Quick answer
A call to action (CTA) is the button, link or line of text that tells a person what to do next, such as "Start free trial" or "Download the guide". It has an action, an outcome and sometimes a friction reducer. Measure it by click-through rate and, more importantly, by what happens after the click.
Formula: CTA click-through rate = CTA clicks ÷ CTA views × 100
What is a CTA in marketing?
A call to action (CTA) is the specific instruction that tells a person what to do next, written as a button, link, banner, form or line of text: "Start free trial", "Download the guide", "Book a demo". It is the point where attention becomes action. Every page, email or ad that is meant to produce a result has at least one, and the quality of that one instruction often matters more to the result than the polish of everything around it.
A CTA has three parts. The action, the verb that says what happens. The outcome, what the person gets. And, sometimes, a friction reducer, a short phrase that lowers the perceived cost, such as "no card required" or "takes 2 minutes". "Get my free report" has an action (get), an outcome (report) and a reducer (free). "Submit" has only an action, and it does not even say what is being submitted.
A CTA is not the same thing as the offer behind it, and it is not the conversion. The CTA is the prompt, the offer is what the person receives, and the conversion is the measured outcome (see conversion rate). A good CTA click that lands on a poor page produces a click but no conversion, which is why measuring clicks alone misleads.
What are the main types of CTA?
| Type | Where it appears | Typical wording | Best used for |
|---|---|---|---|
| Button | Hero sections, product pages, ends of sections | "Start free trial", "Add to cart" | The main action on a page |
| Text link | Inside paragraphs and navigation | "See pricing", "Read the guide" | Secondary actions, low commitment |
| Banner or bar | Top or bottom of page | "Free shipping over $50 — shop now" | Time-limited offers |
| Form | Landing pages, sidebars | "Get the checklist" above an email field | Capturing contact details |
| Pop-up, slide-in or exit prompt | Triggered by scroll, time or exit intent | "Before you go, get the template" | Last-chance offers; easy to overuse |
| Email CTA | Body or footer of an email | "Confirm your subscription" | Moving a subscriber to the site |
| Ad CTA | Ad format button or line | "Learn more", "Sign up" | Sending traffic to a landing page |
| Chat prompt | A chat invitation | "Questions about plans? Ask us" | Help at a moment of hesitation |
There is also a useful distinction by commitment. A soft CTA asks for little ("Read more", "See how it works") and suits visitors early in the journey. A hard CTA asks for a decision ("Buy now", "Book a demo") and suits visitors who are ready. A page with only hard CTAs loses everyone who is not ready; a page with only soft CTAs gives ready visitors nothing to do. Most good pages show one primary hard action and one quieter soft one.
What makes a CTA work?
Six practical rules, each checkable:
- One primary CTA per screen. Competing buttons split attention. Give the main action visual weight, and make secondary ones visibly quieter.
- A specific verb and outcome. "Download the 2026 pricing guide" beats "Submit" and "Click here".
- First-person or benefit wording can help, but test it. "Get my free report" and "Get your free report" both appear in practice; neither is a universal winner.
- Large enough to tap and legible. WCAG 2.2 requires pointer targets of at least 24 by 24 CSS pixels (success criterion 2.5.8) and a contrast ratio of at least 4.5:1 for normal text (success criterion 1.4.3), unless an exception applies (W3C, checked 2026-11-21). A button that fails these fails real users, whatever the colour theory says.
- Place it where the question is answered. A CTA should follow the paragraph that makes the case for it, not sit alone at the top of a page that has not yet explained anything. Long pages usually need the CTA more than once.
- Match the CTA to the page's promise. If the ad said "free template", the button should say "Get the template", not "Request a demo".
How do you measure a CTA?
The first measure is the CTA click-through rate, which divides clicks on the CTA by the number of times it was seen. The second is what happens after the click. Both are needed.
CTA click-through rate = CTA clicks ÷ CTA views × 100.
Sample data: a landing page CTA seen 20,000 times, with two button labels tested on separate halves of traffic, each shown to 20,000 visitors.
| Variant | Views | Clicks | Click-through rate |
|---|---|---|---|
| A: "Submit" | 20,000 | 440 | 2.2% |
| B: "Get my free report" | 20,000 | 560 | 2.8% |
How to recompute: 440 ÷ 20,000 = 2.2% and 560 ÷ 20,000 = 2.8%. The absolute difference is 0.6 percentage points; the relative lift is 0.6 ÷ 2.2 = 27.3%. A two-proportion test, with pooled rate (440 + 560) ÷ 40,000 = 2.5%, gives z ≈ 3.84 and p < 0.001, so the gap is very unlikely to be chance.
Now follow the clicks all the way through, for variant B:
| Step | People | Rate from previous step | Rate from views |
|---|---|---|---|
| Saw the CTA | 20,000 | — | — |
| Clicked | 560 | 2.8% | 2.8% |
| Started the form | 300 | 53.6% | 1.5% |
| Submitted | 120 | 40.0% | 0.60% |
Recompute: 300 ÷ 560 = 53.6%, 120 ÷ 300 = 40.0%, 120 ÷ 20,000 = 0.60%. The CTA did its job, but 260 of the 560 people who clicked never started the form, and that is a landing-page problem, not a CTA problem. A team that watched only click-through rate would celebrate; a team that watched the whole path would know where to work next.
What does the evidence say about CTA changes?
The best-known published test is from the 2008 Obama campaign. The team tested four button labels ("Sign Up", "Sign Up Now", "Join Us Now" and "Learn More") against six media variants above the button. The winning combination, "Learn More" with a family image, had a signup rate of 11.6% against 8.26% for the original page, a 40.6% relative improvement (Optimizely, "Obama's $60 million dollar experiment", checked 2026-11-21).
Read it carefully. The lift belongs to the combination of button and image, not to the button alone, so "Learn More beats Sign Up" is not what the test showed. The write-up comes from a testing vendor with a commercial interest, and the dollar figures attached to it vary between sources, so treat the 40.6% as one real, well-documented result, not a promise. The transferable lesson is about method: the team tested several variants against the original at once and measured by the people who saw each variant.
A second finding is on the user side rather than the test side. In Stanford's 2002 web credibility study of 2,684 participants, the look of a site was the most-mentioned factor, appearing in 46.1% of comments about whether a site was credible (Fogg et al., "How Do People Evaluate a Web Site's Credibility?", October 2002, checked 2026-11-21). A CTA that looks careless can cost trust before anyone reads the words.
What are the common mistakes?
- Vague labels. "Submit", "Click here" and "Learn more" with no object tell the visitor nothing about what happens.
- Too many CTAs. Five equal buttons mean no primary one.
- Hiding it. A CTA below a long, unbroken wall of text, or one that looks like body copy.
- Promise mismatch. The ad promises a template and the button asks for a sales call.
- Measuring clicks only. A click is not a conversion; follow the whole path.
- Changing several things and claiming the button won. As the 2008 example shows, a lift may belong to the combination.
- Overusing pop-ups. An immediate full-screen prompt raises the bounce rate and annoys returning visitors.
How do you track a CTA in TapCub?
Use the sample above as a walk-through.
- Name the click. In the button's click handler call
tapcub.track('cta_click', { id: 'hero-report' }). The event name must match a goal if you want it to count as one; properties keep at most 20 keys, with strings cut at 200 characters. - Make it a goal. Create the matching goal under Web analytics › Goals, so the CTA's conversions appear next to your other goals and in channel reports.
- Follow the path. In All-platform analytics › Behavior › Behavior analysis build a funnel: CTA click, form start, form submit. Each step shows its rate against step one and against the previous step, and the largest drop is marked, which is how you would find the 260 people who clicked and never started the form.
- Compare variants. If you test two labels in your own testing tool, send the variant name as an event property and compare the rates by that property. TapCub does not split traffic or run A/B tests; it measures.
- See where people actually click. All-platform analytics › Behavior › Heatmaps shows clicks on a page, including rage clicks (three or more clicks on the same element within 500 milliseconds), which can reveal a button that looks clickable but is not working.
Two automatic captures help too: outbound link clicks to other domains and file downloads are recorded without code, and form submits can be recorded as ID, name and action when you turn on form tracking; the form fields' contents are not stored.
Sources
- W3C: WCAG 2.2, Target Size (Minimum) and Contrast (Minimum), checked 2026-11-21
- Optimizely: Obama's $60 million dollar experiment, checked 2026-11-21
- Fogg et al.: How Do People Evaluate a Web Site's Credibility? (Stanford Persuasive Technology Lab, 2002), checked 2026-11-21
Was this explanation clear?
We use your answers to fix and expand this page.
Couldn’t send. Please try again in a minute.
Thanks for the feedback. We’ll use it to improve this page.You’ve already sent feedback on this page. Thanks.

Product Analyst
Focuses on event analysis, conversion funnels, and retention, using clear metric definitions and structured analysis to explain product performance.