Thanksgiving Instead of Christmas: Who It Works For and Who It Doesn't

In December, your card lands in a stack of forty. I have watched this from the printing side for a long time, and the arithmetic never changes: everybody sends in December, which means December is the one month your card is guaranteed to be compared to somebody else's.

So here is the contrarian case, and then the honest case against it, because this does not work for everyone and I would rather you not waste the postage.

The mailbox got emptier, and that changed the math

American households received 637.3 pieces of mail in 2025, down from 1,040.6 in 2015 — a 39% decline in a decade, according to the USPS Household Mail Survey — roughly 12,000 recruitment households and about 5,000 completed one-week mail diaries, address-based sampling weighted to Census, administered by NORC at the University of Chicago, fielded October 2024 through September 2025 and published March 2026. USPS funds the survey, which is worth saying out loud, though the fieldwork is NORC's.

Here is the part that matters. As volume fell, attention rose. Households report reading 51% of the advertising mail they received in 2025, up from 46% in 2020 and 47% in 2015. USPS's own explanation is the plain one: "the decline in volume has allowed each individual piece of advertising mail to gain more attention."

Cards specifically have gone scarce. Households sent 11.0 greeting, holiday and birthday cards in 2025, down 38% from 17.8 in 2015. Fewer people are sending. Which means the ones who do are no longer one of many.

That scarcity is real in November and it is not real in December. December is the one month the old volume comes back.

The relationship is the whole advantage

The single most useful number in that survey is this one: mail sent to someone the business already had a relationship with was around twice as likely to be read — 64% of pieces versus 36% — as mail sent cold.

Your clients are not a cold list. They are the 64%. A Thanksgiving card to a past client is the most favorable possible version of a piece of mail: known sender, uncrowded month, no ask attached.

Surprise is the mechanism, and it works hardest at a distance

There is good research on this now, and none of it is funded by anybody selling paper.

Liu, Rim, Min and Min published "The Surprise of Reaching Out" in the Journal of Personality and Social Psychology in 2023. It is pre-registered, and it has been independently replicated at n=742 — which is rarer than it should be. The finding: people badly underestimate how much others appreciate being reached out to — and the effect is magnified between people who are socially distant rather than close.

Read that again with your client list in front of you. The advisor and the client seen once a year. The agent and the buyer from four summers ago. That distance is not the obstacle. It is the reason it lands.

And a Thanksgiving card is a surprise in a way a December card structurally cannot be. In December you are expected. In November you are not.

"It'll feel strange." No, it won't.

This is the objection I hear most, and it is the one the research answers most cleanly.

Kumar and Epley studied exactly this in "Undervaluing Gratitude" (Psychological Science, 2018) — 107 letter writers and 80 recipients, using real gratitude letters sent to real people, plus two further experiments. Writers predicted recipients would feel awkward at 2.95 on their scale. Recipients actually reported 1.95. Writers predicted recipient positivity at 3.11. Recipients reported 4.12.

Every number moved the same direction. The sender expects it to land worse than it does. That gap is not a small effect and it is not a marketing claim — it is why people talk themselves out of sending, every year.

Who this genuinely does not work for

Four cases, and I mean these.

Your list is not all American. Thanksgiving is a US holiday. Canada's falls in October. To a client in London it reads as a slightly confusing American thing. If a meaningful slice of your list is overseas, a Thanksgiving-only program leaves those people with nothing.

You already have a December tradition. If your clients have received your card every December for nine years, that card is a signal, and the signal is reliability. Trading a known signal for a clever one is a bad trade. Add Thanksgiving if you like. Do not swap.

You are planning to send Thanksgiving and then nothing. If every competitor sits on that desk in December and you do not, you have bought novelty and sold presence in the month when renewals get decided. Thanksgiving instead of December is a real strategy. Thanksgiving instead of December by accident, because you felt done, is not.

November is your crunch. If your busiest four weeks of the year are the four weeks before Thanksgiving, the cards will go out rushed, unsigned, or late. A rushed card is worse than no card, because it says the relationship got whatever was left over.

The decision, in one line

If your clients are American, if every vendor they work with mails them in December, and if what you want to say is thank you rather than happy holidays — November is the better month and it is not close.

If any of those three is not true, send in December and send it well.

One practical note, since it is the thing people get wrong: a Thanksgiving card is not a December card with different artwork. The deadline is a month earlier, and that is the part that catches people out. I will lay out the working-backwards calendar next week, once you have had a chance to decide whether you are doing this at all.