How many CTAs should one page have before they start competing?

The honest number is one primary action, not one button, and CTAs start competing the moment the page offers two different actions of roughly equal weight that pull the visitor’s intent in separate directions. The threshold has nothing to do with how many times a button appears and everything to do with how many distinct goals you are asking a single person to choose between. Repeat the same call as often as the page is long, and it never competes with itself. Introduce a rival of comparable prominence, and you have split the decision.

The reason is that a visitor arrives with a vague, fragile intent, and the page’s job is to consolidate that intent into one motion. Every action you present that is not the primary one asks the visitor to stop and evaluate which path is theirs, and evaluation is friction. When two buttons sit side by side at the same visual weight, neither one inherits the page’s full momentum, because the eye and the mind have to arbitrate between them. The cost is not that the visitor picks the wrong action; it is that the pause itself bleeds off the readiness the page worked to build. Worse, an equal-weight alternative often points to the easier, cheaper action, and a visitor on the edge of commitment will gratefully take the smaller ask, which means a rival CTA does not just split intent, it actively downgrades it. One clear goal removes that arbitration entirely and lets the whole page push in a single direction.

Picture a SaaS landing page whose real goal is a free trial signup. The hero has a “Start free trial” button. Lower down, after the feature section, the same “Start free trial” appears again, and again in the footer. That is three buttons and zero competition, because every one of them drives the identical action and simply catches the visitor wherever conviction happens to land. Now imagine the hero instead pairs “Start free trial” with an equally large “Book a demo” and “Download the whitepaper.” Suddenly the page has three goals fighting for one decision, and the trial signup, the action that actually matters, has to share the stage with two alternatives that quietly invite the visitor to do something easier and less committal.

One place the rule bends is the secondary action, which is legitimate when it is genuinely subordinate. A small, low-contrast “Book a demo” text link beside a bold “Start free trial” button does not compete, because the visual hierarchy tells the visitor which path is primary and offers the other only as a fallback for the small slice who need it. Competition arises from equal weight, not from the mere existence of a second option. So the principle is not “never offer more than one action” but “never offer more than one action of equal weight.” A demoted alternative serves the people the primary action would have lost anyway; a co-equal alternative steals from the people the primary action would have won.

When you lay out a page, decide the single action it exists to produce, then make that action the only one wearing the page’s full prominence. Repeat it freely at each point where a different visitor might reach readiness, and let any genuinely necessary alternative live as a quieter, plainly subordinate link. If you find yourself with two buttons you cannot rank, that is the signal you have not yet decided what the page is for, and the fix is upstream of the design. Count goals, not buttons, settle the one that matters, and the page stops competing with itself.