The main findings
- 81% enforce DMARC: 27 refuse failing mail and 48 send it to spam (5 of them on only part of it). 16 only watch (p=none), and 2 have no DMARC record.
- Start-ups prefer “spam folder” to “refuse”: 29% refuse failing mail, against 64% of large listed companies.
- 15% have an SPF problem, and 10 (11%) need more than the 10 DNS lookups SPF allows, so receivers may fail their genuine mail. One record needs 23. At large companies it was 7%. Each new SaaS tool adds an include, and growing companies add many.
- 71% receive mail on Google Workspace and 12% on Microsoft 365, the mirror image of large companies, where 61% use Microsoft.
- MTA-STS is more common than at big firms (12% against 5%), which makes receivers insist on encrypted delivery; BIMI, the logo beside their mail, is less common (27% against 36%).
The numbers
| Check | Start-ups | Share | Large companies |
|---|---|---|---|
| DMARC reject (all or part of mail) | 27 | 29% | 64% |
| DMARC quarantine (all or part) | 48 | 52% | 23% |
| DMARC only watching (p=none) | 16 | 17% | 2% |
| No DMARC record | 2 | 2% | 11% |
| SPF correct | 79 | 85% | 82% |
| SPF over the 10-lookup limit | 10 | 11% | 7% |
| Several SPF records (an error) | 2 | 2% | — |
| DKIM found (at least) | 92 | 99% | 80% |
| MTA-STS published | 11 | 12% | 5% |
| BIMI published | 25 | 27% | 36% |
| Mail on Google Workspace | 66 | 71% | 5% |
| Mail on Microsoft 365 | 11 | 12% | 61% |
What it means for a growing business
Start-ups add tools fast: a CRM, a help desk, a newsletter service, a billing platform. Each one asks to be added to SPF, and after a few the record passes the 10-lookup limit without anyone noticing. Receivers then treat SPF as broken, and genuine mail that depends on it can fail or land in spam.
- Count your lookups with the free email domain checker; it follows every include.
- Remove tools you no longer use, and send bulk mail from a subdomain: see SPF “too many DNS lookups”.
- Rebuild one clean record with the SPF record generator, and move DMARC from p=none to quarantine with the DMARC record generator.
Method and questions
Which companies were checked?
The main web domains of 96 well-known Indian start-ups and unicorns in e-commerce, fintech, education, health, mobility, software and consumer brands, taken from the companies’ own websites. Three were left out because the domain does not receive email or has no website, which leaves 93. We do not name companies: the point is the overall picture, and every company can check itself in seconds.
How was it measured?
On 14 November 2026 we read each domain’s public DNS records (MX, SPF, DKIM at common names, DMARC, MTA-STS, TLS reporting and BIMI) through two public resolvers, with the same code as our free email domain checker.
What does “quarantine for 50% of mail” mean?
DMARC lets a domain apply its policy to only part of the failing mail while it tests. Five start-ups do this; we count them separately from those enforcing on all mail.
Why might DKIM be under-counted?
A DKIM key is published under a name the company or its provider chooses. We tried the names common providers use; a key under a custom name cannot be seen from outside, so the DKIM figure is a floor.
Does a missing or weak record mean a company was hacked?
No. It means mail pretending to come from that domain is easier to deliver, or that genuine mail may fail checks. It says nothing about the company’s own systems.
You may quote these figures with a link to this page.