Leadpages With Causality Engine: Check the 7-Day Offsite Conversions

The short answer: when a Leadpages page reports more store revenue than the campaign behind it seems to justify, check the window before the campaign. Leadpages, the AI landing page builder with testing and Smart Traffic, can count purchases on an external site such as your Shopify store, and credits the page with any that happen within an attribution window that defaults to 7 days after the click, per its Analytics and Insights article. The same orders are claimed again by the ads that sent the visit. GA4 holds a third version, and Causality Engine, a causal attribution tool that reads GA4's export, turns it into a per-channel answer, beginning with a free first finding at app.causalityengine.ai/start.

Disclosure: I founded Causality Engine and own it, and this page recommends it. Every claim about it links to Causality Engine's own pages.

Three things to know before the page's revenue goes in a report:

• The window does most of the work. For example, a subscriber who clicks the page on a Monday and buys from an email on Saturday is inside 7 days: Leadpages counts the order, and so does the email tool. One sale, two claims.
• Your page already captures the source. Leadpages breaks traffic down by UTM campaign, source and medium (analytics), and forms can add UTM, gclid and fbclid values to every submission with Capture UTMs (manual mode guide).
• GA4 needs the hand-off to hold. If the page's domain shows up as a referrer at the store, GA4 has stopped crediting the ad.

The rule behind a Leadpages conversion

The Analytics and Insights article defines three kinds of result, and one setting that stretches the third:

• On the page. A conversion is a visitor who submitted a form, clicked a tracked element or completed a purchase.
• Through Leadpages checkout. With Stripe, PayPal, Shopify, Lemon Squeezy or Gumroad connected, Leadpages records revenue, purchases, average order value and revenue per visitor (checkout article).
• Offsite. Under Dashboard, Conversions, Add Destination tracks a conversion on another site. You set the attribution window, how long after clicking your page a conversion still counts, with 7 days as the default, and the report shows counts, revenue and the pages behind them.

The articles read on 11 November 2026 do not say how a destination detects the purchase, or how one order is handled when a buyer clicked two pages. Ask support before the figure goes in a board report.

The facts in one table

From page to store without losing the ad

Leadpages' dashboard has a Scripts & Pixels area for analytics code (dashboard overview). Put the same Google tag there as on the store. Then follow Google's cross-domain measurement steps: both domains in Configure your domains, under one web data stream, so the visitor keeps one ID as it crosses in the _gl parameter. Without that, GA4 counts a new user and a new session at the store.

An older Leadpages article, written for the classic builder, names a second leak: pages on your own domain are also published on a Leadpages subdomain, and GA4 can register both. Its advice is to exclude the Leadpages domain as a referral (third-party tracking). In today's GA4 that list is called unwanted referrals, with room for 50 per stream.

Do not retag the page's buttons with new UTMs. A fresh source on the outbound link names the page, not the ad, in GA4. The channel should be set once, on the ad: Google's default channel group puts a paid medium from a search engine under Paid Search and medium email under Email.

Worked example: three weeks, four claims

For example, with numbers chosen for illustration and no store behind them: a skincare launch page runs for 21 days, fed by Google Ads, Meta and two emails. The store books 540 orders in that time.

The claims exceed the orders by 673 - 540 = 133. The Leadpages figure overlaps every other row, because its rule credits any purchase within a week of a page click, whatever brought the buyer back.

GA4, with cross-domain set up only in week two, files the 540 as: Paid Search 117, Paid Social 132, Email 61, Organic Search 88, Direct 104, Referral 38. The 38 under Referral carry the page's own domain, sent before the domains were joined.

The read takes that file. Its how-it-works page says it splits Direct's 104 back to the channels that sent those buyers, sets each channel's caused share beside last-click, scores data health from 0 to 100 and names a next step. The Referral 38 stays Referral: the read weighs channels as recorded. Where its Paid Search share and Google's 158 point in different directions, test the change in one region before the account.

Check it free, first

Both checks are by hand, and either is enough on its own.

Shrink the window. Note the offsite conversion count at 7 days, set the window to 1 day, and compare the same dates. A shorter window can only count the same orders or fewer. For example, if 214 falls to 120, then 214 - 120 = 94 of the page's orders came two to seven days after the click, a stretch in which emails, retargeting and searches all had their chance.

Find the page in Referral. In GA4, filter Referral by session source for your page's domains, including the Leadpages one. Purchases there are orders that lost their ad's credit. Fix the hand-off before anything else reads the data.

The A/B testing article supports an Offsite conversion goal and a 50/50 default split, but it decides between page variants, not channels.

Where the pairing does not reach

• A purchase key event in GA4 is required. The one-off read page takes key events from the Advertising tab, and the how-it-works page wants 12 months of export or more.
• Pages are below its line of sight. The read reports Paid Search, never "launch page, variant B". Smart Traffic and heatmaps stay Leadpages' answers.
• Small spend is turned away. The homepage says that when spend is too small to find a signal, the model declines.

Leadpages ships the window at 7 days and adds every purchase inside it to the page's revenue column, the figure on the dashboard you open when deciding whether to keep paying for the plan; a wider default makes that column larger with no change in what the page did.

When both checks pass, export the Attribution paths CSV and drop it on the start page, where the first finding is computed in your browser for free. Causality Engine, which I founded, sells the full read for €99 once, refundable within 30 days, and Pro for €299 a month (pricing, checked on 11 November 2026).

See also

• Leadpages: the rebuilt builder and its testing tiers
• How to use Leadpages: brand kit to first-week analytics
• Leadpages on the marketplace: Leadpages' affiliate program terms
• Causality Engine: what a read shows and costs
• How to use Causality Engine: the export, the read, the test
• Causality Engine on the marketplace: its affiliate program, in the owner's words
• Leadpages vs Unbounce: no traffic cap or visitor caps
• How to find what Direct traffic is hiding: where untagged buyers end up

FAQ

Can Leadpages send its conversions to Causality Engine?

No. The two products share nothing. Causality Engine reads the channels GA4 has recorded (integrations page), so Leadpages appears only as the visits and purchases GA4 logged at the store.

What attribution window should I use for offsite conversions?

My recommendation: the shortest one your buyers' path allows, written next to every figure you report. The default is 7 days (Analytics and Insights).

Why does Leadpages show fewer visitors than my ads show clicks?

Consent and privacy signals. Leadpages collects nothing for a visitor sending Global Privacy Control or Do Not Track, and waits for analytics consent where a consent tool is installed (Analytics and Insights).