If you are asking why nobody is clicking your CTA, the first instinct is a new colour or label. Such a fix only addresses the cause by luck. A button can fail for three different reasons: visitors never reach it, they do not recognise it as a button, or they do not know what happens when they click. Each needs a different change and different data.
This guide covers one button on one page: three questions in order, the decision to move, rewrite or remove competing calls to action, and measuring the click before and after. For a review of the whole path from ad to form, start with a website conversion audit.
Three questions before you change how the button looks
The order matters. There is no point rewriting a button most visitors never see, or moving up one that looks like plain text. Stop at the first question where the answer is no.
Scroll the table sideways to see all columns.
| Question | Where the data comes from | Warning sign |
|---|---|---|
| Do visitors reach the button? | Scroll map for the device in question | Most views end above the section with the button |
| Do they see it as a button? | Element ranking on the click map, plus recordings | Clicks next to the button or on non-clickable elements nearby |
| Do they know what happens next? | Recordings, chat questions, the copy around the button | Someone hovers over the button, hesitates and looks for details elsewhere |
Measure phone and desktop separately: on a phone the button can end up below several screens of photos. Note the page version and dates of your data so you can compare against the same baseline later.
Step 1: do visitors reach the button at all
Mark where the button sits on a page screenshot and read from the scroll map what share of views got that far. If few reach it, a low click rate says nothing about the button itself. How to read the thresholds and spot a false bottom is covered in our guide to scroll depth; here a yes or no is enough.
Reaching the section does not mean the button was visible: on a phone a cookie bar, promo banner or chat window can hide it. If recordings show the button disappearing under a fixed element, check it with our guide on how sticky elements cover buttons. When reach is weak, the decision is usually to move the button up or repeat it higher, not to rewrite it.
Step 2: does the button look like a button
When reach is good but clicks are rare, open the click map and element ranking. Look for two patterns. First, clicks on non-clickable elements nearby: a heading, an icon, a product photo or bold text that someone mistook for a link. Second, a button that ranks below elements irrelevant to the page goal, such as the logo or a footer link.
Both suggest the button does not say "click me", or something else says it louder. Confirm it in a few recordings linked to those clicks. Repeated clicks with no response from the page are a separate problem; see our guide to rage clicks and dead clicks. With low traffic, keep in mind that the Meetlead dashboard marks a map as ready only after 50 page views per device.
In a 2017 Nielsen Norman Group eyetracking study with 71 participants and nine pairs of pages, looking for where to click on pages with weak clickability signifiers took about 22% longer on average than on versions with strong signifiers. NN/g: Flat UI Elements Attract Less Attention and Cause Uncertainty. The study measured finding an element on desktop, not click rates on your site.
Typical causes: a button in the section background colour, plain text with no outline, a tap area too small for a phone, or several look-alike elements when only one leads anywhere. Give the button an unmistakable look and take that look away from non-clickable elements.
Step 3: is it clear what happens after the click
A button can be visible and recognisable, yet nobody clicks it because its promise is unclear. "Next" or "Submit" do not say whether a form, a price list, a sales call or a payment comes next. Recordings show the cursor pausing over the button, a scroll for details, then a return or an exit. Chat then brings questions like "how much does it cost before I sign up?".
In a 2017 article, Nielsen Norman Group describes how the generic label "Get Started" can fit almost any goal and lead users somewhere they did not expect, such as a sign-up form. The authors recommend naming the specific action. NN/g: "Get Started" Stops Users. This is a qualitative observation from usability research, not a measurement of how a label affects click counts.
Instead of: "Submit". Better: "Request a quote". Below the button: "We reply within one business day. No phone number needed." Only make promises your business actually keeps.
Timing matters too. "Book a consultation" right below the headline, before visitors learn what it covers and costs, asks for a decision too early. After the pricing section, the same button reaches someone who knows what they are agreeing to.
Move it, rewrite it or remove competing calls to action
The diagnosis points to one decision. Weak reach: move the button up or repeat it higher. Clicks beside it: change its look. Hesitation and questions about terms: a new label and an explanatory line. Competing calls to action are a separate case: when "Call us", "Download the catalogue", "Join the newsletter" and "Request a quote" sit side by side, clicks often spread across all four, and the main goal loses out.
- Each section has one primary call to action; the others look clearly weaker
- Secondary actions, such as the newsletter, moved lower or into the footer
- The label names an action, not a generic step
- A line next to the button says what happens after the click
- You change one thing at a time and note the date of the change
With several problems across pages, rank them with ICE scores from heatmaps and recordings. For a single button, start with the first question you answered no to.
Measure the click as a goal before and after the change
A fictional example: an accounting firm has a "Submit" button on its services page. Most views reach it, and recordings show people pausing over it and going back to the price list. The firm changes the label to "Request a bookkeeping quote", adds a line about response time and leaves the rest unchanged.
- Name the click.Add a data-livechat-track attribute with a readable name to the button, such as cta-quote. The click then shows up as an event in the visit timeline.
- Create a goal.In Analytics, create an active goal such as cta_quote and call meetlead.goal('cta_quote') after the click or after the form is sent, depending on what you count as success.
- Record a baseline.Count goal completions, section reach and page views per device over full weeks.
- Compare a matching period after the change.Then collect the same number of full weeks and compare completions as a share of views, not raw clicks.
A before-and-after comparison is not an A/B test, because traffic and season change between periods. Treat the result as a strong hint. When traffic allows, comparing both versions at the same time is covered in our guide to A/B testing on your website. In Meetlead Analytics you can view goals alongside click and scroll maps and visit recordings; you install the Heatmaps app for a chosen website, and recording storage and history length depend on your plan.
Questions about a button nobody clicks
Will changing the button colour increase clicks? Only if the look was the problem, meaning the button blended into the background or resembled plain text. If visitors never reach it or do not know what it promises, a new colour changes nothing. Answer the three diagnostic questions first, then choose the fix.
What is a good CTA click rate? No single number fits every page; a contact page and a long guide have different jobs. Compare the button with itself before and after a change, and with other elements on the same page, not with someone else's report.
