Map Extract
Free tool

SPF record checker and lookup counter

Enter the domain you send from. The checker fetches its SPF record from public DNS, follows every include the way a receiving server would, counts the lookups against the hard limit of 10 and tells you in plain words what to change.

Runs in your browser. Domain names are looked up through public DNS over HTTPS (Cloudflare, with Google as a fallback); nothing goes to Map Extract.

Checking a domain step by step

  1. Type the sending domain

    Use the domain after the @ in your From address, for example northfieldsurveys.example, not your website's www host. If you send cold email from a separate outreach domain, check that one.

  2. Run the check

    The page asks Cloudflare's public DNS over HTTPS for the TXT records. If that fails, switch the resolver to Google and run it again.

  3. Read the findings from the top

    Failures (red) break SPF for every message. Warnings (amber) are risks or weak settings. Passes confirm what is already right.

  4. Open the include tree

    The table lists each domain the record pulls in, how many lookups that branch costs and the IP ranges it contributes, so you can see which service is eating your allowance.

  5. Test before you publish

    Switch to the draft mode, paste the record you intend to publish and follow its includes. Optionally enter a sending IP to see whether the record would pass it.

What counts towards the limit of 10

RFC 7208 caps the terms that make the receiving server do another DNS query. Everything inside an included record counts as well, which is how a record with three visible includes can still blow the limit.

TermCosts a lookup?Notes
include:Yes, plus everything inside itEach sending service you add usually brings one, and many bring nested ones
a, a:hostYesAuthorises the host's A or AAAA addresses
mx, mx:hostYesAuthorises your mail servers; often unnecessary for sending
exists:YesRare outside large providers
redirect=YesHands the whole policy to another domain
ptrYes, and deprecatedSlow and unreliable; replace it
ip4:, ip6:NoLiteral address ranges cost nothing
allNoThe fallback for every server not listed

Example: a small agency that added one tool too many

Northfield Surveys (a fictional firm) sends from Microsoft 365, uses a newsletter platform, a help desk, a CRM that sends quotes and, recently, a cold outreach tool. Each vendor's setup page said to add an include, so the record grew one piece at a time.

Expanded, the record needs 12 lookups. Receivers stop at 10 and return a permanent error, so every message, including ordinary replies from Outlook, now fails SPF. The CRM has not been used since spring, and the help desk can be pointed at a subdomain, which brings the total down to 8.

Input
v=spf1 include:spf.protection.outlook.com include:mail.newsletter-vendor.example include:spf.helpdesk.example include:_spf.crm.example include:send.outreach.example mx ~all
Result
Lookups after following includes: 12 of 10  (permanent error)
Heaviest branch: _spf.crm.example (4 lookups)
Fix: remove the unused CRM include and the mx term (mail servers are not senders here),
move the help desk to support.northfieldsurveys.example with its own SPF record
Result: 8 of 10

Why one domain cannot have two SPF records

When a vendor's instructions say "add this TXT record", it is easy to create a second record that also starts with v=spf1. The standard is strict about this: if a receiver finds more than one, SPF evaluation stops with a permanent error, exactly as if the record were malformed. The fix is to merge them into a single record with all the includes and one all at the end.

Other TXT records, such as a Google site verification string or a DMARC record (which lives on _dmarc), do not interfere. Only records that begin with the SPF version tag count.

Before you send any outreach from a domain

  • Exactly one SPF record on the sending domain, ending in ~all (or -all once you are confident).
  • Every service that sends as that domain is included, and services you no longer use are removed.
  • No more than 10 lookups after expansion and no more than 2 includes that point at nothing (void lookups).
  • DKIM signing switched on in each sending service, using your domain rather than the vendor's.
  • A DMARC record published, starting at p=none with reporting, so you can see failures before tightening it.
  • A separate subdomain or domain for cold outreach, so a reputation problem there does not touch your main mail.

When the count is high but nothing seems broken

Some receivers are lenient and some are not, and providers change their own include chains without telling you. A record that sits at 10 today can tip over when a vendor adds a nested include next month. Treat 9 or 10 as a problem to fix now rather than later.

Flattening, which means replacing includes with the literal IP ranges they resolve to, does lower the count, but the ranges then go stale whenever the vendor changes its infrastructure. It is better to remove what you do not use, move secondary senders to subdomains, and flatten only as a last step with a plan to refresh the ranges.

Questions people ask

What does SPF permerror mean?

A permanent error means the receiving server could not evaluate your SPF record at all, so it treats SPF as failed. The usual causes are more than 10 DNS lookups after following includes, two SPF records on the same domain, a syntax mistake such as a missing colon after include, or more than two includes that point at names with no record.

Should my record end in ~all or -all?

Both tell receivers that unlisted servers are not authorised. ~all (soft fail) is the safer starting point while you confirm every legitimate sender is listed; -all (fail) is stricter. In practice DMARC policy decides what happens to failing mail, so getting DKIM and DMARC right matters more than this choice. Never use +all.

Does the ip4 mechanism count as a lookup?

No. ip4 and ip6 ranges are written into the record itself, so the receiver needs no further query. That is why replacing a small include with its address ranges reduces the count, at the cost of having to update the ranges yourself if the provider changes them.

Which domain should I check for cold email?

The domain in the From address of the messages you send, and also the envelope sender (Return-Path) domain if your provider uses its own. If you send outreach from a separate domain such as getnorthfield.example, check that domain rather than your main website.

Why does the result differ from another checker?

Checkers differ in whether they count nested lookups, whether they treat a missing include as a void lookup and when they query DNS. Records can also change between checks because of caching. This checker follows every include and counts the way RFC 7208 describes; the include tree shows exactly where its count comes from.

Is anything stored about the domains I check?

No. The lookups go from your browser straight to the public resolver you choose, as normal DNS queries. Map Extract does not receive the domain names or the results.

Where your data goes

The domain you enter is sent from your browser to Cloudflare's public DNS over HTTPS service (or Google's, if you switch the resolver), exactly like any other DNS lookup. No request goes to Map Extract, and nothing is stored. In draft mode, the record you paste is read on the page and only the domains it includes are looked up.

Checked against the tool on

Top