The blind spot
Between the send button and the inbox, anything can change
Your sending platform adds a tracker. A migration breaks your authentication. An unsubscribe link starts routing through a tracking domain. Nobody on the team decided it, and nobody sees it: your email tool shows you what it sent, not what arrived.
We receive your campaigns and your transactional — An email triggered by an action of the person: purchase confirmation, password reset, legal information. It is the opposite of a campaign, sent to a list at a time chosen by the brand. on real mailboxes, and we measure what they actually contain. Nothing to install, no access to your systems, no line of code to add. No image loaded, no link followed: your own trackers are never triggered.
01 — Observe · free
A report your technical team can fix the same day
24 criteria and 1 bonus, four score families — compliance, deliverability, technical, experience — and for each criterion, the message excerpt that establishes it.
A disputed finding is replayed. It lands on the same value, with the same evidence.
Not met An open-tracking pixel is present
To check No untracked path leads to the brand's site
Not observed Unsubscribe actually processed — beyond the reach of an external measurement
<img src="https://t.maisonclaire.example/o/?m=8f2a" width="1" height="1">
14 liens sur 14 passent par t.maisonclaire.example
List-Unsubscribe-Post: List-Unsubscribe=One-Click
Test · unsubscribe
mc-n01@temoin.example
02 — Test · Version Pro
Your rules, tested on real emails
Sentinels: addresses you place in your CRM, your store, or your forms, that live like real contacts. With them, we unsubscribe on a chosen date, we exercise the objection to tracking, and we observe what happens next.
A silence concludes nothing: every test keeps a control sentinel, which keeps receiving when the other no longer should. Sentinels open up with Version Pro, by batches.
03 — Monitor · Version Pro
Monitoring checks every new email, where an audit describes only one.
Every email is measured as it arrives, campaign or transactional alike. The history builds itself, and you can prove, as of any date, what a message sent six months ago contained.
Full history
The archive is never rewritten. A criterion added is replayed over the entire past.
Sector comparison
Your position among the brands in your sector, on the same criteria.
no tracker detected
<img src="…/o/?m=1" width="1">
Observed between two consecutive emails from the same brand.
Alerts
We catch what changes from one email to the next.
An alert is never about a threshold. It is about a change, observed between two consecutive emails from the same brand, with both excerpts side by side. Sending it to your team opens up with Version Pro.
Impersonation and unexpected uses
We spot messages sent in your name that don't come from you.
Our addresses also receive what you did not send: a message signed from a domain close to yours. And with Version Pro, a sentinel only you hold will tell you if it starts receiving from elsewhere. It is the only alert that holds nothing against you.
An email claims to be from your brand, sent from a domain you do not control.
A sentinel, placed only in your own systems, receives from a third party: an unexpected use, dated, with its evidence.
Your self-reported situation
Legitimate interest · pixel active across the whole base · no preferences link
Regulation
What the rule requires, read on your actual emails
Consent, unsubscribe, authentication, tracking: the rules governing email are tightening, and they change often. We follow them, and we read them against what your emails actually contain, with evidence — a finding, never an opinion on your situation.
Each change is explained simply, with what it concretely changes for your sending platform.
CNIL recommendation — 14 April 2026
Since April, the open pixel has fallen under consent. 89% of measured emails still carry one.
The recommendation says the tracker falls under consent, not the sign-up checkbox. Here is what emails actually do with it, measured across the brands we track, one email at a time.
89 %
of mailings contain an open tracking pixel. 1715 out of 1921
93 %
of brands measured on five mailings or more track all of them. 104 out of 112
80 %
offer no setting for tracking, only a full opt-out. 1541 out of 1921
1921 emails from 163 brands · run on 10/08/2026 · external observation, measurements replayable from the archive.
What we're asked
Is it really free?
Yes. Auditing an email is free: its report reads in full on screen, never blurred, without leaving an address. Only the PDF to take away asks for one.
How do you get this data?
We subscribe to newsletters like any other subscriber, from addresses we control. We receive nothing that an ordinary customer would not also receive, and we have no access to any system: no account, no API, no internal statistics.
Do you open the emails you receive?
No. No remote resource is ever loaded, no tracked link is ever followed. Loading an image would trigger the pixel of the brand we're measuring and distort its own data; following a link would register a click and mark the contact as engaged. Our measurements do not affect the statistics of the brands we track.
Do you use artificial intelligence?
Not for the measurements. No published measurement depends on a language model: each analyzer is a pure function — the message goes in, the finding comes out, with no network and no clock. A figure that cannot be recalculated identically is not defensible the day someone disputes it.
Do you say a brand is at fault?
No. We publish a finding on what a received message contains, with the excerpt that establishes it. The legal qualification is not ours to make: it depends on your purposes and your legal basis, which we do not know.
I work for a monitored brand.
You can claim the brand and access its full history, as well as the criteria we cannot observe from the outside. If a finding seems wrong to you, ask for it to be recalculated: it lands on the same value, with the same evidence.
Audit your next email before your subscribers do
Full report, free, no account. 24 criteria and 1 bonus, with the excerpt that establishes each finding, on your campaigns as much as on your transactional emails.