Landing Page
by
Landing page assignment for ecommerce teams: rank pages on revenue per session rather than bounce rate, then reassign each campaign to the page that actually earns for its traffic.
Landing pages get judged on bounce rate and time on page, which measure whether someone stayed rather than whether they bought. A collection page with a high bounce rate can out-earn a long-form page that holds attention beautifully and converts nobody. The assignment is usually historical too: a campaign points at whatever page it pointed at when it launched, and nobody revisits it. This skill measures revenue per session and margin per session by page and by traffic source, finds where a campaign's traffic converts best, and returns the reassignments worth making.
What It Does
- Measures revenue per session and margin per session for every landing page.
- Breaks the result down by traffic source, because pages perform differently per channel.
- Compares each campaign's current page against the alternatives its traffic sees.
- Separates a page problem from a traffic quality problem.
- Flags pages carrying spend with no path to a conversion.
- Returns a set of campaign reassignments, each with the expected gain.
Use Cases
- Stop judging a collection page against a long-form page on bounce rate.
- Move a campaign off the page it has pointed at since launch.
- Find the page that works for paid search and not for paid social.
- Rank page rebuilds by the revenue sitting behind each one.
Example Session
Which campaigns are pointing at the wrong landing page?
Three reassignments are worth making. One page is doing better than its bounce rate suggests.
- Page | Sessions | Bounce | Rev per session | Margin per session | Action
- Collection, outerwear | 18,400 | 71% | $4.10 | $1.55 | Send paid social here
- Long-form, technology | 6,200 | 34% | $1.90 | $0.62 | Keep for paid search only
- Homepage | 22,100 | 48% | $2.40 | $0.88 | Move two campaigns off
- Quiz, fit finder | 3,100 | 29% | $5.60 | $2.10 | Scale, highest per session
The outerwear collection has the worst bounce rate on the list and earns more than twice per session what the long-form page does, because its traffic arrives ready to buy and leaves once it has. The quiz earns most per session but only runs 3,100 of them: the gain is real and small, so send more traffic there before rebuilding anything else.
Why This Beats Prompting It Yourself
Bounce Rate vs Revenue Per Session: The Hidden Risk
A generic prompt ranks landing pages on the engagement metrics an analytics tool puts first. Bounce rate and time on page both reward a page that holds someone's attention, which is not the same as a page that sells. The recommendation that follows moves traffic away from the page earning most per session, and the numbers supporting it all look reasonable.
When You Still Need The Margin Behind The Session
Traffic is worth what it earns, and pages differ by channel in ways an average hides. This skill states $4.10 of revenue and $1.55 of margin per session on a page bouncing at 71%, next to $0.62 on one bouncing at 34%. Each reassignment carries the expected gain, so the work is ordered by what it is worth rather than by which page looks worst.
Known Limitations
Revenue per session needs session data joined to orders, so heavy cross-device journeys will under-attribute pages that start a purchase someone finishes elsewhere. Margin per session depends on product cost being available at order line level, and without it the ranking runs on revenue and labels the gap. It recommends reassignments and does not change campaign destinations in the ad accounts.
Install It
- Download the ZIP.
It is free and there is no account to create. - Unzip it into your agent's skills folder.
Claude Code reads ~/.claude/skills/, which is hidden by default: the command in the folder block opens it. Other agents scan their own directory, so drop the same folder there instead. - Ask your agent to use it.
Restart the agent if it was already running, then it picks the skill up with no config.
skills folder
Copy
~/.claude/skills/landing-page/
SKILL.md
references/session-attribution.md
reports/
# one report per run, with the reassignments
# macOS: create the folder and open it in Finder
mkdir -p ~/.claude/skills && open ~/.claude/skills
# Windows: paste in the Explorer address bar
%USERPROFILE%\.claude\skills\Frequently asked questions
How is this different from a landing page report in analytics?
Analytics ranks pages on engagement because that is what it measures reliably. This ranks them on revenue and margin per session, split by traffic source, and compares each campaign's current page against the alternatives. The output is a set of reassignments, not a table.
Which agents does the skill run in?
Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision.
What does it need connected to work?
Session data with source attribution, orders, and product cost. With all three it reports margin per session. With sessions and orders only it reports revenue per session and says the margin view is missing, which still catches most misassignments.
Why does it split everything by traffic source?
Because the same page performs very differently depending on intent. A page that converts paid search traffic at a high rate can be the wrong destination for paid social, where people arrive earlier in their thinking. One blended average hides both facts.
Does it change where my campaigns point?
No. It produces the reassignments with the expected gain against each one, and a person applies them in the ad account. A wrong destination applied at scale is expensive to unwind, and the list is short enough to review in a few minutes.
What is included with the skill?
The skill itself, the session attribution reference, the report template, and the per-source benchmarks. It is free and security scanned. We re-publish the ZIP when attribution handling changes, so download it again if your session counts stop matching analytics.
A landing page wins on revenue per session, not on bounce rate.
Polar ties sessions to orders and margin across 45+ sources, so reassigning campaigns follows the page that earns.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from a landing page report in analytics?","acceptedAnswer":{"@type":"Answer","text":"Analytics ranks pages on engagement because that is what it measures reliably. This ranks them on revenue and margin per session, split by traffic source, and compares each campaign's current page against the alternatives. The output is a set of reassignments, not a table."}},{"@type":"Question","name":"Which agents does the skill run in?","acceptedAnswer":{"@type":"Answer","text":"Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision."}},{"@type":"Question","name":"What does it need connected to work?","acceptedAnswer":{"@type":"Answer","text":"Session data with source attribution, orders, and product cost. With all three it reports margin per session. With sessions and orders only it reports revenue per session and says the margin view is missing, which still catches most misassignments."}},{"@type":"Question","name":"Why does it split everything by traffic source?","acceptedAnswer":{"@type":"Answer","text":"Because the same page performs very differently depending on intent. A page that converts paid search traffic at a high rate can be the wrong destination for paid social, where people arrive earlier in their thinking. One blended average hides both facts."}},{"@type":"Question","name":"Does it change where my campaigns point?","acceptedAnswer":{"@type":"Answer","text":"No. It produces the reassignments with the expected gain against each one, and a person applies them in the ad account. A wrong destination applied at scale is expensive to unwind, and the list is short enough to review in a few minutes."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the session attribution reference, the report template, and the per-source benchmarks. It is free and security scanned. We re-publish the ZIP when attribution handling changes, so download it again if your session counts stop matching analytics."}}]}</script>