How Much Should a Page Explain Before It Asks?

How much should a page explain before it asks?

Enough that the reader can tell whether the ask is for them, and no more. That threshold is much lower than most marketing pages assume and much higher than most landing pages allow. The test is not length, it is whether a reader could answer the question is this for me before they hit the button.

Getting this wrong is the most common conversion problem I see, and it fails in both directions. A page that asks too early gets clicks from people who then bounce, which looks like a traffic quality problem. A page that explains too long never gets asked at all, which looks like a copy problem. Neither diagnosis is right.

What follows is how I decide the ratio, built on a principle that has been in the usability literature for two decades and applies to marketing pages much more directly than people realise.

Why does showing everything make a page less persuasive?

Because the act of showing something is itself a message. Nielsen Norman Group's writing on progressive disclosure puts it in one sentence: in a system designed with progressive disclosure, the very fact that something appears on the initial display tells users that it is important.

Turn that around and the cost becomes obvious. If everything appears on the first screen, nothing on the first screen is marked as important. You have not given the reader more information, you have removed your own ability to say which information matters, and they now have to do that sorting themselves.

The underlying tension is real rather than invented, and the same article names it honestly. Users want power, features, and enough options to handle their special needs, and users also want simplicity because they do not have time to learn a profusion of features in enough depth to pick the ones that suit them. Both are true of the same person at the same time.

The resolution proposed is the one I use on marketing pages. Initially show only a few of the most important options, then offer the larger set on request. That is a design instruction, not a copywriting one, and it decides the structure of the page before a single sentence is written.

What does the first screen actually communicate?

Your judgement about what matters. Before a reader has evaluated your argument, they have evaluated your priorities, and they did it by looking at what you put first.

This is why a hero section stuffed with six value propositions reads as uncertainty. It is not that six is too many to absorb. It is that a company which cannot say which one matters most is telling the reader it does not know its own customer, and that impression forms in about a second and is hard to reverse further down the page.

The same logic applies to the ask. A button on the first screen says this decision is available now. If the product genuinely is simple enough to try without explanation, that is honest and it works. If it is not, the early button is a claim your page then spends two thousand words contradicting.

My practical version is a single sentence test. Write down what the first screen claims, including what its layout implies about priority, then ask whether the rest of the page supports that claim or argues with it. Most pages argue with themselves and nobody noticed because nobody read the page as a whole. The narrow version of this is covered in designing a hero section that says what you do.

How do you decide what gets deferred?

By what proportion of readers need it, not by how important it is to you. This distinction is the entire skill, and it is where internal politics usually wins over evidence.

Nielsen Norman Group frames the deferral rule around rarity, describing progressive disclosure as deferring advanced or rarely used features to a secondary screen. Rarely used is a measurable property of your audience. It is not the same as unimportant, and things can be simultaneously critical and rare.

The example the article gives for websites translates directly. It says an ecommerce site might mention a few key product attributes on the primary page and let users click through to a secondary page for the full specifications. The specifications did not become unimportant. They became available on request, to the smaller group who wanted them.

Two implications for a business site. Compliance detail, integration lists, and technical constraints belong one click away rather than absent, because the minority who need them need them absolutely. And the thing you are proudest of is not automatically first, because pride is not a measure of reader demand. The exception, worth naming, is price: Nielsen Norman Group's research on business sites reports pricing as the number one information need, so it is one of the few things that should almost never be deferred.

Where should the ask sit on a long page?

After the section that resolves the largest doubt, and again at the end. Not in the middle by default, and not only at the bottom, because a reader who was convinced at the halfway point should not have to scroll past three more arguments to act.

The way to find that spot is not aesthetic. Look at what people ask you before they buy, put the section answering the biggest one where it belongs in the argument, and place an ask immediately after it. You are placing the button at the moment of maximum readiness rather than at a convenient visual break.

Repeating the ask is fine and it is not pushy when the intervening content is doing real work. What is pushy is repeating the same ask three times with nothing new between the repetitions, because that reads as pressure rather than as availability. The number of asks a page carries is the same question asked from the other end, and the answer is the same: as many as there are genuine moments of readiness, and no more.

One structural note about where this breaks. If your page needs the ask in five places because readers keep reaching the end unconvinced, the problem is not placement. Adding buttons to a page that is not persuading anybody just distributes the failure more evenly.

Does the answer change by traffic source?

Substantially, and this is the part that has changed most in the last two years. A reader arriving from a search result has already been told something about your page before they land, which means part of the explaining was done for you.

The information-foraging research explains why that matters. Nielsen Norman Group describes people choosing where to go based on two estimates, how likely a page is to answer their question and how long it will take to get that answer, both formed before the click. Someone who clicked has already accepted both estimates, so your first screen should confirm them rather than start over.

The practical failure is a page that re-introduces the topic to somebody who already knows what they came for. That is not an explanation, it is a delay, and it burns the goodwill that got them there. Confirm what they expected in the first sentence, then move.

A reader arriving cold, from a mention or an ad or a link with no context, needs the opposite. They have made no estimates and accepted nothing, and the same page has to do all the orienting work. One page cannot be perfect for both, and I would tune the first screen for whichever source actually sends you buyers rather than whichever sends you the most visitors.

What about forms specifically?

A form is an ask that arrives before the explaining is finished, which is why it feels different from a button. Every field is a question you are asking before the reader has decided they want a relationship with you.

The progressive disclosure principle applies here more literally than anywhere else on a page. Ask for what you need to take the next step, then ask for the rest later, when the person has more reason to answer. Nielsen Norman Group's guidance about deferring secondary material as a key guideline for mobile design is doubly true of a form on a phone.

What I would not do is hide required fields to make a form look shorter. If a field is genuinely needed, revealing it after someone has started is worse than showing it upfront, because the form grew after they committed. Deferral is honest when the deferred thing is genuinely optional or genuinely later, and manipulative when it is neither.

The one field worth arguing about individually is the phone number, because it changes what the reader thinks is about to happen. Asking for it converts a request for information into a request for a call, and that reframing costs more submissions than its length suggests. If you need it, say why next to the field.

How do you test this without an experimentation platform?

Read the page out loud from the top and stop at the first moment you would personally close the tab. That is not a substitute for real testing and it catches a surprising proportion of the problem, because the failure is usually structural rather than marginal.

The second method is cheaper than a test and more informative than most. Watch one real person use the page while saying what they are looking for, and do not help them. Five minutes of that tells you whether your first screen answered the question they arrived with, which is the thing you are actually trying to learn.

If you want a number rather than an impression, the honest one available to you is the ratio of people who reach your ask to people who use it, per traffic source, over enough weeks to mean something. That is a real measurement and it does not require an experimentation tool.

What I would avoid is optimising on a single metric like time on page, which moves for contradictory reasons. A reader leaving quickly because they got the answer and a reader leaving quickly because they gave up look identical in that number, and acting on it without knowing which is how pages get longer without getting better. The related discipline of writing for people who never read linearly is in designing for the skim reader.

What should you do next?

Take your most important page and mark every element on the first screen with a number: what percentage of your readers genuinely need this to decide. Anything under about a fifth is a candidate for deferring one click away, with the exception of price, which stays.

Then find the section that resolves your single biggest objection, and make sure an ask sits immediately after it. If you cannot identify which objection is biggest, that is the real task, and your last ten sales conversations will tell you in an afternoon. The sequencing argument generalises, and I worked through the pricing-page version of it in what order a pricing page should answer objections in.

If you have a page that gets traffic and does not convert, and you want a second opinion on whether it is explaining too much or asking too early, reach out. It is usually one of the two and it is usually visible in ten minutes.