What Your Pricing Page Should Say When You Do Not Show Prices

What should a pricing page say if you cannot publish a price?

It has to answer the question the price was going to answer, which is whether this is for someone like me. Give the shape of the model, the range of who it fits, what drives the cost up or down, and what happens on the call. A page that says contact sales and nothing else answers nothing.

I see this page constantly, and the instinct behind it is usually reasonable. Custom scopes, negotiated contracts, or a fear of being compared on price alone. The instinct is fine. The execution is almost always a dead end.

The goal is not to force every visitor into a conversation. It is to let the wrong ones leave and the right ones arrive prepared.

Why do buyers react badly to a blank pricing page?

Because it reads as a refusal. The visitor came with a specific question and left with an invitation to a sales process. That asymmetry feels like a cost imposed on them for information you already have, and people resent paying with their time to learn something you could have told them.

There is a second effect that is worse. In the absence of a number, buyers assume the highest plausible one. Silence does not read as flexible, it reads as expensive, and you have lost the visitors who could have afforded you and never found out.

The people you do get are the ones willing to book a call without knowing the price, which is a smaller and stranger group than your actual market. You have not filtered for fit. You have filtered for patience.

What are you actually competing against?

Against pages that do publish. Your buyer spends all day on software that states its prices plainly. Zapier's pricing page lists a free plan at zero dollars a month including one hundred tasks per month, Professional starting from 19.99 dollars a month, and Team starting from 69 dollars a month.

The same is true in the AI tooling your buyer is now evaluating weekly. Anthropic publishes per token prices for each Claude model, Google publishes per million token prices for the Gemini API with future dated changes attached, and OpenAI publishes a full table of model prices. These are complicated products with public numbers.

That context matters because it sets an expectation. When the complex, usage based, enterprise capable tools in your buyer's stack publish prices, your custom pricing explanation has to work harder than it did five years ago.

What belongs on the page instead of a number?

Four things. The pricing model itself, the variables that move it up or down, the profile of the buyer each option suits, and the process that runs from first enquiry to a written quote. That combination lets a visitor place themselves on your scale without you ever publishing a rate card.

The model is the most important and the most often omitted. Say whether you charge per seat, per usage, per site, per outcome, or as a fixed engagement. That single fact changes how a buyer thinks about the cost of growth, and it is not commercially sensitive.

The variables come next. Tell people what makes a deployment expensive. Volume, integrations, data migration, compliance requirements, support levels. A buyer who knows the drivers can estimate their own position on the scale, which is most of what the number would have done.

Then describe the process honestly. How many calls, how long until a quote, who needs to be in the room. Uncertainty about the process is a bigger barrier than uncertainty about the price, and it is free to remove. A well built qualification form can do part of this work, as I described in routing enquiries by budget.

How specific can you be without publishing a rate card?

Much more specific than most companies allow themselves to be. A starting point is not a rate card. Saying that engagements typically begin at a certain level tells a visitor whether to keep reading, and it commits you to nothing about any particular deal you might negotiate later.

Ranges work too, if they are honest. A wide range with an explanation of what moves a buyer within it is more useful than no range at all, and it is far more useful than a narrow range you quietly ignore in negotiation.

If you genuinely cannot publish anything, publish the disqualifier instead. Saying that you work with teams above a certain size, or that you are not a fit for early stage budgets, does the same job from the other direction and costs you nothing you wanted.

What does not work is the hybrid page that shows three tiers with the prices replaced by the words contact us. That layout promises a comparison and withholds the one column that would make it a comparison. If you are going to show tiers, show what differs, as I argued in how many tiers B2B software should have.

What does hiding prices cost you in AI search?

It costs you the answer. When someone asks an assistant what a category of tool costs, the systems answering can only work with what pages actually state. A page that says contact sales contributes nothing to that answer, and your competitor who published a starting figure becomes the one described.

This is a genuine shift in how pricing research happens. Buyers increasingly form a shortlist before visiting anyone's website, from summaries built out of published pages. Being absent from that summary is a quieter loss than a lost deal and a larger one over time.

The defensive move is not to publish a number you do not want to publish. It is to make sure the page states something concrete and quotable about your model, your buyer, and your starting point. My longer argument about pricing pages in this environment is in designing a pricing page for the AI answer era.

When is hiding prices the right call?

When the price genuinely depends on a discovery you have not done yet, and when the range is wide enough that any published number would mislead more people than it helps. That situation is real, particularly for implementation heavy products and for regulated industries.

It is also right when your pricing is changing faster than your site can keep up. A wrong number is worse than no number, because it becomes a promise in the buyer's mind and an argument later.

What is not a good reason is fear of competitors. Your competitors already know what you charge. They have prospects who tell them, and sales cycles that reveal it. The only person your blank page is keeping in the dark is your buyer.

How do you test whether your page is working?

Look at what happens after the page rather than on it. Count how many enquiries arrive with a budget already stated, how many calls end with an obvious mismatch, and how far into the process a price surprise appears. Those three numbers tell you whether the page is qualifying or just collecting.

Then do the reader test. Show the page to five people outside your company and ask them to guess what you cost and who you are for. If the guesses vary wildly, the page is not communicating, whatever it says.

Watch the sales team's first email too. If it repeatedly explains something the page could have said, you have found the missing section. The best pricing page content is usually already written, sitting in a sales inbox.

What should you do next?

Open your pricing page and check whether a visitor can answer three questions from it alone. How do you charge, what makes it cost more, and is this plausibly for someone like me. If any answer is missing, write that section this week before you touch the layout.

You do not have to publish a number to be useful. You do have to stop treating the page as a door with a bell on it. Every sentence you add that helps someone disqualify themselves saves both sides a call neither wanted.

If you want an outside read on whether your pricing page is qualifying or just blocking, send me the link and I will tell you what a first time visitor would take away from it.