Use this when
- Paid or partner traffic arrives with a promise the homepage does not keep in the first screen.
- A webinar, launch, or comparison campaign needs a URL that will die when the campaign dies.
- Capture ads land on
/and bounce.
Do not use this when
- The promise is the standing story. That is the homepage.
- You need the comparison or the price. Send them there; do not clone those pages under a UTM.
- The motion is unnamed and you are making pages to feel busy. Channel strategy first.
- You want to A/B button color before the promise matches the ad.
A few useful terms
Keep this in mind
The first screen must keep the promise that bought the click. If the ad said “{job} for {seat} vs {alternative},” the hero cannot say “welcome to the platform.” Message match is the page. Everything else is decoration.How to do it
Step 1: Define the promise and next step
Complete: people who saw {source} were promised {sentence}; this URL keeps it by {proof or explanation}; they do {one action}; we unpublish or redirect on {date}. If you cannot finish that, you do not need a landing page.Step 2: Keep the product story consistent
Category, seat, alternative, result—same as positioning. The landing page narrows (this campaign’s trigger or offer). It does not reposition. Navigation can shrink; the constraint link should still exist if the promise touches security or implementation.Step 3: Match the action to visitor intent
Two primary buttons mean two campaigns got taped together.
Step 4: Keep the form proportionate to the request
Awareness register: email. Hand-raise: the demo request rules (including HDYHAU if this is the hand-raise). Do not put a 12-field gate on a creation click and call the fills pipeline.Step 5: Review or retire the page after the campaign
When the campaign ends, redirect to the standing page that kept the same promise (homepage, comparison, or content URL). Orphan landing pages with old prices are how trust dies in search.Worked example (illustrative)
Paid search on “shared inbox vs queue for ops.” Capture.Copy: landing-page brief (fill)
- Source (ad / partner / email) and the exact promise:
- How this URL keeps it (one proof or explanation):
- Door (one):
- Fields, if any—and why:
- Standing page we redirect to on kill date:
- Kill date:
- What we will not add (second CTA, pop-up, extra SKU):
Before you start
- Promise on the page matches the source in the first screen.
- Positioning is not rewritten.
- One primary door.
- Form length matches intent.
- Kill date and redirect target are written.
- This is not a clone of
/with a UTM. - Creation traffic is not forced through a capture form.
Metrics
Bounce rate is a hint that the promise broke. It is not the scoreboard. Button-color wins are not this playbook; tests that change a decision sit in experimentation.
Common mistakes
- Sending every ad to
/. - A new value prop that sales has never said.
- Gating a creation click.
- Letting the page live a year after the campaign.
- Calling this CRO and testing headlines before the promise matches.
What to read next
The standing scan is the homepage. The durable URLs a campaign should eventually earn are content strategy. Whether you should have paid for the click is paid media. How to interpret the visit alongside evidence beyond its last recorded click is measurement model. A named customer’s proof, when you have approval, is a case study; use only approved, accurate customer material on the landing page.Sources and evidence boundary
This is an owner-maintained operating synthesis. Message match (ad promise = page hero) is standard paid-capture hygiene, not a Unbounce recipe. One door and kill dates are this library’s judgment, paired with homepage and paid media. Consumer CRO lists (urgency clocks, fake social proof) are refused.Copyright © 2026 Ivan Xu. All rights reserved. See the copyright and reuse terms. Canonical source: github.com/weilun88313/B2B-Playbook