Why Isn't My Landing Page Converting?

Your ads earn clicks, but the page they land on turns too few into leads or sales. So you tweak the button, rewrite the copy, swap the hero image, and the number barely moves. Every tweak burns another week of paid traffic, and you still can't say whether the page, the ad or the audience is at fault.
A landing page that isn't converting usually fails on relevance, not design. It's the wrong destination, it breaks the ad's promise, or its first screen can't say who it's for. Audit in that order, then check trust. Start with the page losing the most conversions in absolute terms, not the one with the worst rate.
Which landing page should you fix first?
If you run more than one page, the first decision is where to look. The instinct to start with the lowest conversion rate usually points you the wrong way. Conversion rate is a ratio, and ratios hide volume. A page converting at 1% on a few hundred sessions a month can only recover a handful of conversions, however well you fix it. The worst-rate page is often doing a different job anyway, such as a guide fed by cold social traffic, or a campaign page left running after its offer expired. Comparing raw rates across pages like that compares audiences, not pages.
Score lost value instead. Multiply each page's monthly sessions by the gap between its current and target conversion rate to get lost conversions. Then multiply that by what a conversion is worth.
Here's a hypothetical account. A paid search page gets 15,000 sessions a month and converts at 2% against a 3% target, so it loses 150 conversions a month. A webinar page gets 1,200 sessions and converts at 0.8% against the same target. It looks far worse but loses only about 26. The average-looking page is the bigger leak.
Two inputs decide whether the score is honest:
- The target should come from your own data. Use the best rate a comparable page achieves on the same traffic source. Published landing page benchmarks mostly come from page-builder vendors' customer bases, with a traffic mix you can't see.
- The value should come from downstream. In e-commerce, that's order value times margin. In lead generation, it's the lead-to-opportunity rate from your CRM.
When two pages lose similar value, fix the paid one first, because you're paying for every session it wastes.
Assembling that table is the slow part. Sessions sit in Analytics, spend sits in each ad platform, and lead quality sits in the CRM. SproutMe Companion answers a question like "which landing page lost us the most qualified leads last month" from Google Ads, Meta, Analytics and HubSpot data held in one place. The ranking doesn't have to start from three exports.
Valuing conversions also settles two pieces of testing advice that look like they conflict. One says to judge first-screen variants on conversions rather than scroll depth. The other says conversion rate alone misleads, and that cost per conversion and lead quality should decide a test. Once value sits in the score, the two agree. A variant that lifts form fills while lowering lead quality has lost, whatever its rate says.
The ranking has two exceptions:
- Pages with too little traffic to measure. Park them or group them by template. Two extra signups can double their rate and mean nothing.
- Breakage on a final-step page. A low-traffic page at the last step before a purchase still gets checked for outright faults, such as a form that won't submit. That repair jumps the queue.
Finally, segment the top-ranked page by traffic source and device before you redesign it. If paid search converts and paid social doesn't, the page may be fine and the promise feeding it isn't. The full method, including ICE scoring to break ties between fixes, is in which landing page to fix first.
Should the ad land on your homepage at all?
Once you know which traffic matters most, check where it lands. For non-brand search, it should almost always land on a dedicated page built around the one thing the ad promised. Keep the homepage for branded and exploratory searches.
A homepage is a routing page. It has to orient prospects, existing customers, job applicants and journalists all at once. Picture a visitor who clicked an ad about same-day boiler repair. They meet a navigation bar, a banner about the company's story and several competing calls to action. Each one is a way to wander off, and paid visitors arrive mid-task with little patience for detours.
A landing page serves one audience, one offer and one action. It repeats the ad's promise, removes most exits, and gives you a single goal you can measure and test cleanly.
The destination also changes what you pay. Quality Score has three components:
- expected click-through rate
- ad relevance
- landing page experience
Landing page experience is Google's judgment of how relevant and useful the page is to someone who clicked that ad. A homepage is relevant to everything you sell, so it closely matches very few keywords. A below-average rating makes your click dearer than it is for a competitor with a matched page.
Keep two caveats in view. The 1–10 score in the interface is a diagnostic proxy, not the auction input itself. And Google doesn't publish how the components are weighted.
You'll also see confident claims that landing pages convert some fixed percentage better than homepages. We haven't found one that traces back to a study with a stated sample, so treat those figures as marketing.
The homepage is the right destination when the visitor wants your brand rather than a specific offer. That covers:
- branded search
- broad queries where you can't yet tell what the visitor needs
- long sales cycles where buyers want to verify you first
- single-offer businesses whose homepage already works as a landing page
The quick test is whether the ad's headline could be the page's headline without anyone noticing the jump.
To act on it, export final URLs by ad group, filter for the homepage and sort by cost. The homepage tends to become the default for every ad group nobody built a page for, and that tail can hold more budget than you'd expect. Group keywords by intent rather than building a page per keyword, and start with the highest-spend theme. Run the homepage against the new page as a split before switching everything over. The detailed case is in whether your Google Ads should go to your homepage or a landing page.
Does the page say what the ad promised?
A dedicated page fixes the destination, not the message. Your page delivers on the ad only when the first screen repeats the ad's offer in the ad's own words. In practice, that comes down to four elements:
- Headline: restates the ad's main claim, recognisably if not word for word.
- Offer: price, discount, trial length and terms identical to the ad.
- Terms: the noun the searcher typed, not your internal product name.
- Call to action: continues the ad's action, so "get a quote" lands on a quote button rather than "book a demo".
On paid social the visual counts too. If the ad showed one product in one colour, the page should show that product.
The common failure isn't a wildly wrong page. It's a generic one, usually a homepage or category page where everything is true and nothing answers the question the ad raised.
Run the landing page audit per ad rather than per page:
- Start from the ad as it's actually served.
- Open the page on a phone.
- Look only at the first screen, and mark each element as a match, a partial match or a miss.
- Repeat for every ad pointing at the same URL.
The last step matters most. A page written for one campaign quietly becomes the destination for four more, matching the first ad and contradicting the rest.
A mismatch has a recognisable signature in your data. Look for an ad group with a healthy click-through rate and a conversion rate that lags its siblings. The ad is doing its job, and the page isn't following through.
The cleanest evidence comes from two ads that share a page, because the page is held constant and only the promise differs. Bounce rate on its own can't tell you this, since slow loading and confusing layouts produce the same early exits.
In Google Ads, look for ad groups rated well on ad relevance but below average on landing page experience. That rating also reflects speed and mobile usability, so treat it as a pointer, not a verdict. A mismatched page charges you twice: once in a pricier click, and again in lost conversions.
Matching words still isn't the whole job:
- Keyword echo isn't the promise. Pasting the keyword into the headline can still dodge the offer the ad led with.
- Some promises belong to the ad. If the ad promised a discount that has ended, fix the ad rather than the page.
- Creative outpaces pages. Teams that rotate creative weekly and update pages quarterly drift out of match by default, so the page check belongs in every new ad launch.
- One page, one promise. Ads that share an offer and a key term can share a page, but a page carrying several promises matches none of them.
The four-element audit is worked through in whether your landing page says what your ad promised.
What has to fit on the first screen?
Message match decides whether a visitor stays. The first screen decides whether they scroll. Attention is front-loaded and drops sharply past the fold, so that space isn't there to hold your whole argument. It has one question to answer at a glance: is this for me? Five elements answer it:
- An outcome headline that states what the visitor gets, matched to the ad that sent them.
- A subheadline saying who it's for, what it replaces or what makes it different.
- One concrete call to action, such as "Book a 20-minute audit" rather than "Learn more".
- A visual of the product in use, compressed so it doesn't slow the screen it's meant to sell.
- One trust cue: a logo strip, a rating or a short quote from a peer.
Everything else works better one scroll down. Feature lists, pricing tables, long testimonials and FAQs belong below the fold. A second button of equal weight turns a decision into a dilemma. Sliders are the clearest casualty, because each visitor sees whichever slide loaded and no message gets the space. Auto-playing video, chat pop-ups that open on arrival and consent banners covering half a phone screen all compete with the headline for the same few seconds.
There's no single fold, either. It sits wherever the viewport ends. Design for the smallest screen your traffic commonly uses, which for paid social usually means a phone, and check what survives the crop.
The layout bends in two cases:
- Expensive or complicated purchases. An aggressive buy button on the first screen asks for commitment before the page has earned it. The first screen should work as an introduction instead, asking for something smaller, such as a guide or a short consultation.
- Warm visitors. Intent can override price. Someone who searched your brand plus "pricing", or clicked a retargeting ad, came for a transaction and should find the form without scrolling.
If you run both cold and warm traffic, build both kinds of page rather than one compromise.
Before running a test, try a quick check. Show the first screen to someone outside the project for a few seconds, hide it, and ask what's on offer and what to click. If they can't say, more traffic won't help. Then test one element at a time, starting with the headline, so you learn why a variant won. The element-by-element version is in what should be above the fold on a landing page.
Do visitors trust you enough to convert?
A page can match the ad and nail the first screen and still lose a visitor who isn't sure you're real. That doubt weighs most on the visitor paid ads mostly bring: someone who has never heard of you. Trust builds in sequence. Design earns the first few seconds, visible identity proves a real business stands behind the page, and reviews prove you deliver.
Design is a gate, not proof. Visitors who can't judge your expertise judge your presentation instead. A cluttered or dated page loses people before they reach your credentials, and a clean one only buys the next few seconds.
Identity does the persuading. Stanford's guidelines for web credibility put showing a real organisation near the top of the list. That means a physical address, clear contact details (phone, address and email), and bios of the people behind the site. The guidelines date from June 2002, but the problem they answer hasn't changed: online, a visitor can't walk into your premises. Today, meeting them means:
- a named contact
- an email address on your own domain
- real photos instead of stock imagery
- an About page that says who runs the business
Reviews matter more the less people know you. A 2023 conjoint study by Sung, Chung and Lee found that consumers preferred high star ratings, a large number of reviews and a trustworthy platform, even when similar products were cheaper. For unfamiliar brands, they were willing to pay more for all six attributes tested.
The study covered shopping platforms rather than merchant sites, so applying it to your page is an inference. The direction is still clear:
- Volume and rating both count.
- Reviews verifiable on an independent platform tend to outweigh quotes you host yourself.
- Reviews work hardest beside the form or the pricing.
Here the earlier advice pulls against itself, and it's worth reconciling. The first-screen rule allows one trust cue. The landing page rule strips the navigation that would lead a doubtful visitor to your About page and team. Both are right for most paid traffic, but an unknown small brand loses the most when its proof sits a click away. Bring identity onto the page itself:
- one trust cue above the fold
- reviews next to the decision
- a named contact, phone number and address in a minimal header or beside the form
Then clear the small failures that quietly hand trust back: typos, broken links, stale notices, a missing privacy policy, no https, or a button that does nothing on mobile. Each one reads as a preview of how you'd handle an order, and each is cheap to fix. More on the signals that move sceptical visitors is in what makes website visitors trust a small business.
Conclusion
A landing page that isn't converting is rarely fixed by redesigning it. Work the problem in order:
- Rank pages by the value they lose, not by their rate.
- Segment the leader so you know whether the page or its traffic is at fault.
- Check that non-brand ads land on a dedicated page.
- Check that the page repeats each ad's promise in the ad's own words.
- Check that the first screen tells the right visitor, at a glance, that it's for them.
- Make sure a stranger can see you're real before you ask them to commit.
Most pages fail one of the first few checks long before button colour matters. That's why the checklist runs in this order.
See how Companion's daily brief reports changes in spend and conversion across your channels, so a page that starts losing value shows up the next morning rather than at month-end.
Frequently Asked Questions
There's no honest universal figure. Most published landing page benchmarks come from page-builder vendors' own customer data, with an unknown mix of industries, offers and traffic sources. A more useful target is your own best comparable page on the same traffic source, or the page's own rate before it slipped. That reflects your audience, offer and price point, which no cross-industry median can.
Long enough to cover a normal sales cycle. Make one change at a time and compare conversion on the same traffic source. Many small businesses don't get enough traffic for a clean A/B test, so a before-and-after comparison is often the realistic option. If you change reviews, the headline and the hero image in one week, you won't know which moved the number.
Yes, in two ways. Load speed and mobile usability feed into Google's landing page experience rating, which affects what you pay per click. A heavy hero image also slows the very screen meant to persuade, and the effect is worst on mobile. Compress images and cut decorative ones first, but speed alone won't rescue a page that doesn't match its ad.
Get a complimentary audit to uncover AI opportunities hidden in your data.
Put these strategies to work


