Landing page conversion rate: how to measure it and what moves it
You want more signups, trials or leads from the pages people land on. This guide shows how to measure landing page conversion rate per page, what really moves it, and a simple routine to improve it without guesswork.
By Théophile Louvart, founder of Porteur · Updated 14 September 2026 · Markdown
Why averages are not your target
Ignore published averages. They mix industries, traffic sources, definitions and timeframes. They do not tell you what your visitors do on your pages.
The only rate that matters is yours. Measure it per landing page over a period long enough to gather a few hundred sessions. Then decide where to work.
What “landing page conversion rate” means
Use clear definitions so you do not chase the wrong number. Here are the ones you need for product sites as of 2026.
- Landing page conversion rate: conversions divided by sessions that began on the page.
- Signup conversion rate: accounts created divided by visitors of the signup path.
- Trial to paid rate: paying accounts divided by trials started in the same cohort.
Free trial and freemium are different models. A free trial is time limited with the full product. Freemium is a free tier forever with limits. Expect different rates and support costs for each.
How to measure your rate per landing page
You want sessions that begin on each page and the conversions from those sessions. Do this in your analytics and, where possible, in product analytics that hold the signup event.
Pick the conversion event
Choose one conversion per page type. For /pricing, use signup or trial started. For a guide, use the next step you offer, such as “start free trial” or “see the demo”.
Build or open a landing page report
In GA4, use the Landing page dimension with Sessions and your chosen conversion event. Filter by source if you want only search or only ads.
Use sessions that began on the page
Make sure the metric is sessions where that page was the first page. Do not use all pageviews. The rate is conversions divided by these landing sessions.
Set a minimum sample
Wait until each page has a few hundred landing sessions. If traffic is low, extend the window to 28 days or more. Avoid reading noise from tiny numbers.
Join to product analytics
Where you can, join the first page in the session to the signup event in product analytics. This tells you which landing pages bring signups and which bring readers who never try.
A working view looks like this: a table with /pricing, /, /guides/getting-started, each with Landing sessions, Signups, and a rate column you add. Sort by sessions to focus on impact first, then by rate to find wins you can copy.
Minimum sample, timeframe and low-traffic pages
Make decisions with enough data. A few hundred landing sessions per page is a sensible floor. Below that, one or two signups can swing the rate wildly.
- Low traffic: extend the window to 28 days or more.
- Group similar pages: combine variants, such as /templates/invoice, /templates/quote, to see a pattern before drilling down.
- Directional checks: track micro-steps like clicks on the primary button when you cannot wait for final conversions.
- Control for source: separate organic, paid and referrals when rates differ by intent.
The five levers that move landing page conversion
You control five things that move the rate: intent match, the first screen, the call to action, friction and trust. Fix them one by one, with a clear example on a real page on your site.
Intent match: answer the reason they came
Search traffic does not convert for three usual reasons. The query is informational and the page sells. The next step is missing or hidden. The visitor is not the buyer. Product analytics that join the landing page to the signup event is how you tell them apart.
Map each page to a query or segment
Write the main query or audience above the page. For /guides/getting-started, it might be “how to set up yourproduct” or support readers from email.
Serve the intent first
If the query is “invoice template”, a page that opens with pricing will fail. Show the template, let them download or copy it, then offer the product that automates it.
Add the next step where the answer ends
Place one call to action at the end of the answer. Phrase it as the next step, such as “create your first invoice in yourproduct.com” rather than “sign up”.
Worked example: yourproduct.com/blog/invoice-template ranked for “invoice template”. Add a live preview and a “Use this in yourproduct.com” button right below the preview. Keep the article above the fold for readers who came to learn.
The first screen: clarity and speed
The top of the page must say who the page is for and what happens next. It must also load fast and stay stable while loading. That first impression sets the tone of the session.
- State the outcome for a clear audience in one line. Example: “Time tracking for solo consultants, with invoices in one click”.
- Show the primary action once. No carousel, no rotating headlines, no autoplay video up top.
- Make the largest element load within about 2.5 seconds for LCP, keep layout shift around 0.1 for CLS, and keep interactions near 200 ms for INP.
- Use a simple hero image or a screenshot instead of heavy animation. Avoid layout shifts from late-loading fonts or banners.
Worked example: /pricing had a vague header and a giant SVG animation. Replace with “Plans for freelancers and small teams” and a static shot of the billing page. Run PageSpeed Insights and fix any layout shift from banners or cookie notices before the plans table.
The call to action: one next step, in the reader’s words
Content pages convert when they carry one clear next step. The call to action sits where the reader has just got the answer. It is phrased as the next step in their terms. Avoid modals that interrupt reading. Google penalises interstitials that cover content on mobile, and your rate will suffer too.
- Use one primary button per page. Secondary links can sit in the footer.
- Place it right after the section that answers the query. On /guides/compress-images, put “Compress your images in yourproduct.com” below the compression steps.
- Use labels like “Check your site”, “See the report for your domain”, “Start a 14 day trial” if that is your model. Avoid “Submit”.
- Keep the button visible without scrolling on short pages like /pricing. Do not hide it under tabs or accordions.
Worked example: /guides/getting-started ended with a summary. Add a single button, “Create your account to follow this guide”, right after the last step. Remove the floating bar that blocked headings on mobile.
Friction: remove each block between interest and first use
Signup friction kills momentum. Each extra field, a card before the product is seen, an email verification before the first screen, a password policy that rejects real passwords, no single sign on, and vague errors will lower conversion. Fix them first on the pages that bring the most sessions to signup.
Cut fields to the minimum
On /signup, ask for email and password. Ask for name and company later, when trust is earned. Defer any survey until after first value.
Move the card wall
If you require a card for a free trial, show the product first. Ask for payment details at a natural point, such as when a user reaches a limit or invites a team.
Fix verification and password pain
Allow login with Google, Apple or GitHub. If you must verify email, let the user proceed to the first screen and block only actions that send email until verified. Accept strong passphrases without fussy character rules.
Write helpful errors
Tell the user what to fix. “Password is too short, use 10+ characters” is better than “Invalid password”. Keep errors close to the field that failed.
Worked example: /signup asked for six fields and blocked the first screen on email verification. Cut to two fields, allowed SSO, and let unverified users reach the dashboard with a banner to verify later. Expect higher completion and a faster first action, with no need for a new design.
Trust: show real signals a small site can earn
Visitors decide quickly if they should trust you. Show real signals you can stand behind. Add them to /pricing and to any high-traffic landing page that asks for signup or a trial.
- An About page with real names and roles.
- A postal address and a company registration.
- A security page that says what is encrypted, where data lives and how to delete it.
- A status page and a changelog with dates.
- Screenshots of the real product, not mockups.
- Customer logos only with permission, and quotes with a name and a company.
- A visible support email address.
- Terms and privacy in plain language.
- Links to reviews on third-party sites such as G2, Capterra or Product Hunt.
No customers yet? Show the founder’s own use, real numbers you have, a public roadmap, the open source parts, and a comparison that is honest about what you lack. Add a short trust section to /pricing above the fold on mobile. Link to security, status and reviews below the plans table.
Pricing pages that convert: what to show and where
Pricing is a common landing page. There is strong guidance on what converts. Keep the structure simple and consistent with your app, docs and emails.
- Two to four plans, with one recommended.
- Each plan explained in one line: who it is for and the limit that matters.
- Price visible without a click. No “contact us” for core plans.
- Annual and monthly as a toggle, with the saving stated.
- An FAQ that answers real objections support hears: cancellation, card, limits, what happens at the end of a trial.
- Use the same plan names everywhere: app, docs, emails and /pricing.
Worked example: /pricing hides annual prices behind a tooltip, calls the plan “Starter” on the page but “Basic” in the app, and has no FAQ. Fix the names to match, put the price in the cell, add a toggle with “save two months”, and write four FAQ entries from your support inbox. Expect fewer confused chats and a clearer choice.
Speed and stability: check them because they change behaviour
Speed is a conversion problem as much as an SEO problem. Slow pages leak visitors on mobile and make your calls to action feel heavy. Check and fix Core Web Vitals on the landing pages that matter most.
- Largest Contentful Paint: aim for about 2.5 seconds or less.
- Cumulative Layout Shift: around 0.1 or less so buttons do not jump.
- Interaction to Next Paint: near 200 ms so clicks feel instant.
- Test with PageSpeed Insights and your own devices on a 4G connection.
- Tackle images, web fonts and third-party scripts first on / and /pricing.
Worked example: / loaded a hero video from a third-party player. Replace with a compressed poster image and a click-to-play inline video. Preload the main font. Remove a chat widget on mobile. LCP drops, CLS settles, and the primary button becomes tappable faster, which helps your rate and your rankings.
A monthly routine to move one page at a time
Stop random tweaks. Run a simple routine once a month. Pick one page, one lever, one change. Measure against a clean baseline and write down the result.
Pick the page
Sort your landing page report by landing sessions. Choose the top page with a rate you are not happy with, such as /pricing or a high-traffic guide.
Pick the lever
Choose one: intent match, first screen, call to action, friction or trust. Decide based on what is off on the page now.
Write the hypothesis
Example: “Adding a single next-step CTA below the answer on /guides/image-compression will raise signups from search sessions.”
Baseline for 28 days
Record landing sessions, conversions and the rate for the last 28 days. Screenshot the page state. Note any traffic shifts, launches or outages.
Ship one change
Make only the planned change. Avoid shipping multiple changes together so you can attribute the result.
Measure for 28 days
Compare the same metrics for the next 28 days. If traffic is very low, run longer. If it is seasonal, compare to the same period last year.
Log and repeat
Keep a simple log per page: date, change, lever, before and after. Roll winners to similar pages. Revert changes that did not help.
A fixed page is obvious when you read it aloud. It says who it is for, what to do next, loads quickly and lets a new user in without fuss. If your notes show movement, do the same change on the next page in the list.
Diagnose non-converting search traffic fast
Do this when a page brings search traffic but few signups. It will save you weeks of fiddling with button colours.
- Pull sessions that began on the page from organic search only.
- Join landing page to signup in product analytics. If you cannot, track button clicks as a proxy.
- Check the query and the headline. If intent is informational, answer it first, then offer the next step.
- Move or add one call to action where the answer ends. Make the label a next step, not “sign up”.
- If the visitor is not the buyer, accept it. Students and competitors read and leave. Do not contort the page for them.
Worked example: /blog/utm-parameters ranks for “what are UTM parameters”. Add a short definition, a copyable example and a button, “Build UTMs for your campaign”. Keep the tutorial intact above that button so the page serves the query first. This is in line with Google’s page experience and helpful content guidance.
Questions
There is no universal good rate. Sources use different industries and definitions. Measure your own rate per landing page over a few hundred landing sessions, then improve it page by page.
Published averages vary by source and by what they count. Do not set targets from them. Use your own data by landing page, channel and model, then decide what a good result is for your business.
It can be great or poor depending on the page, the traffic and your model. Look at sessions that began on the page and the conversions from those sessions. Compare to your other pages, not to someone else’s chart.
Use the Landing page dimension with Sessions and your chosen conversion event. Filter by source if needed. Compute conversions divided by sessions that began on each page. Set a minimum sample of a few hundred sessions before you judge a change.
Use at least 28 days and a few hundred landing sessions per variant. Low traffic needs longer. If traffic is very low, use before and after tests and track micro-steps such as button clicks until you gather enough final conversions.
Not to start. Ship one meaningful change per month and measure before and after. When you have the traffic and a testing habit, add an A/B tool for parallel ideas on your highest impact pages.
Sources
Check my site, free
Paste your URL to see three findings from your site, the searches around it and rivals in about thirty seconds, then decide which landing page to fix first.
- Free check, no card
- Read-only, your own accounts
- Readable by your agent
Read next
- GuideWebsite conversion rate: from the visit to the signup
- GuideThe GA4 landing page report: the pages that bring people, and what they do
- GuideLanding page SEO: one page, one search, one promise
- GuidePages that get search clicks and no signups: the three causes
- GuideCalls to action on guides and blog posts: one, in the right place
- GuideCore Web Vitals for a product site: the three numbers
- GuideWriting a landing page that ranks and converts
- AlternativesSurfer SEO alternatives: content optimisation tools and what they measure