The default window credits email for orders it did not cause, and Apple's privacy changes made the problem substantially worse.
Klaviyo says email drove $412,000 last month, 22.3 percent of your revenue. Change one setting, re-run the identical date range, and the number becomes $198,000.
Nothing about the business changed. You switched attribution from Klaviyo's default, which credits email for an open, to click-only. The $214,000 difference was resting entirely on opens, and since 2021 a large share of those opens were performed by Apple rather than by a person.
Klaviyo's out-of-the-box conversion setting is a 5-day click window and a 5-day open window. An order is credited to email if either condition is met.
The click half is defensible. Someone clicked your email, arrived on your site, bought within five days. You can argue about the length of the window, but there is a real interaction.
The open half is the problem, and it is doing more work than most people realize. It credits email when the recipient did nothing but open, or appeared to. No click. No visit from the email. They could have opened your campaign on Monday, forgotten it entirely, searched your brand name on Thursday and bought direct, and Klaviyo would record that order as email revenue.
In September 2021 Apple shipped Mail Privacy Protection, on by default and accepted by most users during setup. When enabled, Apple Mail pre-fetches all images in a message through a proxy the moment it arrives, whether or not the recipient ever opens it.
Open tracking works by embedding an invisible image. When Apple fetches it, your email platform records an open. There is no way to distinguish this from a human opening the message, because from the server's perspective they are the same event.
On a typical DTC list, Apple Mail accounts for somewhere between 50 and 70 percent of the audience. So more than half of your opens are now machine-generated, they fire on essentially every send, and each one starts a fresh 5-day attribution window for that subscriber.
The compounding effect matters. If you send twice a week, every Apple Mail subscriber is inside an open-attribution window permanently. Any purchase any of them make, from any channel, gets credited to email.
Stated plainly: for the majority of your list, Klaviyo's default settings credit email with every order they place, regardless of what actually drove it. That is not a Klaviyo defect. It is the default configuration doing exactly what it says, in a world where opens stopped meaning anything.
Crestline Co.'s month, same orders, two attribution settings.
| Attribution basis | Attributed | Share of net sales |
|---|---|---|
| Default: 5-day click or 5-day open | $412,000 | 22.3% |
| Click-only, same 5-day window | $198,000 | 10.7% |
| Resting on opens alone | $214,000 | 11.6% |
Broken down by flow and campaign type, the distortion is uneven:
| Type | Default | Click-only | Overstated by |
|---|---|---|---|
| Abandoned cart | $61,000 | $44,900 | 36% |
| Welcome series | $38,000 | $24,700 | 54% |
| Post-purchase | $47,000 | $12,400 | 279% |
| Weekly campaigns | $266,000 | $116,000 | 129% |
Crestline Co. is a simulated DTC brand used for demonstration. These figures are illustrative and are not a client result.
Post-purchase is the clearest case. Those emails go to people who just bought. Under a 5-day open window they collect credit for the order that preceded them, plus any repeat purchase in the following five days. The flow is being paid for revenue it arrived after.
Broad weekly campaigns are the largest absolute distortion simply because they reach the most people, so they open the most windows.
The number itself is harmless. The decisions made on it are not.
The setting is easy. Noticing it matters is the hard part. An inflated email number does not look like an error, it looks like a channel performing well, and it stays unexamined for years. Postlytix reads your email data against actual order and margin data rather than against the platform's own attribution, and flags where a reported contribution and a real one have separated.
This is one of eight places the numbers come apart. Why your e-commerce revenue numbers never match explains why fragmentation happens across the whole stack, and links to the rest.