Direct Mail's Comeback: Why Brands Are Returning to Mail

The Mailbox Beat the Inbox to the Trust Problem

A well-targeted direct mail piece to a house list still pulls a response rate of roughly 5%, while a house-list email is lucky to clear 0.6%. That gap, tracked for years by the Association of National Advertisers and the old Data & Marketing Association before their 2018 merger, used to be a footnote agencies cited to justify a postcard budget line. In 2026 it has become the headline, because the channel it's being compared against just got a lot noisier.

I spent eleven years building outdoor, signage and experiential budgets for retailers and public sector clients across the Prairies, and the conversation I keep having this year in Winnipeg boardrooms isn't about billboards. It's about postcards, catalogs and the kind of tactile mail piece that a CMO's own kids don't get anymore. The reason isn't nostalgia. It's that email, the channel that was supposed to make mail obsolete, broke something in itself.

Generative AI made it essentially free to write, personalize and blast a marketing email. Every brand did it at once. Gmail and Yahoo responded in early 2024 by tightening the rules for bulk senders, requiring proper authentication and capping complaint rates, which pushed a meaningful share of legitimate marketing mail into spam or the promotions tab overnight.

What the Response Rate Data Actually Says

That table is an older benchmark and worth treating with some skepticism as a static number, but it's still the reference figure list brokers and mail vendors quote, because nobody has produced a more recent public study that contradicts the order of magnitude. What's actually new for 2026 is the corroborating evidence on the ground: the U.S. Postal Service's Informed Delivery product, which emails subscribers a scanned preview of their physical mail each morning, now has more than 66 million registered users.

That's tens of millions of households opening a digital email specifically to check what's arriving in their physical mailbox, an engagement behavior email marketers would sell a limb for. You can register for the free USPS Informed Delivery service yourself in under five minutes and see exactly what a brand's mail piece looks like next to the grayscale scan competitors get.

Canada Post's own neuromarketing research with True Impact Marketing, conducted with an insights team it still publishes updates through, found that physical mail requires roughly 21% less cognitive effort to process than digital media and produces stronger recall days later. That's not a marketing claim from a mail vendor with something to sell. It's the finding that makes tactile media worth a line item even when digital cost per impression looks cheaper on paper, a caveat I've made before about fleet graphics and their real cost per impression: the cheapest number on the invoice isn't always the cheapest number that matters.

The Brands Already Spending Here

Chewy built part of its customer loyalty reputation on handwritten condolence cards and pet portraits mailed to customers, a program chronicled in Business Insider and Adweek profiles over the years, and it's the kind of tactile gesture no AI-drafted email replicates convincingly. Financial services never actually left mail: Mintel Comperemedia has tracked credit card issuers like Capital One and Chase holding or growing direct mail volume even as digital ad spend rose, because acquisition offers with real APR terms convert better on paper that customers keep on the counter.

Direct-to-consumer brands are the more interesting case, because they're the ones who supposedly proved digital-only acquisition could scale. Casper's early growth playbook, well documented in trade press case studies from its first few years, leaned on direct mail specifically for retargeting cart abandoners and lapsed trial customers, a segment too small and too high-value to blast with generic email but perfect for a personalized postcard.

Where Mail Fits Now, and Where It Doesn't

None of this makes direct mail cheap. A standard postcard drop runs $0.50 to $1.20 per piece once printing, list rental and postage are counted, against fractions of a cent for an email send. That math only works when the audience is narrow enough and the offer valuable enough that a 3% to 5% response beats a much larger but nearly free email blast on total contribution margin, not on cost per impression alone.

• Use it for house-list win-back: customers who lapsed six to eighteen months ago are worth more per contact than a stranger, so the higher cost per piece is easier to justify.
• Use it for high-consideration or high-LTV categories: financial products, healthcare, home services, education, anywhere a physical piece signals legitimacy an email can't.
• Pair it with an OOH or fleet campaign in the same geography rather than running it in isolation. If you haven't audited what those formats actually cost per market, our out-of-home formats and pricing guide breaks down the current numbers by format.
• Skip it for low-margin, high-frequency purchases where the per-contact cost can't be recovered even at a strong response rate.

Treat the channel the way you'd treat any line in a broader physical mix. Our out-of-home formats and costs buyer's guide is a useful companion here, since the same discipline about true cost per contact applies whether you're pricing a transit wrap or a mail drop.

Measurement Is Still Mail's Weak Spot, and That's Fixable

The honest objection to direct mail has never been that it doesn't work. It's that attribution is harder than a click. A postcard doesn't fire a pixel, and unless a customer types in a specific promo code, the causal chain from mailbox to purchase gets murky fast, a problem our overview of marketing attribution models for 2026 budgets covers in more depth for any offline channel.

That said, digital attribution isn't the clean alternative it once claimed to be either. Google itself has said its AI Overviews and AI Mode send 'billions' of clicks weekly without releasing the underlying data to back it up, which is a strange position for anyone lecturing mail marketers about measurement rigor. Every channel has a black box right now. The question is which black box you can at least partially open.

A Framework for 2027 Planning

1. Start with a holdout test on your existing house list, not a new prospect file. A 90/10 split against a lapsed segment tells you incremental lift within one quarter.
2. Set the bar at the category benchmark, not folklore. If your test doesn't clear roughly 2% to 3% response on a cold list or 4% to 5% on a warm one, the math likely won't beat email or paid social for that segment.
3. Budget for design and personalization, not just postage. A generic postcard performs like a generic email; the lift comes from the same targeting discipline you'd apply to a paid media buy.
4. Review your martech stack before adding a mail vendor. If your CRM can't trigger a mail send off a lapsed-customer flag or sync with a fulfillment partner, fix that plumbing first; our martech stack audit guide walks through exactly this kind of gap analysis before budget season.
5. Reassess annually, not quarterly. Mail's cost structure and response curve move slowly compared with digital, so judge it on a full campaign cycle, not a single month's numbers.

The Bottom Line for 2027 Budgets

Direct mail isn't replacing email, and any CMO pitching it that way is selling something. What's actually happening is a rebalancing: AI made email cheap to produce and, in the process, made a good chunk of it worthless to the recipient, while the mailbox stayed exactly as scarce and trustworthy as it's always been. That scarcity is now worth paying for again, in the segments where the math supports it.