How to match a landing page to the promise that sent people there

Someone sees an advert for 20% off winter coats, clicks it, and lands on a homepage with a hero banner about the new spring range. The coats are in there somewhere, two clicks deep, and the discount is mentioned in small print at the checkout. Every person who made that click wanted coats at a discount, and the page made each of them start their search again. Plenty won’t bother.

That gap between the promise and the page is what people mean by message match. The advert, email, social post or affiliate link makes a promise, and the landing page either keeps it within a glance or it doesn’t. Nielsen Norman Group’s work on information scent describes how people judge whether a page is worth their time from the cues in front of them, and how, when a landing page offers too little context, they decide quickly the page is no use to them and leave. A visitor who has just clicked an advert arrives with a very specific scent in mind. If the first screen doesn’t carry it, they assume they’re in the wrong place.

Google treats this as part of how it rates your ads. Landing page experience is one of the three components of Quality Score, described as how relevant and useful the page is to the people who click your ad. Google describes Quality Score as a diagnostic that plays no direct part in the ad auction, though a below average rating is a fair hint that your page and your promise have drifted apart.

What usually goes wrong

The most common failure is sending every campaign to the homepage, or to one generic landing page, because building specific pages is slow and nobody owns them. The second is drift. The campaign team changes the offer or the creative on Tuesday and the page still says what it said last month. The third is the page that tries to do everything. It wants the visitor to buy, sign up for the newsletter, follow on social, read the brand story and download the app, so it does none of them well. A landing page with one job is easier to scan, easier to measure and easier to test, because you know what success looks like before anyone arrives.

There’s also the mismatch of tone and audience. An email to lapsed customers that says “we miss you” and lands on a page that treats them as strangers, explaining what the company does from scratch, breaks the promise in a quieter way.

Where to look for evidence

Start in analytics, broken down by campaign and source, because a blended landing page report hides the problem. A page can look fine overall while one campaign feeding it bounces at twice the rate of the others. Look at engagement on the first screen, scroll depth, how many people reach the next step, and whether the visitors from a given campaign end up searching or clicking away from the page they landed on. Site search terms from landing page visitors are especially useful, because people often type the thing the advert promised.

Then do the boring check by hand. Put the live advert next to the page and read them side by side, on a phone, because that’s where most paid social traffic arrives. Session recordings of visitors from one campaign will usually show you within a few minutes whether people are looking for something the page doesn’t show them.

What is worth testing

The highest-value test is usually matching the first screen to the promise: the headline repeats the offer in the advert’s words, the product or category shown is the one advertised, and the next step is obvious. After that, try reducing the page to one primary action, or routing a campaign to a deeper page such as the relevant category or product instead of the homepage.

A test on a landing page has a particular trap. The traffic mix changes constantly, because campaigns start, stop and change budget while the test runs. Ask for the result to be reported by campaign or source as well as overall, and make sure the split is applied to the same traffic, so one arm isn’t getting more of the cheap search clicks than the other. If the page serves several campaigns, a win for one audience can be a loss for another, and the blended number will hide it.

Measure the right outcome too. A landing page that reduces bounce rate but sends more people to the product page without any change in orders has made the funnel longer, which may not help anyone. Agree the primary metric before launch, usually the next meaningful step or the order, with bounce as a supporting signal.

The version to refuse

The manipulative version of message match is the advert that promises something the page can’t deliver. A headline price that only applies to one size nobody stocks, a “free” offer that turns out to need a subscription, or a discount that ended last week. In the UK, the Digital Markets, Competition and Consumers Act 2024 lists bait advertising, inviting people to buy at a price you have reason to think you can’t supply in reasonable quantities, among the practices banned outright, alongside bait and switch and false limited-time offers. If a test idea relies on the advert overpromising so the page can catch the click, turn it down. The cost lands later, in refunds, complaints and customers who don’t come back, and none of that shows up in the test result.

What to ask whoever builds the test

Ask which campaigns feed the page and whether results will be split by them. Ask who owns keeping the page in step with the creative when the campaign changes, and what happens to the test if an offer changes mid-run. Ask what the primary metric is and why, and whether bounce rate is being used as a stand-in for something more useful. Ask to see the advert and the variant side by side on a phone before launch. And if the agency proposes a single “optimised” landing page for every campaign, ask how it will keep the promise of each one, and which campaign it was written for.

Put it to work

Pick your three highest-spend campaigns. For each one, put the advert or email next to the page it lands on and write down the promise in the ad, and the first thing the page says. Mark each pair as a match or a mismatch.

Your answer

Show the worked example

There is no right answer. Write yours first, then compare.

Sources

1. Nielsen Norman Group: Information Scent, How Users Decide Where to Go Next
2. Google Ads Help: About Quality Score (landing page experience)
3. Digital Markets, Competition and Consumers Act 2024, Schedule 20 (banned practices)