Call to action

A call to action is the point on a page where you ask the visitor to do something. It is usually a button, sometimes a form or a link, and its job is narrow: to make the next step obvious to somebody who has just decided they are interested. The term gets used loosely for the button itself, and the button is the smallest part of it.

In short

• The part of a page that asks the visitor to do one specific thing.
• One action per page, repeated, rather than several competing ones.
• The wording matters more than the colour.
• It only works if the page has already given a reason to act.

The failure is almost never the button

Businesses redesign buttons. They change the colour, round the corners, make it bigger, and the number of inquiries does not move, which is discouraging and entirely predictable.

Three things have to be true before a button does anything.

• The visitor understands what is being offered.
• They have a reason to want it.
• The step being asked of them feels proportionate to how interested they are.

A button is only the third of those, and it is the one that gets all the attention because it is the easiest to change.

A page that has not explained the service will not start producing inquiries because the button turned orange.

Say what happens, not what the form does

The words on the button are the cheapest improvement available on most sites and the most frequently skipped.

Submit, Send, Click here and Learn more all describe a mechanism. They tell the visitor what the software will do, which nobody wanted to know. The alternative is to name the outcome from the visitor’s side, so that the button reads as the end of a sentence beginning with “I want to”.

• Submit becomes Send my inquiry.
• Learn more becomes See what is included.
• Contact us becomes Ask about your project.
• Sign up becomes Get the monthly summary.

The second version in each pair is longer, which is fine. Buttons are not read the way prose is read, and a specific four-word button beats a vague one-word button every time.

There is a second gain hiding in this. Writing the outcome forces you to state what the visitor actually gets, and a surprising number of pages cannot finish that sentence, which is useful to discover.

Where it goes

The rule most people have heard is that the action belongs above the fold. That is half right and the half that gets misapplied.

Two kinds of visitor arrive on any page. One already knows what they want, often because they searched for exactly this, and any delay is friction for them. The other needs to read first. Serving only the first group puts the button on screen immediately and leaves the second group with nothing to act on when they finally finish reading, four screens down.

So the action appears early, and again after each part of the page that does persuading. On a long page that might be three or four times. This is repetition of one action, which is entirely different from offering several, and it is why a scroll on a phone should never take somebody past the point of deciding without something to tap. How little fits in that first screen is the subject of above the fold, and it is less than most people picture.

One action, not a menu

The most damaging version of this problem looks like generosity. A page offers a quote request, a brochure download, a newsletter, a Messenger link and a phone icon in the corner, on the reasoning that different people prefer different things. What the visitor experiences is a decision they did not ask to make, at the exact moment they had decided to act. Some of them will pick the easiest option instead of the most valuable one. Some will pick nothing.

Every page can carry one primary action. Whatever else exists should look noticeably quieter: a plain link rather than a button, in the body rather than beside the main action. If two things genuinely deserve equal weight, that usually means the page is trying to serve two audiences and should be two pages.

Making it easy to tap

On a phone, most of this is physical rather than persuasive. A button that a thumb can miss is a button that gets missed, and the reader will not conclude that they mistapped. They will conclude the site is broken. Material Design, which is Google’s own design system, sets 48 CSS pixels as the minimum for anything tappable and asks for 8 pixels of clear space around it. The accessibility standard is lower, at 24, and that is a floor rather than a target.

The form behind the button is where most of the loss happens. Each field is a small tax, and a required field asking for something the visitor is not ready to give is a large one. Name, one way to reply, and a message box is enough to begin a conversation. Anything else can be asked in the reply, when a person is on the other end rather than a page.

Words you will hear

• Primary CTA. The one action the page exists to produce.
• Secondary CTA. A quieter alternative for a different reader, visually subordinate to the primary one.
• Sticky CTA. A button fixed to the bottom of the screen so it stays reachable while scrolling. Effective on phones and easy to overdo.
• Micro-copy. The short reassurance near the button, such as what happens after sending. It removes hesitation more often than the button text does.
• Friction. Anything between wanting to act and having acted. Extra fields, account creation, unclear next steps.
• Above the fold. What is visible before scrolling.

How you know whether it is working

The number that matters is not clicks on the button. It is completed actions divided by visitors to the page, which is the conversion rate, and it is the only version that accounts for people who tapped and then abandoned the form.

Set that up before rewriting anything, because otherwise every change is a matter of opinion and the loudest opinion in the room wins.

Then start with the cheapest edit available. Go to the page that matters most, read the button, and if it says Submit or Learn more, replace it with the thing the visitor gets. That takes a minute and it is the one change on this page that needs no design work, no developer and no meeting. Everything the rest of a landing page has to do to support that button is a longer conversation.