When Someone Asks AI 'How Much Does This Cost?' — Winning the Pricing Query

Price is one of the first things a buyer asks about — and increasingly they ask AI before they ask you. If your pricing isn't legible to the model, it won't stay quiet out of politeness. It will answer anyway: with a guess, a category average, or a number it borrowed from a competitor who was more transparent than you. Winning the pricing query means being the clear, honest source on what you cost, so the model quotes you instead of filling the gap with someone else.

I talk to a lot of business owners who have built their entire approach around not saying the price. The logic is familiar: get them on a call, build value first, then reveal the number once they're invested. That instinct made sense in a world where the buyer had to come to you to learn anything. It makes a lot less sense when the buyer's first move is to open ChatGPT or Perplexity and ask what things like yours cost — and the model gives them an answer whether you participated or not.

Silence doesn't create a conversation — it creates a vacuum

Here's the shift that trips people up. In the old model, withholding price created a conversation: the buyer had to reach out to find out. In AI search, withholding price creates a vacuum, and the model fills it. Ask any major engine what a service costs and it will do its best to answer, pulling from whatever it can find — industry ranges, forums, review sites, and crucially, competitors who published their numbers. Your silence doesn't hold the buyer in suspense. It just hands the pricing narrative to whoever was willing to talk.

And the answer the model gives in your absence is often worse for you than any number you'd have chosen. It might quote a figure that's years out of date. It might anchor the buyer to a cheap competitor and make your fair price look inflated by comparison. It might hedge so vaguely — 'prices vary widely' — that the buyer leaves with no reason to pick you over anyone. None of those outcomes is one you'd choose on purpose, yet they're exactly what you get by default when you decline to be the source.

The buyer, meanwhile, genuinely wants this information. 'Can I afford this?' is not a rude question or a sign of a bad lead — it's the gate almost everyone walks through before they'll engage. When you answer it honestly and early, you're not giving something away. You're removing the single biggest reason a good-fit buyer hesitates.

You can win the pricing query without publishing a rigid menu

The objection I hear next is that pricing is complicated — it's custom, it varies, there's no single number to publish. Fair. But 'I can't publish one exact price' is not the same as 'I can't be legible on price,' and conflating the two is how businesses talk themselves into saying nothing at all.

The model doesn't need a rigid price list. It needs something extractable and honest. A few forms that work:

• A range. 'Most engagements run between $X and $Y' is enormously more useful to both the model and the buyer than silence. It sets expectations and gives the model a confident, quotable fact.
• A starting point. 'Projects start at $X' anchors the low end honestly without committing you to a number for every scenario.
• The drivers. Explain what moves the price up or down — scope, size, complexity, location, add-ons. This lets the model (and the buyer) reason about where they'd likely land, and it's often the most genuinely helpful thing you can offer.
• What's included. Price without context invites the wrong comparison. Tying the number to what it covers reframes it from a raw figure into a value statement the model can carry.

Any of these makes you citable on the pricing query while preserving all the flexibility you actually need. You're not locking yourself into a number. You're giving the model enough to quote you accurately instead of guessing.

The mistake I see most often isn't choosing the wrong form — it's choosing none of them because no single number felt safe to commit to. Perfect precision was never the bar. The bar is being more useful and more legible on price than the vacuum the model falls back on, and a thoughtful range clears that bar easily. Pick the form that fits how you actually sell, state it honestly, and you've already done more than most of your market is willing to do.

Structure the number so the model can lift it cleanly

Being willing to share price is necessary but not sufficient — the information also has to be structured so a model can extract it without ambiguity. I've seen businesses that technically disclose pricing but bury it in a PDF, trap it inside an image, or scatter it across a page in a way no extraction process can reassemble into a clean answer.

A few principles that help the number actually land:

• Put it in real text, near the thing it prices. The price should sit in plain, selectable text right beside the service or product it belongs to — not in a graphic, not three sections away. Models quote passages; make the passage self-contained.
• State it plainly before you dress it up. Lead with the number or range, then explain the value. If the reader (or the model) has to wade through three paragraphs of positioning to reach the figure, the figure is easy to miss and easy to misattribute.
• Frame it with what it includes and who it's for. 'Our $X package includes A, B, and C, and is built for businesses at [stage]' does double duty: it gives the model a clean fact and it pre-qualifies the buyer against your range.
• Keep it current. A stale price you forgot to update is almost as bad as a competitor's number — it trains the model, and your buyers, on the wrong figure.

That last point about framing is where the pricing query quietly becomes a qualification tool. When the model can tell a buyer not just what you charge but what that buys and who it's for, the people who reach out afterward have already self-selected. They've seen the range, decided it's in bounds, and contacted you anyway. That's a fundamentally better conversation than one that opens with sticker shock — and you got there by being transparent, not by hiding the ball.

Price is also a trust signal the model reads

There's a dimension to this that goes beyond the raw number, and it's one I think a lot of owners underrate. Willingness to state your price reads as confidence — to buyers, and I'd argue to the model forming a picture of you as an entity. A business that clearly says what it costs and why comes across as established, legitimate, and sure of its value. A business that treats price as a secret the buyer has to pry loose can read as evasive, or simply as thin: there's less concrete, factual information about it for anyone to work with.

Remember that AI is assembling an understanding of your business from the signals available, and specific, concrete facts are the raw material. A clear price point is one of the most concrete facts you can offer. It's unambiguous, it's quotable, and it slots neatly into the kind of factual profile the model builds when it decides how confidently to talk about you. Vague positioning language gives the model very little to hold onto; a stated price, a defined package, a clear range — those are the hard edges that make an entity feel real and knowable rather than fuzzy.

I'm not claiming a price tag by itself vaults you up the rankings — that would overstate it. But in a landscape where the model rewards the legible and hedges on the murky, refusing to say what you cost pushes you toward the murky end for no good reason. The businesses that feel most citable tend to be the ones that are easiest to state plain facts about, and what you charge is one of the plainest, most buyer-relevant facts there is.

Check what AI says you cost right now

Everything above is testable, and I'd rather you test it than take my word. Go ask the major engines what your business or service costs, phrased the way a real buyer would. Read the answer honestly. Is the number right? Is it current? Is it even yours, or did the model quote someone else because they were the clearer source? Is it so vague the buyer gets nothing?

Whatever comes back is a live report on how well your pricing is actually landing in the places buyers now look first. If it's wrong or borrowed or absent, that's not a reason to hide harder — it's the clearest possible signal that the gap is being filled by someone else, and that the fix is to become the honest, legible, well-structured source on your own price. My read, not a guarantee: the businesses willing to answer the cost question clearly tend to earn the citation on it — and the buyer who arrives already comfortable with the number.

About the author

Scott Tischler is the Founder & Chairman of AIrecommend.ai and a practitioner-authority on AI search and Answer Engine Optimization. With 20+ years in marketing technology — including American Express, MetLife, and UBS — and executive and professional study at Wharton, Harvard, and Oxford, he helps businesses become the ones AI recommends.