Why we publish no inbox placement ranking

What an inbox placement claim would have to survive

Suppose we wanted to say that one platform reaches the inbox more often than another. To mean anything, the test would have to hold everything else still: the same sending domain with the same authentication records, the same list of real recipients who signed up the same way, the same content and subject line, the same sending history, the same time of day, repeated enough times to separate a difference from noise.

Change any one of those and the test measures that thing instead. Send from two different domains and you measured two domain reputations. Use a purchased seed list on one side and your own subscribers on the other and you measured two audiences. Most published deliverability rankings have at least one of these problems, and almost none of them publish the setup in enough detail for you to tell which.

There is also a timing problem. Reputation is earned over weeks and it is domain and IP specific. A test that runs for a day on a fresh account measures a cold start, which is a real thing worth knowing about, but it is not the same question as where your established newsletter would land.

What the mailbox providers actually publish

The companies that decide where your mail goes do publish requirements, and those are the only public, enforced thresholds in this field. They are also the part of deliverability that is genuinely under your control.

Taken from the mailbox providers' own published requirements. These are the only thresholds in this field that are both public and enforced by the party that actually decides where your mail lands.

Note what is missing from that list: neither provider publishes an inbox placement rate, for you or for anyone else. Gmail's own tool reports spam rate, reputation, authentication and delivery errors, and it will not tell you what share of your mail reached the inbox rather than the spam folder. If the party doing the filtering does not publish that number, a comparison site claiming it for nine platforms should be read with suspicion.

Where the platform genuinely does make a difference

None of this means the tool is irrelevant. It decides what you can set up and what you can see. Whether you can authenticate your own sending domain rather than borrow the provider's. Whether the unsubscribe header is handled properly on every send. Whether you can separate hard from soft bounces, suppress cleanly, and export the raw events to check. Whether a dedicated IP is available, and whether it makes sense at your volume.

Those are capability questions, and they are answerable from documentation without a test. Where we can answer them, we do, on the platform pages and in the deliverability basics guide. Where a provider does not publish an answer, we say so instead of filling the gap.

The test we would run, and why we have not

Our protocol, if we ran it, would be: one sending domain, the same SPF, DKIM and DMARC records pointed at each platform in turn, one identical list of real opted in recipients, one identical message, sent in alternation over several weeks, with placement checked across the major providers each time and the spam rate read from each provider's own tooling rather than inferred.

That requires a paid account on every platform in the comparison, a warmed domain, and a list of real people willing to be part of it. We have not done it. Running it badly would produce a number that looks authoritative and means nothing, which is worse for you than an honest gap. So this page exists instead of a ranking, and the tables elsewhere on this site stay on ground we can source: prices, published limits, documented capabilities and migration objects.

What you can check yourself, for free

• Authentication and alignment: SPF, DKIM and DMARC on your own sending domain, with the From domain aligned to one of them. Both providers above require it.
• Your spam rate: verify your domain in the mailbox provider's own tooling and read the number they act on, rather than a third party estimate.
• Your unsubscribe path: one click, visible in the message, working without a login, and honoured quickly. It is a published requirement, not a courtesy.
• Your own list: how people joined, how recently they engaged, and how cleanly bounces are suppressed. This is the lever with the largest effect and the one no platform can pull for you.

If a vendor or a review site tells you their platform has better deliverability, ask what was held constant. The answer tells you how much of the claim to keep.