The figures we cite, with their sources.

Best and worst segments

The provider is only half the story; the sending profile matters as much. Transaction-heavy, highly engaged segments clear the high-90s, while sectors with weaker engagement and colder lists sit near the bottom — the same infrastructure delivering very different outcomes depending on who is sending and to whom.

The authentication gap

Authentication is now mandatory for bulk senders, yet adoption across the wider web is still thin. Of the top ten million domains, fewer than one in five publishes a valid DMARC record, and fewer than one in twelve enforces it at quarantine or reject — which is why a published p=none meets the letter of the rule while protecting almost nothing.

The bulk-sender rules

The requirements that decide whether bulk mail is accepted at all, side by side. The principles rhyme across the three providers; the specifics — thresholds, dates and codes — do not, and the gaps are where compliant-looking mail fails.

The numbers worth memorising

A handful of figures do most of the work in day-to-day deliverability decisions.

• Spam complaints: keep below 0.10% as a working target; 0.30% (three per thousand) is the line where providers act regardless of authentication.
• Bounce rate: hold under 2%; the cross-industry average sits around 2.33%.
• Bulk threshold: 5,000 messages a day to one provider’s consumer inboxes makes you a bulk sender, counted per provider.
• Delivery is not inbox: at scale, technical acceptance can overstate real inbox reach by roughly 40% — mail accepted, then filtered out of view.
• High volume, harder placement: senders past a million messages a month have seen inbox placement fall well below 30% when authentication and hygiene slip.

Sources: Validity 2025; GlockApps 2025–2026; Unspam 2025 deliverability report; industry B2B benchmarks 2025. Figures from different panels are not directly comparable; we cite each so you can judge the method.