Sender With Causality Engine: Check the 5-Day Revenue Claim

The short answer: Sender's Ecommerce section credits email with far more orders than GA4 files under Email, and both screens are working as designed. Sender, an email and SMS platform, counts an order for the last message a contact opened or clicked in the 5 days before buying, per its sales attribution article. That is a rule about timing. It says nothing about what the email caused. If your shop records purchases in GA4, the route is short: tag Sender's links, keep one group out of a send, then set the claim beside Causality Engine, a causal attribution tool that takes the GA4 Attribution paths export as its only input. Its first finding costs nothing 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 sums before the report goes to anyone:

1. Price an open against a click. For example, a campaign to 8,000 contacts records 2,960 opens and 240 clicks. Every opener sits inside the window, so 2,960 people can earn email a sale where 240 visited the shop: 2,960 ÷ 240 = 12.3 times as many.
2. Start from what is wired up. The shop is already listed under Connected stores, GA4 already holds the purchase event, and one toggle adds the tags.
3. Fix the yardstick now. Orders per 1,000 contacts, mailed against held out, counted in the shop's own order export.

Sender and the read, side by side

Neither product connects to the other. Sender reaches the read only as tagged sessions inside the GA4 file.

What Sender counts as its own sale

The sales attribution article is plain about the mechanics. A sale is attributed when a contact engages with a message and then orders inside the window; receiving a message is not enough. For email, an open or a click qualifies, each for 5 days. For SMS, a click qualifies for 5 days and a delivery for 12 hours. Where a contact touched several messages, the most recent one takes the sale, and every order placed inside one window goes to that message.

Three details widen the count. Opens that Apple Mail Privacy Protection pre-loads are recorded as opens, and they qualify. A refunded or cancelled order keeps its attribution. And the purchase needs no click, so an order that arrived through a Google ad still belongs to email if a newsletter was opened that week. The article itself says Sender's revenue will not match the store platform's, and calls such differences expected.

The figures appear in the Ecommerce section of each Campaign overview, per the attribution basics guide. Asked whether the windows can be changed, the sales attribution FAQ restates the defaults and names no setting. Plan access is described two ways: the pricing table lists Ecommerce Reports from Standard upward, while the attribution basics guide places the Enable ecommerce tracking toggle on Pro plans, so check your own account.

Getting Sender's clicks into GA4 as Email

Go to Account settings, General settings, and find the Google Analytics section. With the toggle on, Sender appends three parameters to every tracked campaign link: utm_source=newsletter, utm_medium=email and utm_campaign={$subject}, which inserts the subject line (attribution basics guide). GA4's default channel group puts a visit in Email whenever its medium or its source says email, so the default medium does the job.

What the guide warns about:

• The toggle is off. No tags go out, and email clicks land in another row.
• The subject line is the campaign name. Every send becomes its own line in GA4. Set custom campaign name, on the Campaign settings page, gives a stable one.
• Nobody looked before sending. The Review and send step shows the value as Appears on Google Analytics as.

A caution on the defaults. Sender's older Google Analytics article gives the account username as the default source, where the newer guide says newsletter. Open the field and read it. Both pages describe campaign sends, so click a link in a live automation email and confirm the address carries the tags.

A worked example: 640 orders, 777 claims

For example, as an illustration and not a store's results: in one month the shop's order export shows 640 orders. Sender's Ecommerce sections add up to 212. Meta's dashboard claims 301 and Google Ads 264, each by its own window.

• Claims: 212 + 301 + 264 = 777.
• Claims per order: 777 ÷ 640 = 1.21.
• Claimed at least twice: 777 - 640 = 137 orders.

GA4's paths report, whose chart Google's report help says is drawn on a last-click model, gives Email 58 of the 640. Sender's figure is 212 ÷ 58 = 3.66 times GA4's. The gap is 154 orders: buyers who opened or clicked within 5 days and then arrived by another road, or were on their way already.

Neither figure answers what email caused. 212 is everyone who was near an email. 58 is everyone whose last tracked step was one. The read adds a third column: Email's share of credit once Direct has been handed back to whichever channels brought those buyers, beside last click rebuilt out of that same CSV, with a 0 to 100 data-health score and a suggested next step, as how it works describes. If that share lands far from Sender's claim, move no budget on either number. Test on a held-out group first.

The free check: one group that gets nothing

A holdout is enough on its own, and Sender's recipient picker supports it. Deselect All active subscribers and two columns appear, Send to these groups/segments and Exclude groups/segments; anyone in an excluded group does not receive the campaign (include and exclude guide). The random draw is by hand: export the list, add a random-number column in a spreadsheet, and import one contact in ten into a group named holdout. Sender's A/B experiments compare versions of an email, so they cannot play this part.

Exclude that group for a month, then count orders for both groups in the shop's export. For example, 7,200 mailed contacts place 252 orders and 800 held-out contacts place 20: 252 ÷ 7.2 = 35.0 per 1,000 against 20 ÷ 0.8 = 25.0. Email added 10.0 orders per 1,000 mailed, 10.0 × 7.2 = 72 orders, where the window claimed 212 and GA4's Email row held 58.

Then the channel view. Give the start page your paths export: a preview, computed inside the browser, names the channels behind the buyers that last click parks under Direct. If email is among them, Causality Engine, which I founded, runs the full read on the same file from the start page: €99 once, before VAT, with a 30-day refund (pricing, checked on 11 November 2026).

Where the pairing stops

The read needs a purchase key event in GA4 and asks for 12 months or more of history, per the export steps on its start page. It works at channel level and will not say which campaign or automation earned the credit; Sender's per-campaign Ecommerce section and its Purchases tab do that. The one-off read page advises against buying when analytics history is thin or broken, and says a channel under 1% of credit on both methods is called too small to act on. An open that never becomes a click never reaches GA4, so the read cannot see it.

One structural fact about the default. Ecommerce Reports are missing from Sender's free plan and begin on Standard (pricing table). The attributed-revenue figure therefore first appears on the plan a shop has just started paying for, under a rule in which a pre-loaded Apple Mail open counts the same as a click. A wide window suits a number whose first job is to sit next to the invoice.

See also

• Sender: plans, EU hosting and the rivals it meets
• How to use Sender: domain records, import and a first campaign
• Sender in the marketplace: Sender's affiliate program terms
• Causality Engine: what it reads, costs and declines
• How to use Causality Engine: from GA4 export to a tested verdict
• Causality Engine in the marketplace: its affiliate program terms
• How to measure email incrementality: the holdout method at flow level
• Email marketing ROI calculator: turn caused orders into a return figure

FAQ

Can Sender's 5-day attribution window be changed?

The sales attribution article poses that question and answers with the defaults: 5 days for email opens, email clicks and SMS clicks, 12 hours for SMS deliveries. It describes no control for altering them.

Does Causality Engine plug into a Sender account?

No. There is no connector, login or API link in either direction. Causality Engine's integrations page says it reads every channel the GA4 export records, so Sender shows up as Email once its links carry the tags.

Why does GA4 show fewer email orders than Sender?

GA4 credits Email only when a tagged click led to the session. Sender also counts orders that followed an open with no click, including Apple Mail's pre-loaded opens, for 5 days.