Call to action

A call to action is the element asking the user to do the thing: a button, a link, a form submit. The term comes from direct marketing, where the whole point of a mailing was the response it produced.

The label is the design

The most common improvement available on any CTA is its wording. A label describing the outcome outperforms one describing the mechanism, because the user is deciding whether they want the outcome.

“Submit” describes what the form does. “Book the pickup” describes what the user gets. “Download the report” tells you what happens next; “Click here” tells you nothing and is useless out of context, which is exactly how a screen reader user encounters it when listing links.

First person can help where the action is a commitment, but the difference is smaller than the difference between naming the outcome and not naming it.

One primary action per view

If everything is emphasised, nothing is. A view should have one obvious next step, with alternatives visibly secondary. This is a hierarchy decision before it is a colour decision.

In practice

Two tests. Read the label alone, out of context: does it say what will happen? And count the competing primary actions on the screen: if there is more than one, the page has not decided what it wants.