The Context Box

The Most Neglected Line in Your Template

A context box is one short, plain-language sentence near the footer telling the recipient why they are receiving this email. It has one purpose: to close the gap between "I don't recognise this sender" and the spam button.

• It is a memory aid, not a legal instrument. No mailbox provider gives you credit for having one — there is no header or standard that carries it. It works entirely through the human reading it, so the only thing that matters is whether that person finds it credible.
• Its leverage is on complaints, the metric with the least forgiveness. Google wants your user-reported spam rate under 0.10% and never at 0.30%; Yahoo holds bulk senders under 0.30%. Complaints are the recipient's answer to a question — "why am I getting this?" — and the context box is your best chance to answer it first, at the exact moment the recipient is deciding.
• It earns the most where recall is weakest: long gaps between signup and first send, sub-brands that don't match the signup domain, re-engagement of dormant segments, and lists that changed hands in an acquisition. These are precisely the sends where most programs skip it.
• It is not the unsubscribe link. Keep them adjacent but distinct: the box explains the relationship, the link ends it. A recipient who wants out and finds only an explanation will complain.

Store the Source, Then Merge It

The multi-guess footer is a symptom. The disease is that acquisition data was never captured per record — Al's "do it right" is, underneath, a data-model requirement.

• Capture at signup: the specific source (form, page, purchase flow, event), the timestamp, the exact consent language shown at that moment, and the submitting IP for web forms. The consent language stays in your records as proof; the source and date are what the footer will use.
• Merge source and date into the footer at send time, so every recipient sees the one reason that applies to them. A dynamic context box is the only kind that stays accurate as the program adds channels.
• Never write a fallback that lists possibilities. If a record has no source, the honest fallback is not a menu — it is routing that record out of the send.
• Audit inherited lists hardest. Data that arrived through an acquisition, an agency handover, or a platform migration is the most likely to have lost its provenance in transit. Treat missing source data on an inherited segment as a suppression trigger.

Write One True Reason: Al's Five Checks

Al's post ends with five checks a context box must pass — specific, accurate, permission-based, verified, useful. In practice:

• Be specific — name the event, the place, and the date. "You're receiving this because you signed up for our newsletter at example.com on 14 March 2026" gives the recipient three separate hooks for recognition; any one may land. Don't hand them the phone book of every opt-in route you operate.
• Be accurate — let it be unflattering. A purchase, a donation, a badge scan — say exactly that, and accept the unsubscribes it produces. If the box doesn't match reality, recipients assume you lied or bought their data, and the complaint rate goes up, not down.
• Be permission-based — a confession is not a permission. "We obtained your address legally" is accurate and still fatal: legally compliant spam is still spam to the person reading it, and the report button is available to them no matter how lawful your send was. If the true sentence describes a purchase rather than a signup, the fix is not better copy — it is not sending (next section).
• Be verified — or don't assert. Without double opt-in, signup forms get forged, pranked, and mail-bombed. Telling an angry recipient "you signed up on our site" when a bot or an enemy submitted their address only fuels the fire. Confirmed opt-in is what makes the sentence in your footer safe to write.
• Be useful — plain language, legible type. The job is to jog a memory at a glance. The moment it reads like a lawyer wrote it, the recipient treats it as a defence — the song's "fine-print alibi" — and grey 8-point type at the bottom of a long footer is functionally absent, especially on a phone.

If You Cannot Name the Source, Do Not Send

The song's outro is the rule, stated without decoration.

• Suppress unknown-provenance records before the campaign, not after the complaints arrive. One send to a segment you cannot account for can undo months of reputation building on a domain and IP.
• Segment by acquisition source and monitor each source separately. Bounce and click rates are available per segment directly. Complaint attribution is split by provider: Yahoo's Complaint Feedback Loop and Microsoft's Junk Mail Reporting Program return a per-message ARF report for each complaint, which your ESP maps back to the campaign; Gmail sends no per-message reports — its FBL is aggregate, keyed on the Feedback-ID header you set per source, and surfaced in Postmaster Tools. Tag your mail for both, and remember the per-campaign complaint rate in an ESP dashboard undercounts reality, because Gmail — usually the largest share of any list — is missing from it.
• Read the right dashboards. Google Postmaster Tools reports spam rate per authenticated domain; Microsoft SNDS reports per IP — on a shared IP it describes the pool, not you. Keep hard bounces under the ~2% industry warning line.
• Expect traps in the unexplained records. Pristine spam traps never opted in to anything — they enter lists through scraping, purchased data, careless imports, or bots stuffing unprotected signup forms. Recycled traps arrive through age. Both concentrate exactly where nobody can name the source.

The Compliance Boundary

A context box is good practice, not compliance — conflating the two is how senders come to believe a footer protects them.

• CAN-SPAM requires truthful headers and subject lines, ad identification where applicable, a physical postal address, and an opt-out honoured within ten business days. It requires neither consent nor any explanation of how you got the address.
• GDPR — where consent is your lawful basis, it must be freely given, specific, informed, and unambiguous, and demonstrated by a stored record, not a template sentence. Article 14 is the provision worth knowing: when data came from somewhere other than the data subject, you must tell them the source at the latest at first communication. That is a legally required context box for EU-facing senders using third-party data.
• CASL requires express or defined implied consent, sender identification with valid contact information, and an unsubscribe honoured within ten business days — with the burden of proving consent on the sender.
• One-click unsubscribe (RFC 8058) — the List-Unsubscribe / List-Unsubscribe-Post header pair — has been required since February 2024 by Google for senders above 5,000 messages a day to Gmail and by Yahoo for bulk senders, with opt-outs processed within two days. It sits alongside the context box, not in place of it: the box explains, the header exits.

Conclusion

Al Iverson's framing survives every technical test applied above: the context box is powerful, widely botched, and only as good as the data behind it. Generated from real per-record acquisition sources, it is a free, immediate reduction in the complaint rate — the metric with the narrowest margins in modern deliverability. Written from a template, it is a list of guesses announcing that the sender never knew who you were. His closing advice is the standard: if you can't make it specific and accurate, skip the disclaimer entirely and focus on sending content people actually remember signing up for. And if the honest version of the sentence cannot be written at all, that is not a copywriting problem — it is a permission problem telling you not to send.

• Store source, timestamp, consent language, and signup IP on every subscriber record.
• Merge source and date dynamically into the footer — one true reason per recipient, never a menu.
• Write it plainly, name a date and place, keep it readable on a phone.
• Run double opt-in, so the signup you cite is one the recipient actually performed — forms get forged and mail-bombed.
• Suppress every record whose source you cannot name — before the send, not after the complaints.
• Attribute complaints per source through the feedback loops; keep Gmail's spam rate under 0.10%.
• Remember what it is: complaint prevention, not legal cover — except under GDPR Article 14, where naming the source is the law.