Plerdy With Causality Engine: Same-Session Revenue by Channel

The short answer: Plerdy says paid social earns a fraction of what Ads Manager claims, and Direct looks like the store's strongest channel. Both readings follow from one documented rule. Plerdy, a heatmap, session replay and ecommerce analytics tool that sits on the site, counts a purchase only in the session it happens in and gives that session to the channel it arrived from, while Google Ads credits the ad for up to 30 days after the click by default, per Plerdy's own wasted ad spend guide. That shows which pages turn visits into money. Which channel caused the buyer to come is what Causality Engine reads from a GA4 export, and its first finding is free 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 checks before a channel row moves a budget:

• Rebuild the row from its parts. Plerdy's Ecommerce Analytics defines revenue per session as AOV times eCR. For example, a channel with an AOV of €72 and an eCR of 1.5% earns 72 x 0.015 = €1.08 a session; against €0.95 a click, it returns 1.08 / 0.95 = 1.14 before product costs.
• Use the purchase tag you already run. Plerdy picks up Shopify orders by itself and, through Google Tag Manager, an existing GA4 purchase tag (same page), so both tools read the same orders.
• Read the channel list first. Plerdy sorts sessions into Google Ads, Google Organic, Paid Social, Social, Email, Referral, Direct and Other from UTM tags and referrers (Traffic Value FAQ). GA4's groups are not the same eight.

Plerdy and the read, side by side

How Plerdy assigns a sale

Plerdy publishes its method. The Ecommerce Analytics page gives the formulas: AOV is revenue divided by orders, eCR is orders divided by all sessions, and revenue per session is AOV times eCR, each compared with the previous period of the same length. Every Plerdy report shares that one calculation, and none shows profit, because Plerdy imports no costs.

The window is a single session. The guide's own example: a click on an ad on Monday and a return through Direct on Friday sends the order to Direct. The Traffic Value page offers filters for country, device and source, and no attribution model to switch.

Two more rules matter. Traffic Value labels pages Good, Traffic Problem, Landing Problem, Complex or Critical, and gives no diagnosis under 30 sessions (Traffic Value FAQ); the guide calls verdicts under 100 preliminary. Element purchase lift divides the conversion rate of sessions that clicked a block by the rate of those that did not, needs 20 clicking sessions, and is shown as correlation, not cause (Ecommerce Analytics).

What Plerdy's channels become in GA4

Plerdy sends no traffic. Each visit keeps the source and medium it arrived with, and Google's default channel group files it. The two lists part ways at the edges. Plerdy's single Google Ads row holds what GA4 splits into Paid Search, Paid Shopping, Paid Video and Cross-network, where Performance Max lands. Plerdy's Social is GA4's Organic Social, which accepts a social source or a medium such as "social" or "sm". A medium that matches none of Google's rules is filed as Unassigned, a bucket Plerdy's list does not have.

What crosses into GA4 is Plerdy's own events and goals: a checkbox in the site's settings transfers them, automatically where GA4 runs through Google Tag Manager, with a trigger still to configure (event tracking). They arrive as events and move no session between channels.

Worked example: one month, three ledgers

For example, with figures chosen for the arithmetic and no store behind them: the store books €96,300 over 28 days. Plerdy's channel rows add up to €91,740, the revenue it saw inside tracked sessions. The difference is worth a look before anything else: the pricing page caps E-commerce tracking at 500 on Startup and 1,500 on Scale.

14,200 + 9,860 = €24,060 for the two paid rows. Google Ads claims €38,600 for the same days and Meta €27,950, so 38,600 + 27,950 = €66,550, which is 66,550 - 24,060 = €42,490 more than Plerdy gives them. With the email platform's €21,400, claims reach 66,550 + 21,400 = €87,950, 91% of the €96,300 booked, before organic search.

Direct holds 31,450 / 91,740 = 34.3% of Plerdy's revenue. Some of it belongs to paid clicks made days earlier, and Plerdy cannot say how much, because its rule ends with the session.

That gap is what the read puts numbers on. Per its how-it-works page, it sets each channel's caused share beside a last-click rebuild from the same export, grades each with a data-health score and adds a next step: raise, cut or hold, tested in one region first.

Suppose the read puts Paid Social above Plerdy's €9,860 and well below Meta's €27,950: keep the page fixes, and move Meta spend in one country first.

The free check: one country, two periods

Plerdy's reports filter by country and compare with the previous period of the same length (Ecommerce Analytics), which makes a geographic test free. Pause one paid channel in one country for 14 days, change nothing else, and read that country's orders and revenue per session against the period before, beside a country left alone. If revenue barely moves where the channel went quiet, its claim was mostly buyers who would have come anyway. Seasonality sits inside the result, so treat it as a direction.

For the page question, Plerdy's A/B testing tool targets a test by device, country or traffic source, and the pricing page lists A/B testing as unlimited on every plan, Free included. Its page names no confidence threshold for a winner, so give each variant at least the 100 sessions the guide wants before a firm verdict.

For every channel at once, drop the GA4 Attribution paths CSV on the start page; the first finding is free and computed in the browser. Causality Engine, which I founded, charges €99 once for the full read, excluding VAT, refundable within 30 days (pricing, checked on 11 November 2026).

Limits on both sides

Nothing connects the two: Causality Engine reads no Plerdy report, only the channels a GA4 export records (integrations).

• GA4 history. The read wants 12 months of data or more (how it works), while Plerdy keeps 3 months on Startup and Scale (pricing).
• Key events. Purchase must be a key event; the read takes key events from GA4's Advertising section (one-off read).
• Thin channels. A channel holding less than 1% of credit by both the causal and the last-click method gets no verdict, only the note that it is too small to act on (how it works).
• What Plerdy keeps. Rage clicks, scroll depth, element lift, the Critical pages. A channel read never sees a click.

The Traffic Value page tells buyers they are buying budget decisions rather than a report, and its same-session rule makes paid channels look weakest exactly where Google Ads' 30-day default, as the guide records it, makes them look strongest: each default argues for what its seller sells, page fixes on one side and more clicks on the other.

See also

• Plerdy: per-site plans, Traffic Value and the review record
• How to use Plerdy: paid landing pages that sell nothing, found in order
• Plerdy's affiliate program terms: what Plerdy pays referrers, on the marketplace
• Causality Engine: prices, method and the cases it declines
• How to use Causality Engine: from GA4 file to a one-region test
• Causality Engine on the marketplace: referral terms for the owner's product
• Exporting GA4 data for attribution: the Attribution paths file, step by step
• Break-even ROAS calculator: the return a channel must clear

FAQ

Can Plerdy's revenue report feed Causality Engine?

No. The read takes one file from GA4, not from Plerdy, and there is no connector in either direction; Plerdy's sales show up there only as the GA4 purchases the same tag records (integrations).

Why does Plerdy's Google Ads revenue sit far below the Google Ads account?

Plerdy keeps a purchase in the session it happened in, and Google Ads counts it for up to 30 days after the click by default, per the wasted ad spend guide. A buyer who returns through Direct is Direct to Plerdy and an ad conversion to Google.

Does a high purchase lift prove a page block sells?

No, and Plerdy says so: the Ecommerce Analytics page labels element lift as correlation, not cause. Test the block with an A/B test before rebuilding the page around it.