Flight log / Transmission 042
Landing page copy that converts paid clicks
Most of the words on your landing page will never be read, and that's fine if they're the right ones. Jakob Nielsen's 2008 analysis estimated visitors read about 20% of the text on an average web page.
> Landing page copy that converts paid clicks repeats the searcher's words in the headline, backs up every claim in the ad, names one outcome and one next step, and stays plain. Unbounce's 2024 benchmark found pages written at a 5th to 7th grade level converted at 11.1%, versus 5.3% for professional-level copy.
Paid traffic changes the copywriting brief. An organic visitor might be browsing. A paid search visitor typed a specific problem into a search box a few seconds ago, read your ad and clicked because the ad seemed to answer it. The page doesn't need to introduce your company. It needs to finish the sentence the ad started.
This post covers how much visitors actually read, what reading level tends to convert, how to write the headline, the order of the rest of the page, a worked rewrite and a manual process you can run on your top pages this week.
// 01 · the briefingWhat makes paid landing page copy different?
Three things: the visitor arrives mid-thought, they arrive with words already in their head, and you paid for them to be there.
Mid-thought means the copy should skip the warm-up. No "Welcome to," no company history, no mission statement in the first screen. Words already in their head means the strongest copy often borrows the searcher's own vocabulary rather than your internal category name. And the cost of the click means every sentence that makes a qualified visitor hesitate is spend you didn't get back.
Google's own guidance for advertisers points the same way. It asks for a landing page that "closely matches your ad and keywords," that will "mirror the call-to-action in your ad text," and that is "clear about your product or service and what it does." That's a copywriting brief in three lines. The message match guide covers the handoff from ad to page in more depth.
// 02 · telemetryHow much landing page copy do visitors actually read?
Less than you wrote. In 2008, Jakob Nielsen analyzed browsing data from a study by Weinreich and colleagues: 25 users whose browsers were instrumented in 2005, with 45,237 page views left after filtering. His conclusion was that users had time to read at most 28% of the words during an average visit, and that about 20% was more realistic.
The more useful finding for copywriters is the marginal one. The best-fit formula showed people spent only 4.4 seconds more for each additional 100 words on a page. At a typical reading speed, that's about 18 of those 100 words. Add a paragraph and most of it goes unread.
The data is old and the sample was small and above-average in education, which Nielsen acknowledged. But nothing since has suggested people started reading web pages more carefully. The practical takeaway holds: you're not writing a document to be read, you're writing a page to be skimmed, and you get to choose which 20% gets seen. Put the important word first in every headline, line and button.
Every sentence on the page is competing with the back button. Most of them lose quietly. Write the ones that need to win first.
What reading level should landing page copy be?
Plainer than feels comfortable. Unbounce's 2024 Conversion Benchmark Report covered more than 41,000 landing pages and 464 million unique visitors, and compared conversion rates by the reading level of the copy.
Source: Unbounce, 2024 Conversion Benchmark Report, 2024
View data
| Conversion rate | |
|---|---|
| 5th to 7th grade | 11.1% |
| 8th to 9th grade | 7.1% |
| Professional | 5.3% |
Pages written at a 5th to 7th grade level converted at 11.1%. Pages at an 8th to 9th grade level converted at 7.1%, and professional-level copy at 5.3%, less than half the rate of the plainest group. The same report found that difficult words, reading time and word count all correlated negatively with conversion rate, with difficult words showing the strongest link.
Read that carefully. It's a correlation across many industries, not a controlled test. Pages written plainly may also tend to have simpler offers, and simpler offers convert better. But the direction lines up with the reading data in the previous section: if visitors skim, short words and short sentences survive the skim better than long ones.
In B2B, "plain" doesn't mean dumbed down. A security engineer searching "SOC 2 Type II evidence collection" uses those terms because they're precise. Keep the searcher's vocabulary. Cut your own: the "leverage," "seamless," "end-to-end" and "solutions" that nobody typed into a search box.
// 04 · payloadHow do you write a landing page headline for paid search?
Start from evidence, not inspiration. Open the search terms report for the ad group and the headlines that are actually serving. Then write a headline that does two things: uses the searcher's core phrase or a clear synonym, and states the outcome or offer the ad promised.
The illustrative chart below shows how that logic tends to play out. It's a hypothetical test on a B2B scheduling page, with five headline styles for the same search.
In this hypothetical, "Build a week of shifts in 10 minutes: shift scheduling for hourly teams" (query words plus outcome) beats either half on its own, and both beat the category name ("Workforce management platform") and the brand tagline. The pattern is the point, not the numbers: the headline that confirms the search and promises a result gives the visitor two reasons to stay.
Responsive search ads add a wrinkle. Google lets you supply up to 15 headlines of up to 30 characters each, shows up to 3 at a time, and assembles the combinations itself. Any headline in the ad could be the one this visitor saw, so every claim in every headline needs to be true and visible on the page. If the page can't back one up, change or pin the ad headline. See responsive search ads best practices for how pinning works.
// 05 · mission logWhat should the rest of the page say, and in what order?
A paid landing page answers the visitor's questions in roughly the order they ask them. The first three rows belong on the first screen, which our above the fold guide covers in detail.
| Section | The visitor's question | Common miss |
|---|---|---|
| Headline | Am I in the right place? | Brand tagline instead of the search |
| Subhead | What exactly do I get, and is it for me? | Restates the headline in longer words |
| Button | What do I do next? | "Submit" or "Learn more" instead of the action |
| Proof | Does this work for people like me? | Vague praise with no name, role or number |
| Benefits | What changes for me? | Feature list with no outcome attached |
| Objections | What's the catch? | Price, setup time and contract terms left unanswered |
| Closing ask | OK, now what? | A different action than the top button |
Two rules make every section work harder. Start each block with its point, because skimmers read the first words of a line and move on. And name the button for what happens after the click: "Get the checklist," "Start my free trial," "See pricing." A button that says "Submit" asks the visitor to do your job of explaining the action.
If the ad says "free trial" and the button says "Contact sales," the page just renegotiated the deal without telling anyone.
What does a landing page copy rewrite look like?
A hypothetical example. Say a compliance consultancy bids on "SOC 2 readiness checklist" and runs an ad offering a free download.
| Before | After | |
|---|---|---|
| Headline | "End-to-end compliance solutions for modern enterprises" | "Your free SOC 2 readiness checklist" |
| Subhead | "We leverage deep expertise to streamline your journey" | "62 controls, grouped by owner, so you know what to fix before the auditor arrives." |
| Button | "Submit" | "Send me the checklist" |
| Proof | "Trusted by industry leaders" | One quote from a named head of security at a company the visitor would recognize as a peer |
| Form | Name, email, company, size, budget, timeline | Work email and company |
Nothing about the offer changed. The "after" version uses the searcher's words, puts the specific outcome in the subhead, names the action on the button and stops asking for information the visitor wasn't ready to give. Cutting the form to two fields matters as much as any headline.
// 07 · pre-flight checklistHow do you rewrite landing page copy by hand?
Copy rewrites are wax-on, wax-off work: dull repetitions that turn out to be the whole skill. Here's the process for one page.
- Collect the visitor's words. Export the top search terms for the ad groups that point at the page, plus every serving ad headline and description.
- Highlight the promises. Mark every specific claim in the ads: price, trial length, outcome, audience, offer. Each one must appear on the page.
- Write the headline and subhead first. Use the search vocabulary and the ad's promise. Write five versions and pick the plainest.
- Cut by a third. Delete adjectives, warm-up sentences and anything that describes your company rather than the visitor's outcome.
- Check the reading level. Most word processors can show a readability score. Aim plainer than your current page, without dropping terms your buyer actually uses.
- Read it aloud. Anything you wouldn't say to a prospect on a call doesn't belong on the page.
- Test it against the old version. Then wait for a real result.
That last step matters because copy changes often look better in week one than they are. The illustrative chart below shows a hypothetical account where a rewrite went live in week 6.
Illustrative data, not from a real account.
View data
| Conversion rate | |
|---|---|
| W1 | 3.9% |
| W2 | 4.1% |
| W3 | 3.8% |
| W4 | 4.0% |
| W5 | 4.0% |
| W6 | 4.6% |
| W7 | 4.8% |
| W8 | 4.7% |
| W9 | 4.9% |
| W10 | 4.8% |
In this illustrative example, conversion rate moves from around 4.0% to around 4.8% and holds for five weeks. Week-to-week noise of a few tenths of a point is normal, which is why a before and after comparison is weaker evidence than a split test run at the same time. Our guide on landing page testing covers how to set that up.
// 08 · where this breaksWhere copy rules break down
- Plain isn't always short. An expensive or risky purchase needs detail. Long copy can win when the decision is big, as long as each section leads with its point.
- Jargon is relative. The words your buyer searched for are plain language to them, even if they look technical to you. Cut your jargon, keep theirs.
- Benchmarks are averages. The Unbounce reading-level numbers span many industries and offers. Your page's best reading level is something you find by testing, not by looking up.
- Copy can't fix the offer. If the price, product or ask is wrong for the audience, better words just get you to "no" faster.
- Low volume limits proof. A page with a few hundred clicks a month may take a long time to show a clear winner. Judgment and the checklist above carry more weight there.
Something's warming up on the launch pad.
We're building something for people who read posts like this. Early access opens in waves.
Early access, by invitation.
FAQ
What makes good landing page copy for paid search?
It repeats the searcher's words in the headline, backs up every claim the ad made, states one specific outcome, names one next step and stays plain enough to understand while skimming.
What reading level should landing page copy be?
Plainer than most teams write. Unbounce's 2024 Conversion Benchmark Report found pages written at a 5th to 7th grade level converted at 11.1%, compared with 7.1% at 8th to 9th grade and 5.3% for professional-level copy. That is a correlation, not proof of cause.
How much landing page copy do visitors read?
Not much. Jakob Nielsen's 2008 analysis of browsing data estimated people read about 20% of the words on an average web page, and at most 28%.
Should landing page copy be long or short?
As short as the decision allows. A free checklist needs a few lines. An expensive, high-risk purchase needs more detail, but it should still be skimmable, with the key point at the start of every section.
How do you write a landing page headline for Google Ads?
Start from the search terms and serving ad headlines, use the searcher's core phrase or a clear synonym, and add the outcome or offer the ad promised. Avoid brand taglines and internal category names.