the range. You will lose a few price shoppers you were never going to keep.
Third, handing the writing to someone who has never done the work. All the value in these pages lives in the exceptions, and only the person who answers the phone knows them. The workable division of labour is that the operator talks the answer out loud — a voice memo on the drive home is enough — and someone else turns it into clean prose. Drafting tools are genuinely good at that conversion now. They are terrible at inventing the exception, and they will make one up if you let them.
Keep it honest, keep it current
A wrong answer is worse than no answer. Once a year is not enough; read the whole set every quarter and fix the pages where the policy, the price band or the turnaround has moved. When a page is right, you will know, because the question stops arriving by phone. That is the measurement you actually have, and it is better than most dashboards.
These pages also pull their weight alongside your reviews. A customer who read a straight answer and then got exactly what was described is a customer inclined to say so publicly, which is most of what I meant in earning the first hundred reviews honestly. Published answers set an expectation. Reviews confirm you met it. Neither works nearly as well without the other.
The answer you'd give at the counter
The test for every page is whether it says the same thing you would say to someone standing in front of you with their keys in their hand. Not the polished version. The real one, including the part where you tell them the cheaper option is fine for what they need. Businesses spend years trying to manufacture trust through advertising while giving it away free on the phone forty times a day and never writing it down. We should all be doing the boring version first: answer the question, publish the answer, keep it true.