The Keep List, The Pen Pass and The Postmark Plan — prepared by Edmund Bogen
When anyone can produce a warm message in seconds, the words stop being the hard part. What matters is whether people believe a person took the time. In seven preregistered experiments, people who believed an emotional marketing message was written by AI reported less positive word of mouth and less loyalty, and the penalty was smaller when AI only edited a message a person wrote (Kirk & Givi, Journal of Business Research, 2025). The problem is not that AI writes badly. A warm message loses its value when people think no one wrote it. This session gives you three systems for Dec 9 to Jan 8. The Keep List: who gets a handwritten note, a card, the farm piece or nothing, sphere first. The Pen Pass: your words first, AI tidies at most, then you write it out by hand, sign it and stamp it. The Postmark Plan: the year’s mail on one page, with a pre-send checklist. USPS’s First-Class date for Christmas delivery is Dec 17, eight days after the session, so a New Year card is the practical choice for most of what you send.
When anyone can produce a warm message in seconds, the words stop being the hard part. What matters is whether people believe a person took the time. In seven preregistered experiments, people who believed an emotional marketing message was written by AI reported less positive word of mouth and less loyalty, and the penalty was smaller when AI only edited a message a person wrote (Kirk & Givi, Journal of Business Research, 2025). The problem is not that AI writes badly. A warm message loses its value when people think no one wrote it.
We will master 3 systems live, together:
1 · The Keep List — Who gets what: a handwritten note, a card, the farm piece or nothing. By relationship, never by who people are.
2 · The Pen Pass — Your words first. AI may tidy, never write the warm part. Then your hand, your signature, your stamp.
3 · The Postmark Plan — The year’s mail on one page: Dec 17 or the New Year this season, and a ten-point check before anything is mailed.
Register for the SessionEvery page of the 51-page course study is on this page, in order — the PDF is the same content, formatted to print. Jump to any section:
Three systems, and the three habits underneath them: choosing who gets what, writing the warm part yourself, and checking every piece before it goes in the mail.
| # | Module | The skill it teaches | Pages |
|---|---|---|---|
| 1 | What the research actually says | Gratitude, AI authorship, effort, handwriting, relationship mail | 4–10 |
| 2 | The rules of mail | Mailbox, mail class, postage, EDDM, brokerage name, “sold”, fair housing | 11–17 |
| 3 | System 1: The Keep List | Hand, Card, Route or None, by relationship, rules attached | 18–20 |
| 4 | Writing the warm part | Specific, about them, in your words; where AI helps | 21–23 |
| 5 | System 2: The Pen Pass | Your words, a tidy, your hand, your stamp; robot pens and costs | 24–29 |
| 6 | System 3: The Postmark Plan | The calendar, Dec 17 or the New Year, the ten-point check | 30–34 |
| 7 | Your 30 days | Dec 9 to Jan 8, three counts, what to re-check | 35–36 |
| ✓ | Assessment and appendices | Answer keys, reference card, prompts, practice lines, questions, glossary, sources | 37–50 |
Six exercises sit where the skill was just taught. All use sample data, labelled as such: the people, addresses and notes are invented. Do them on paper first, then check Appendix A. The keys give the reasoning, not just the answer.
This session builds on this fall’s workbooks instead of repeating them. Use these when you need them.
| When you need… | Use this | From the workbook |
|---|---|---|
| The personal January message, its Jan 4–15 window, and the gift and RESPA rules | The January Callback; Lesson 2.8 | Holiday Networking (Dec 2, 2026) |
| The research on reaching out and on fresh starts | Lesson 1.8 (Liu et al. 2023; Dai, Milkman & Riis 2014; Kumar & Epley 2018 in one line) | Holiday Networking (Dec 2, 2026) |
| AI drafts that sound like you, and the honest answer if a client asks | The Voice Card; the honesty line | Own Your Voice (Sep 2, 2026) |
| Who in your database has gone quiet | The Monday Call List and its touch-neglect scoring | The Sunday Night Workflow (Nov 4, 2026) |
| Postcard art | The Landmark Library; the Style Card | Neighborhood Famous (Sep 16, 2026) |
| A fair-housing check on AI copy; what never goes into an AI tool | The Fair Housing Firewall; the Data Line | Keeping AI Compliant (Sep 23, 2026) |
| State AI disclosure laws; fake reviews and the FTC penalty | Module 3; the testimonial rule | AI in 2027 (Nov 25, 2026) |
| A payment or rate figure on a mailer; a lender who offers to pay for mail | The Reg Z lesson; RESPA in Lesson 2.5 | The Golden Handcuffs Conversation (Nov 18, 2026) |
| Mail to expired listings | The Article 16 guardrails | The Expired Goldmine (Aug 12, 2026) |
| Choosing a farm or a niche | The Niche Finder; the Referral Farm | Death of the Generalist (Sep 9, 2026) |
No earlier workbook teaches these: the mailbox rule (18 U.S.C. 1725); which mail class a handwritten note must use; 2026 postage and sizes; what EDDM can carry; the “sold” rule (SOP 12-7) and Florida’s brokerage-name rule in print; mailing lists and stop requests; and the research on warm messages people believe AI wrote. The live discussion at the end of The Price-Cut Conversation (Oct 14, 2026) turned to handwritten notes. This is the full version.
The core finding behind this session is about gratitude, not paper. It comes from experiments with real letters and real recipients.
Amit Kumar and Nicholas Epley asked people to write a letter of gratitude to someone who mattered to them, then surveyed the people who received it. Writers underestimated how surprised recipients would be, by the letter and by what was in it. They overestimated how awkward recipients would feel. And they underestimated how positive recipients would feel. Writing the letter also lifted the writer’s own mood.
| Experiment | Who wrote | Recipients who replied |
|---|---|---|
| 1 | 107 MBA students, in a voluntary class exercise | 80 of the 92 contacted (87%) |
| 2 | 100 lab participants, writing to anyone | 58 of 92 (63%) |
| 3a, 3b | 50 and 98 people choosing whom to write to | They chose recipients they expected to feel less awkward and more positive |
| 4 | 127 students | 78 of 96 (81%) |
Kumar & Epley, Psychological Science 29(9), 2018; full text read. The authors note that imperfect response rates could bias results; they found no difference in predictions between writers whose recipients did and did not reply.
In Experiment 4, writers predicted how recipients would rate them. Recipients rated the writers more competent, meaning more articulate, with words that were just right, than writers expected (9.28 against 6.99 predicted), and warmer (9.54 against 8.40). The gap was biggest on competence: writers worried about what they would actually write and how articulate they would be. The authors call that worry “an unwarranted barrier to expressing gratitude.” Their reading is that recipients focus more on a thank-you’s warmth than on its wording. The study compared predictions with ratings; it did not measure which one recipients weighed more.
Paper. Writers “wrote and sent” their letters, and in one experiment (3b) chose e-mail or a notecard the researchers mailed. The paper did not compare handwritten letters with e-mailed ones. Samples were students and lab participants, and we found no independent direct replication. Holiday Networking (Dec 2, 2026) listed this study in one line; this page is what is inside it.
The penalty attaches to the belief that AI wrote it, not to the quality of the words. When readers don’t know, AI’s words often score as well as a person’s.
| What the study found | Study | Limits |
|---|---|---|
| People who believed an emotional marketing message was written by AI gave less positive word of mouth and less loyalty. The penalty was weaker for factual messages and when AI only edited the message. | Kirk & Givi, J. Business Research, 2025; 7 preregistered experiments | Mostly hypothetical scenarios; online US and UK panels |
| People who disclosed that they used AI were trusted less, in 13 experiments. Being exposed by someone else hurt trust more than disclosing. | Schilke & Reimann, Org. Behavior and Human Decision Processes, 2025 | Workplace and professional settings |
| A label such as “This text has been edited by AI” reduced how sincere a thank-you message seemed. A 90-second video about AI in social messages, which mentioned AI holiday greetings, made people more suspicious of unlabelled messages. | Cao & Li, J. Computer-Mediated Communication, 2026; 2 preregistered studies | Brief vignettes; a teacher receiving a student’s thank-you |
| AI-written messages made people feel more heard than human-written ones, until they learned AI wrote them. | Yin, Jia & Wakslak, PNAS, 2024 | Not about mail |
| Third-party raters judged AI-written empathetic responses more compassionate than those of selected humans, expert crisis responders included. | Ovsyannikova, Oldemburgo de Mello & Inzlicht, Communications Psychology, 2025 | Raters were not the person the message was written to |
| When possible AI use was not raised, recipients showed no skepticism. Disclosing it hurt. | Zhu & Molnár, 2025 | Preprint, not peer-reviewed |
One study up close, from the authors’ preprint. In Kirk & Givi’s first study, 300 US participants imagined buying weights at a local store and getting a warm follow-up e-mail from the salesperson. In the AI version, a line under the signature said the letter was prepared by ChatGPT. Positive word of mouth was 4.11 with the label against 5.45 without; loyalty was 5.02 against 5.85. In their third study, a condolence e-mail that a person wrote and AI edited did better than AI-written versions. These figures come from the OSF preprint. We could not open the published paper’s full text; its abstract reports the same direction.
Why the belief matters: people read effort as care. Take the effort away, or seem to, and the message says less about you.
| What the study found | Study | Limits |
|---|---|---|
| When a friend used AI help for a support message, advice or a birthday celebration, people saw less effort, felt less satisfied with the relationship and more uncertain. Getting help from another person was not significantly different from using AI. | Liu, Kang & Wei, J. Social and Personal Relationships, 2024; N = 208 | Hypothetical scenarios with a fictional close friend |
| Help given with AI assistance was seen as less warm, created less sense of obligation, and made people less likely to return the favor. | Dishop, Brown, Chao, Kuznetsov & Woolley, J. Business and Psychology, 2025 | Coworker setting |
| An “AI-assisted” label weakened what a message said about the sender’s character. An AI-assisted apology made the sender seem less warm. | Khadpe, Wenzel, Loewenstein & Kaufman, AAAI/ACM AIES, 2025; N = 399 | Two vignette studies |
| Using one of three AI tools on an apology did not reduce how authentic it seemed. Heavier use did. | Glikson & Asscher, Computers in Human Behavior, 2023 | Read via a university research-portal summary |
In 2004, Kruger and colleagues found people rated a poem and paintings as better, and worth more money, when told they took longer to make (J. Experimental Social Psychology; student samples). Two preregistered replications with 1,405 online participants were mixed: the paintings result held with smaller effects; the poem result failed on one panel and, on the other, held for liking but not for money (Ziano et al., Collabra: Psychology, 2023). The fair summary: people tend to rate things higher when they know effort went in, more weakly than first reported, and least reliably on price.
A related finding: websites that showed the work they were doing were valued more, even with longer waits for the same results, and people felt a pull to reciprocate (Buell & Norton, Management Science, 2011; simulated online services). “I pulled the three sales on your street myself” is the mail version. Applying it to mail is our inference.
The largest preregistered field test on this page found handwritten thank-you cards did no better than printed ones. The details explain why.
In May 2023, 1,132 past donors to a German opera house got a fundraising letter with a postcard thanking them for past gifts. Every handwritten card was written individually, with a pen that needed some pressure, and deliberately imperfect lettering. It made no measurable difference. The printed cards in fact brought the larger average gift (€114.19 against €86.50) and more money per letter (€21.30 against €16.72), though neither gap was significant. In a separate test, adding “Thank you in advance” to 8,800 letters made no difference either (1.8% gave with the phrase, 1.6% without; not significant). The authors’ explanation, which they did not test: the cards “were not perceived as personal because they were sent by an institution rather than by a genuine peer” (Adena, Huck & Neyse, Experimental Economics, 2025; full text read).
| What the study found | Study | Read it as |
|---|---|---|
| A handwritten sticky note asking for help raised survey returns: 76%, against 48% for the same handwritten message on the cover letter and 36% with no note. It also beat a blank sticky note. | Garner, J. Consumer Psychology, 2005; numbers via The Decision Lab summary | The form of a personal favor mattered, not just the ink. Surveys, not purchases. |
| A server writing “thank you” on the check was followed by higher tips (16% to 18%). Adding her first name added nothing. | Rind & Bordia, J. Applied Social Psychology, 1995; numbers via Lynn (Cornell) | Thanks versus nothing, not handwriting versus print. One server. |
| Handwriting-style fonts made products feel more human and better liked. The effect shrank for brands people were already attached to, and reversed for functional products. | Schroll, Schnurr & Grewal, J. Consumer Research, 2018 | Fonts, not real handwriting. Products, not notes. |
| A handwriting-style font on charity appeals raised willingness to take part, more so for viewers not strongly connected to the brand. | Chu, Tok, Zhou & Chen, Psychology & Marketing, 2023 | Fonts on ads |
We found no controlled study that compares real handwriting, a handwriting font and plain print in one experiment.
USPS’s own survey shows a wide gap between mail to existing customers and mail to prospects. It is the strongest number in this book for sphere first.
| USPS Marketing Mail, FY2025 (quarters 2–4) | Read | “Very likely to respond” |
|---|---|---|
| Sent to existing customers | 64% | 25% |
| Sent to prospects | 36% | 3% |
| All recipients | 50% | 13% |
USPS Household Mail Survey FY2025 (formerly the Household Diary Study), prepared for USPS by RCF Economic & Financial Consulting, dated March 2026; Table E.5 / 5.5. About 5,200 households a year; the survey was redesigned in January 2025, so FY2025 mixes old and new methods. USPS paid for it and earns money from mail. “Very likely to respond” is what households said, not what they did. Not about real estate.
| Also from the FY2025 survey | Figure |
|---|---|
| Advertising pieces read, all ad mail (46% in 2020; 47% in 2015) | 51%; 10% set aside; 39% thrown away unread |
| Ad mail read by households earning $150,000 or more | 42% (about 58% under $50,000) |
| Greeting, holiday or birthday cards a household sent in 2025 | 11.0, down from 17.8 in 2015 (−38%) |
| Personal letters or cards a household received in 2025 | 19.5 (24.9 in 2015) |
| Correspondence received: homeowners against renters | 105 against 68 pieces |
| All mail received or sent per household | 637 pieces (1,040.6 in 2015) |
In 1976, two researchers mailed Christmas cards to strangers, and about 20% sent one back, as a later paper describes it (Kunz & Woolcott, Social Science Research). When a 2016 study mailed cards to 755 people from a supposed stranger, 2% reciprocated, and a follow-up survey linked lower reciprocation with suspicion (Meier, J. Social Psychology). A card from a stranger can read as a reason to be wary.
The most-quoted mail statistics come from studies commissioned by postal operators. They are worth reading. Say who paid when you quote them.
| What it found | Who did it | Read it as |
|---|---|---|
| People said they liked postcard ads and e-mail ads about equally, and their stated willingness to pay did not differ. A week later they remembered where they had seen postcard ads more accurately and confidently. Brain value regions were more active for those items. | USPS Office of Inspector General with Temple University, Report RARC-WP-15-012, June 15, 2015; 56 people analyzed, then 39 in the scanner | Government-commissioned, not peer-reviewed. Small lab sample; brain results at an uncorrected threshold. USPS: ad mail was 31% of its revenue in FY2014. |
| “Direct mail requires 21% less cognitive effort to process than digital media”; unaided brand recall 75% after mail against 44% after a digital ad. | Canada Post, with True Impact Marketing, July 31, 2015; 270 people, 30 per format, mock brands | Commissioned by a postal operator; the vendor’s own metrics; no significance tests reported |
| People valued digital versions of goods, such as souvenir photos, books and films, less than physical versions. | Atasoy & Morewedge, J. Consumer Research, 2018; 5 experiments | Products, not messages. A printed closing photo is closer to what was studied than a card. |
| Merely touching an object raised people’s sense of owning it. | Peck & Shu, J. Consumer Research, 2009 | Handheld objects. Anything about houses is a leap. |
| A texture in a communication persuaded more when it connected to the message. | Peck & Wiggins, J. Marketing, 2006 | Abstract only; field design not found |
| Heavy clipboards made job candidates seem more important. | Ackerman, Nocera & Bargh, Science, 2010 | Failed a preregistered replication (Selker, Beek et al., 2017) |
The honest version of this session’s premise is narrower than the slogan. Here is the edge of the evidence.
| A line you can say | Source |
|---|---|
| “People who write thank-you letters underestimate how good the recipient will feel.” | Kumar & Epley, Psychological Science 29(9), 2018 |
| “Thank-you writers worried most about their wording, yet recipients rated their letters more articulate, and warmer, than the writers expected.” | Kumar & Epley, 2018, Experiment 4 |
| “When people believe AI wrote a warm message, they respond worse. When AI only edits your draft, the penalty is smaller.” | Kirk & Givi, J. Business Research, 2025 |
| “In one big field test, handwritten thank-you cards from an institution did no better than printed ones.” | Adena, Huck & Neyse, Experimental Economics, 2025 |
| “Households said they read 64% of ad mail from businesses they deal with, and 36% from others.” | USPS Household Mail Survey FY2025 (USPS-funded) |
| “Households sent 11 greeting, holiday or birthday cards in 2025, down from 17.8 in 2015.” | USPS Household Mail Survey FY2025 |
Six lines you might hear at an office meeting this month (sample wording). For each, write Keep, Fix or Drop, and then the version you would say. Use only facts in this module.
Answer key: Appendix A, page 38.
A federal criminal statute and the USPS mail manual say the same thing: nothing without postage goes in, on or hanging from a mailbox.
The USPS mail manual is broader. Mail receptacles “may be used only for matter bearing postage,” and nothing without postage may be “placed upon, supported by, attached to, hung from, or inserted into a mail receptacle.” Anything found there “is subject to the same postage as would be paid if it were carried by mail” (DMM 508.3.1.3). And “Any advertising on a mailbox or its support is prohibited” (DMM 508.3.2.5).
| The piece | In or on the mailbox? | At the door? |
|---|---|---|
| A thank-you note you deliver by hand, no stamp | No | Yes |
| A door hanger | No, and not hung from the box or its post | Yes. Check HOA and city rules first. |
| A one-page market report | No | Yes, or stamp it and mail it |
| A stamped card that went through USPS | Yes. That is mail. | — |
We did not research HOA or city rules on door hangers, and did not compute a dollar figure for a fine.
| The rule | Source |
|---|---|
| “Mail containing handwritten or typewritten material must be mailed as First-Class Mail, USPS Ground Advantage — Retail, Priority Mail, or Priority Mail Express.” | DMM 133.3.4 |
| Mail with “personal information,” meaning “any information specific to the addressee,” must go by the same classes. | DMM 133.3.3 |
| EDDM-Retail may not carry handwritten or typewritten matter, or personal information. | DMM 143.3.4, 143.3.2 |
| Marketing Mail may not carry handwritten or typewritten matter, except limited additions such as “Do Not Open Until Christmas” and “Happy Birthday, Mother.” | DMM 243.2.4, 243.2.6 |
So a real handwritten note travels First-Class with a stamp. It cannot ride in an EDDM farm drop or a bulk Marketing Mail run. A personal line on a bulk postcard (“Great seeing you at the block party, Maria”) turns it into personal correspondence. We found no USPS ruling on whether robot-pen script or a handwriting font counts as “handwritten matter,” so if you use a service, ask the vendor and USPS which class applies.
Stamp prices rose on July 12, 2026. Shape and contents can add a surcharge. Say the date whenever you quote a price.
| The piece | Postage | Note |
|---|---|---|
| Letter or card in an envelope, 1 oz (Forever stamp) | $0.82 | Was $0.78 before July 12, 2026 |
| Each additional ounce | $0.29 | A 2-oz stamped letter is $1.11 |
| Postcard, rectangular, card size | $0.65 | Was $0.61 |
| Square card, or one with a ribbon, clasp, coin, pen, magnet or other lump | $0.82 + $0.49 = $1.31 | Nonmachinable surcharge |
| Large envelope (flat), 1 oz | $1.69 | |
| Metered letter, 1 oz | $0.78 | Office postage meters |
| EDDM-Retail flat, up to 3.3 oz | $0.26 | Postage only, before printing |
USPS news release, Apr 9, 2026 (prices effective July 12, 2026; the Postal Regulatory Commission approved them on May 27, 2026, per UMass Lowell, June 1, 2026). USPS Notice 123, page header “Effective October 04, 2026”; the Oct 4 change touched parcels only. A book of 20 Forever stamps is $16.40 (our arithmetic).
We found no USPS filing for a January 2027 stamp increase. In July 2026 USPS asked the Postal Regulatory Commission for a waiver so it could move to a January cycle (Docket R2027-1). Trade press reports the Commission denied it on Aug 21, 2026 (PRINTING United Alliance, Aug 25, 2026). The next increase is expected in mid-2027 at the earliest. We read industry reports, not the order itself, and those sources have an interest in mail volume.
Every Door Direct Mail reaches whole carrier routes without names. Your own lists carry their own rules.
| EDDM-Retail rule | Source |
|---|---|
| Flats only: longer than 10½ inches, or higher than 6⅛ inches, or thicker than ¼ inch; no more than 3.3 ounces. | DMM 143.2.5 |
| At least 200 pieces (or 50 pounds) and up to 5,000 pieces per day per 5-digit ZIP Code. Fewer is allowed only when the whole ZIP has fewer than 200 deliverable addresses. | DMM 143.2.5 |
| Entered at the Post Office that serves the routes, in bundles of 50 to 100 with a facing slip, sorted by route from the EDDM tool’s counts. | DMM 143.2.5; 145.1.3 |
| Addressed to “Postal Customer.” No forwarding or return service. | usps.com EDDM page; DMM 143.2.3 |
| Nothing personal and nothing handwritten. | DMM 143.3.2, 143.3.4 |
The EDDM online tool can filter routes by age, income and household size from Census data. Our advice: leave the filters off and mail whole routes you chose by geography. Choosing locations so that groups are denied housing information because of a protected class is prohibited (24 CFR 100.75(c)(3); page 15).
| Situation | What applies | Source |
|---|---|---|
| A Florida parcel shows no owner name | The owner may hold a protected-address exemption; the appraiser removes the name, not the address. Address the piece to “Current Resident” and leave the record alone. | Fla. Stat. 119.071 (2026) |
| Mailing owners named in county records | Florida’s Digital Bill of Rights reaches only “Controllers” with more than $1 billion in revenue, and excludes information lawfully available through government records. | Fla. Stat. 501.702 (2026) |
| A list bought from a data broker | Since Aug 1, 2026, data brokers must check California’s deletion system at least every 45 days. Ask your vendor how it handles deletions and opt-outs. | Cal. Civ. Code 1798.99.86(c), per CPPA |
| “Is there a Do Not Mail list?” | We found no federal one. DMAchoice is a voluntary service of the ANA, a marketers’ trade group. PS Form 1500 orders cover only sexually oriented ads. | FTC (Jul 2022); dmachoice.org; 39 U.S.C. 3008 |
| Anyone asks you to stop mailing | Flag them in your CRM that day and take them off every list, your vendor’s included. | Our advice |
Before any mailing, check your addresses for moves. USPS requires lists for presorted First-Class and Marketing Mail prices to be updated within 95 days before mailing, NCOALink being one method (PostalPro, read via a search excerpt). AI cannot fix a stale address list. E-mail follows different rules: CAN-SPAM opt-outs apply only to e-mail, as Holiday Networking (Dec 2, 2026) taught. Other states’ privacy laws were not reviewed.
Florida’s advertising rule and the NAR Code both reach printed mail. A brand or team name alone is not enough.
| Rule | What it means for mail | Source |
|---|---|---|
| QR codes and web pages | A page your postcard points to follows Florida’s internet rule: the brokerage name “adjacent to or immediately above or below the point of contact information.” | 61J2-10.025(3)(a) |
| Discipline | A licensee may be disciplined for advertising that is “fraudulent, false, deceptive, or misleading in form or content.” | Fla. Stat. 475.25(1)(c) |
| Team names, phone numbers | We found no Florida rule on team names; workshops in 2016 and 2017 adopted nothing. The 2007 amendment removed the phone-number requirement. | Rule effective 2/5/2007 |
| Another state: Texas | The broker’s name “in at least half the size of the largest contact information” for any agent or team in the ad. Communications to a current client are excluded. | 22 TAC 535.155 (LII mirror, amended 9/4/2018) |
| NAR: the firm’s name | In any medium, print included, disclose your firm’s name “in a reasonable and readily apparent manner.” | SOP 12-5 |
| NAR: true picture | Be honest and truthful, present “a true picture,” and make your status as a real estate professional apparent. | Article 12 |
| NAR: other lines | Do not describe your services as costing clients nothing unless you receive no compensation from any source (12-1). Spell out a giveaway’s terms in advance (12-3). Advertise property only with authority (12-4), and only credentials you hold (12-13). | SOPs 12-1, 12-3, 12-4, 12-13 |
A neighbor’s sale handled by another brokerage cannot appear as “we sold.” You may report it as a market fact, “a home on your street closed at $X (source: MLS),” if your MLS rules allow; we did not check MLS rules on sold data in print. A handwritten note to a past client is safest with the brokerage on the card or the envelope (our advice). The NAR Code quoted here is the 2026 Code; the 2027 Code was not posted on Nov 3, 2026, so re-check it before you print in January.
A mailer is a “flyer” under the fair housing rules. Choose the area by geography, and keep people out of the pictures.
| Rule | What it says | Source |
|---|---|---|
| No preference in ads | Unlawful to make, print or publish any notice, statement or advertisement about the sale or rental of a dwelling that “indicates any preference, limitation, or discrimination based on race, color, religion, sex, handicap, familial status, or national origin.” | 42 U.S.C. 3604(c) |
| Mail counts | Written notices include “flyers, brochures… or any documents used with respect to the sale or rental of a dwelling.” | 24 CFR 100.75(b) |
| Words and pictures | “Using words, phrases, photographs, illustrations, symbols or forms which convey that dwellings are available or not available to a particular group of persons” because of a protected class. | 100.75(c)(1) |
| Where you mail | “Selecting media or locations for advertising… which deny particular segments of the housing market information about housing opportunities” because of a protected class. | 100.75(c)(3) |
| REALTOR® members | The same, with sexual orientation and gender identity added. | SOP 10-3 (2026 Code) |
Where HUD guidance stands. HUD withdrew its April 29, 2024 guidance on digital advertising, effective Sep 17, 2025: it “should not be relied upon as authoritative,” and actions that do not comply with the Act “continue to be subject to enforcement” (91 FR 17291, Apr 6, 2026). The 1995 Achtenberg advertising memo is not on that withdrawal list, but its old HUD web address no longer works, so its status is unclear. We found no guidance on human models in ads in force. The withdrawn guidance never changed the law.
A “general mailing or distribution addressed to all prospects in a given geographical area” is allowed even if some recipients are listed with another REALTOR® (SOP 16-2). What Article 16 targets is written solicitation of owners whose properties are listed with another REALTOR®, aimed at them through current listings, signs or MLS information (the paragraph after SOP 16-2), and soliciting a listing that is currently exclusive with another broker (SOP 16-4). So for any targeted piece, such as a note to selected owners, an AI-built “likely to sell” list or a home-value offer, remove active listings first. The Expired Goldmine (Aug 12, 2026) and The Golden Handcuffs Conversation (Nov 18, 2026) already teach this filter.
Four questions that come up as soon as AI and vendors touch your mail. Where no specific rule exists, we say so.
| Question | What we found | Source |
|---|---|---|
| Can a robot pen imply I wrote it? | No specific rule. We found no FTC rule, guidance or case, and no state law, on machine-written “handwritten” mail. The general test: the FTC finds deception in “a representation, omission or practice that is likely to mislead the consumer acting reasonably in the circumstances, to the consumer’s detriment,” judged by “the net impression.” Florida’s FDUTPA gives FTC interpretations “great weight.” | FTC Policy Statement on Deception (Oct 14, 1983); Fla. Stat. 501.204 (2026) |
| Can AI polish a client’s quote? | Not into anything they did not say. A business may not “write, create, or sell” a consumer testimonial that materially misrepresents that the person exists, used the service, or had the experience described. Use their words as given, with permission. A quote from your own team member needs that relationship disclosed. Civil penalties run up to $53,088 per violation (16 CFR 1.98(d)), as AI in 2027 taught. | 16 CFR 465.2(a), 465.5(b)(1); 1.98(d) |
| Must I label AI-drafted mail? | We found no state law requiring it for printed letters, cards or postcards. The laws taught in AI in 2027 cover chatbots, AI “synthetic performers” in ads, and decisions. Your broker’s AI policy still applies. | AI in 2027, Module 3 (not re-verified here) |
| A lender wants its name on my postcard | Each party pays its own share at fair market value. A partner paying for your mail is a thing of value, and the CFPB’s FAQs name paying an agent for “direct mail campaigns” it does not perform, or above market value, as prohibited. | 12 CFR 1024.14(b), (g)(2); CFPB RESPA FAQs (updated Oct 7, 2020) |
Where this book stops. Cost-sharing with lenders, title companies and local businesses is the subject of Split the Bill (Jan 6, 2027, planned). RESPA and gifts are in Holiday Networking (Dec 2, 2026) and The Golden Handcuffs Conversation (Nov 18, 2026, Lesson 2.5). Any payment or rate figure on a mailer goes through The Golden Handcuffs Conversation first. As it put it: “The same number can escape the ad rule in a meeting and trigger it on a postcard.”
Keep this page by the stamps. Every line points back to its lesson.
| When | The rule in one line | Page |
|---|---|---|
| The mailbox | Only stamped mail goes in, on or hanging from it. Hand-delivered pieces go to the door. | 11 |
| Handwritten or personal | First-Class with a stamp. Never EDDM, never Marketing Mail. | 11 |
| Postage | As of Oct 4, 2026: card 82¢, postcard 65¢ (card size only), square or lumpy $1.31. | 12 |
| EDDM | Flats, at least 200 pieces a mailing and up to 5,000 per ZIP per day, whole routes, nothing personal. | 13 |
| Brokerage | The licensed brokerage name on every piece of advertising (Florida); the firm name in any medium (NAR). | 14 |
| “Sold” | Only on deals you were on as listing or cooperating broker; before closing, the listing broker’s consent. | 14 |
| Fair housing | Whole areas by geography. Homes, streets and landmarks in the pictures, not people. | 15 |
| Article 16 | General mail to an area is fine. Targeted pieces skip owners listed with another broker. | 15 |
| Honesty | Real quotes, with permission. A machine’s writing never claims your hand. | 16 |
| Stop requests | Flag that day, on every list, vendors included. | 13 |
Six mail situations (sample data). For each, name the rule and say what you would do instead.
Answer key: Appendix A, page 38.
The Keep List decides who gets which kind of mail before anyone writes a word. Sphere first. Strangers last.
| Tier | Who (sample) | What they get | How it travels | Rules that ride along |
|---|---|---|---|---|
| Hand | Clients you closed with this year; anyone who sent you business; people you would call friends | A note you wrote yourself, in your hand | First-Class, stamped. 82¢ for a rectangular card. | Real memories only (Module 4). Brokerage on the card or envelope (our advice). |
| Card | Past clients from other years; the rest of your sphere | A card you sign by hand, with one handwritten line; or a service card in your words that makes no claim to be your hand | First-Class, stamped. A handwritten line requires it; ask the vendor about service cards. | DMM 133.3.4. Owners listed with another broker come off this tier for anything about your services. |
| Route | Every address on the farm routes you chose | One identical piece, no names | EDDM-Retail ($0.26), or a saturation Marketing Mail run | Nothing personal or handwritten. Brokerage name. “Sold” only for your deals. Whole routes by geography. |
| None | Anyone who asked you to stop; strangers you have no reason to mail | Nothing | — | A do-not-mail flag on every list, your vendors’ included. |
You do not need a new database. You need four labels on the one you have, and the rules that ride with them.
Start with the Monday Call List from The Sunday Night Workflow (Nov 4, 2026). Its scoring already includes “touch neglect,” which catches people on a mailing list who have had no real contact. Use it to find the quiet past clients; do not rebuild the scoring here. Then add one field: Hand, Card, Route or None.
As big as the notes you can write in your own hand, with real memories, by the date you need them. Count your hours, not your contacts. Practitioners have taught versions of this for years, with no statistics behind them: Tom Ferry’s 2007 lead-generation book gives each past client or center of influence “3 personal notes (handwritten by me personally),” a holiday card and a New Year letter a year, beside newsletters and branded postcards, and tells agents to “send out a handful of handwritten notes each day.” Buffini & Company’s 2018 “Personal Notes” sheet lists when to write: “After you’ve received a referral,” “To re-establish contact,” “To simply make someone’s day.” Holiday Networking quoted The Close (“nothing beats a handwritten note to garner attention”) and kept it for a select group, not the whole list.
Active listings. Owners listed with another broker come off the Hand and Card tiers for anything about your services (NAR Article 16; SOP 16-4). A general Route piece to their whole route is fine (SOP 16-2).
Route tier by geography. Whole routes around your listings or your farm, chosen for your business reasons. Never by who lives there (24 CFR 100.75(c)(3)).
Stop requests. Anyone who asked you to stop is None, on every list, vendors included.
Florida parcels with no owner name. “Current Resident,” and leave the record alone (Fla. Stat. 119.071).
Addresses. Checked for moves before you mail. AI cannot fix a stale list.
No sorting fields for age, family, faith, health, national origin or disability. If your CRM has one, do not use it to tier.
Referral partners. A thank-you note to a lender, inspector or title rep is fine. Anything of value tied to referrals is not: RESPA has no small-gift exception (Holiday Networking, Lesson 2.8, covers the gift rules).
AI and the list. Prompt 1 (page 42) sorts an export by relationship fields only. Follow The Data Line (Keeping AI Compliant, Sep 23, 2026): replace names with ID numbers or role labels (“Buyer A”) before anything goes into an AI tool, and never paste an exact address or anything that pairs an address with a person.
Sample data: ten rows from an invented agent’s database, each with the reason for its tier.
| Contact (sample) | What you know | Tier | Why |
|---|---|---|---|
| A | Bought with you in 2025; you talk every few months | Hand | A close, recent client |
| B | Sent you a buyer in October | Hand | A referral this year. A note, nothing of value. |
| C | Sold with you in 2019; no contact since | Card | Gone quiet; flagged by touch neglect. Hand, if you remember them well. |
| D | An inspector you work with often | Hand | A referral partner. A note only. |
| E | Past client; home now listed with another broker | None for now | Article 16. Ask your broker about a purely personal greeting. |
| F | Asked you in March to stop mailing | None | Do-not-mail flag, everywhere |
| G | Signed in at an open house in May; one conversation | None | Close to a stranger. The Route piece reaches them if they live on your routes. |
| H | A neighbor you know by name | Card | Sphere |
| Oak Lane routes | 1,100 addresses around your listing | Route | Whole routes, chosen by geography |
| Parcel, no owner name | On your farm, Florida | Route | “Current Resident” |
Eight entries from a sample database. Give each a tier, Hand, Card, Route or None, and the reason. Name any rule that applies.
Answer key: Appendix A, page 39.
Your words land better than you fear. Warmth is easier to write than you think, and harder to fake than AI makes it look.
| What the research says | Study | How it shows up in a note |
|---|---|---|
| Writers underestimated how articulate and how warm recipients would find them, most of all on wording. | Kumar & Epley, Psych. Science, 2018 | Plain words are fine. |
| In couples’ conversations, praising the partner was linked with the partner seeing them as responsive, feeling good and feeling loved. | Algoe, Kurtz & Hilaire, Social Psych. and Personality Science, 2016; 370 couples; correlational | “You were so patient through three inspections” is closer to this than “Thanks for your business.” An association, not proof. |
| The giver’s thoughtfulness counted only when something prompted receivers to think about it. | Zhang & Epley, J. Exp. Psych.: General, 2012 | A specific detail makes the thought visible (our inference). |
| Showing the work raised how much a service was valued. | Buell & Norton, Management Science, 2011 | “I pulled the three sales on your street myself” (our inference for mail). |
No pitch in the warm part. This is our rule, not a finding: a thank-you note is a thank-you. Buffini’s list of reasons to write includes “To simply make someone’s day.” If you have business to raise, raise it another day, by a channel the person agreed to. The January Callback (Holiday Networking) is that channel for the people you met in December.
Sample wording. Each Better version uses only a memory the writer actually has, written in the writer’s own voice.
In a field experiment and lab chats, conversations by phone or video built stronger bonds than e-mail or text, with no more awkwardness, yet people’s expectations still led them to prefer text (Kumar & Epley, J. Exp. Psych.: General, 2021). Paper was not tested, and a note followed by a call was not tested either; the combination is only consistent with the research. Practitioners pair them: Buffini teaches a handwritten note as a follow-up to a call, visit or mailing, and Tom Ferry’s 2007 book recommends following postcards with a phone call or door contact.
People who did a small, unexpected kindness, like handing out hot chocolate in a park, underestimated how positive recipients would feel (Kumar & Epley, J. Exp. Psych.: General, 2023). Compliment-givers underestimated how good the recipient would feel and overestimated how uncomfortable (Boothby & Bohns, Personality and Social Psychology Bulletin, 2021). The same pattern runs through this whole module: senders hold back because they expect awkwardness that recipients do not feel.
The barrier to a thank-you note is starting, and remembering what to say. AI can help with both, as long as the memory and the words stay yours.
| The job | AI may help? | Why |
|---|---|---|
| Reminding you who is due a note: closings, referrals, people gone quiet | Yes | Getting started is the practical barrier (our inference; the study found the worry is about wording, page 4). Lists and dates are factual work. |
| Asking you questions so you remember a moment | Yes, questions only | Prompt 2 asks; you answer. The memory must be yours. |
| Tidying your draft: spelling, length, a clumsy sentence | Lightly | A message a person wrote and AI edited held up best of the AI options tested (Kirk & Givi, 2025). Prompt 3. |
| Writing the warm part | No | People respond worse when they believe AI wrote something personal (page 5), and you could not honestly say you wrote it. |
| Supplying a memory, a feeling or a detail | Never | It did not happen, or you do not know that it did. |
| Planning the year’s mail calendar | Yes | From dates you give it. Prompt 4. |
| Checking a farm piece or a card against your rules | Yes, as flags | It flags; you and your broker decide. Prompts 5 and 6. |
Five sample notes, each with a sample memory. Rewrite each as three sentences: the moment, the person, the wish. Use only what the memory gives you. Add nothing.
Answer key: Appendix A, page 39.
The Pen Pass keeps the warm part yours from the first word to the stamp. AI tidies spelling and clarity, at most.
| Why each step | The evidence | Source |
|---|---|---|
| Your words first | When people believe AI wrote an emotional message, word of mouth and loyalty drop. A message a person wrote and AI edited was the least-penalized AI option tested. | Kirk & Givi, 2025 (page 5) |
| Tidy lightly | Using one of three AI tools on an apology did not reduce how authentic it seemed; heavier use did. | Glikson & Asscher, 2023, via a research-portal summary |
| Not your assistant either | Help from another person looked no better than AI help in a relationship task. | Liu, Kang & Wei, 2024 (page 6) |
| Your hand | Handwriting alone did nothing for an institution’s cards; the authors think a card from a genuine peer might differ, which they did not test. A personal request on a sticky note helped in another study. | Adena et al., 2025; Garner, 2005 (page 7) |
| First-Class, stamped | Handwritten matter must go First-Class, or Ground Advantage Retail or Priority. | DMM 133.3.4 (page 11) |
If AI writes the warm part and you copy it out by hand, the ink is yours but the words are not. That is the belief that costs goodwill when people suspect it, and you could not honestly say “I wrote this.” With your words first, the honest answer is simple. Own Your Voice (Sep 2, 2026) gave you the line for AI-assisted work: “I write with AI, but every word is mine before it ships.” For a Pen Pass note you can say, “Yes, I wrote it.”
Six services, in their own words. Dated examples, not recommendations. None of them says how to tell recipients.
| Service | How the writing is made (their words) | What they say about recipients knowing |
|---|---|---|
| Handwrytten | “custom-designed handwriting robots that hold real pens” | Nothing. A testimonial says “You would never know that these weren’t actually handwritten.” |
| Simply Noted | “written by a robot holding a real ballpoint pen with real ink, not printed in a handwriting font” | FAQ: “recipients cannot tell it apart from a note written by hand.” |
| Thankster | Handwriting fonts: “system handwriting fonts,” or a custom font from your scanned writing | Nothing |
| Postable | Printed: “We print, stamp and mail”; “handwriting fonts so realistic…” | Nothing |
| Felt | Your own handwriting, entered in the app, then printed on card stock; “We hand apply a first class stamp.” | Calls it “100% authentic.” |
| thanks.io | “digitally rendered handwriting” / “Handwriting-style font” | Nothing |
Vendor claims, from each vendor’s own site or app listing, checked Nov 3, 2026. Not independently verified.
Several add AI writing. Handwrytten: “use our AI to perfect your message.” Simply Noted: “write your message (or let our AI draft it).” Thankster has an “AI Message” tool, and its own FAQ suggests keeping it to about 100 words and says you will likely want to edit the result: a vendor admitting the draft is not the final card.
Our arithmetic on published prices, checked Nov 3, 2026. Vendors change prices; check before you order.
| Method (dated example) | How it is written | Math | Total for 100 |
|---|---|---|---|
| Your own cards, Forever stamps | Your hand | 100 × $0.82 | $82.00 + cards and your time |
| Your own postcards, postcard stamps | Your hand | 100 × $0.65 | $65.00 + cards |
| Your own square cards | Your hand | 100 × ($0.82 + $0.49) | $131.00 + cards |
| thanks.io notecard, pay as you go | Handwriting-style font | 100 × $3.04 (Marketing Mail) | $304.00 |
| thanks.io notecard, First-Class upgrade | Handwriting-style font | 100 × ($3.04 + $0.21) | $325.00 |
| Simply Noted folded card | Robot pen | “100 credits for $362” | $362.00 |
| Felt | Your handwriting, printed | 100 × $4.50, postage included | $450.00 |
| Handwrytten, retail, plus postage | Robot pen | 100 × ($3.75 + $0.82) | $457.00 |
| Postable, 20–199 cards | Printed handwriting font | 100 × ($4.59 + $0.82) | $541.00 |
Sources: USPS Notice 123 (effective Oct 4, 2026); each vendor’s own pricing page or app listing, Nov 3, 2026. The cost of a box of cards is not included; we found no published retail price to use. Standard cards only: custom cards, gift cards, inserts and international postage are extra. The Handwrytten line assumes postage at the 82¢ USPS rate, which is how its plans bill, and shows no sign-up discounts. Simply Noted labels its 100-credit tier “Incl. $0.82/card postage” ($2.80 + $0.82 a card) and its FAQ calls package prices “all-in,” yet the same answer says postage is “$0.82 per card on top of the credit price.” We used the all-in $362. Felt’s price is “start at just $4.50 (including postage),” before volume credits.
Sample data. Sam’s note to a buyer who closed in August, taken through all four steps.
dana — thanks again for being so patient w/ the 3 inspections in aug. you never lost your cool even when the roof guy was late twice. hope the yard is everything you wanted and the first year in the house is a great one. sam
Before pasting into Prompt 3, Sam replaces “dana” with “Buyer A.”
| Check | Result | Fix |
|---|---|---|
| Every detail is mine | Yes: the inspections, the late roofer, the yard | None |
| Nothing was added | Yes. The tool listed every change and added none. | None |
| It sounds like me | “Roofer” is not how Sam talks. | Put “roof guy” back. |
| No pitch | Yes | None |
| It fits the card | Three sentences | None |
Sam copies it onto a rectangular card, writes Dana’s name back in by hand, changes nothing else, and signs it “Sam.” The card has Sample Realty Group printed on the back, and the return address on the envelope names the brokerage (our advice). One Forever stamp, 82¢ as of Oct 4, 2026. Into a USPS collection box, First-Class. Logged in the CRM: date mailed, Hand tier.
Dana, Sam and Sample Realty Group are invented for this example.
The kind of card a robot-pen app and its AI can produce from a CRM record. Find the problems before you read the table.
| # | The line | What is wrong | Page |
|---|---|---|---|
| 1 | “Dear Karen… Your friend,” to someone he has never met | A stranger written to as a friend. Strangers are last on the Keep List; cards from strangers can read as suspicious. | 18 |
| 2 | “I sat down by the fire tonight to write this… by hand” | A robot wrote it. It claims effort he did not make (our reading of the FTC standard). | 16 |
| 3 | “toasting your new patio with champagne” | Invented by the AI. They have never met. | 21 |
| 4 | “Thank you in advance for all your referrals” | An ask dressed as thanks. It had no effect on 8,800 fundraising letters. | 7 |
| 5 | “JUST SOLD: 22 Oak Lane” | Another brokerage’s sale. SOP 12-7. | 14 |
| 6 | “worth $85,000 more” | A number with no source; not a true picture (Article 12). | 14 |
| 7 | “— A Happy Seller” | An invented testimonial. 16 CFR 465.2. | 16 |
| 8 | “families like yours” | Wording that ties housing to family status. 24 CFR 100.75(c)(1). | 15 |
| 9 | Signed “Sam,” no brokerage anywhere | Florida requires the licensed brokerage name on advertising; NAR requires the firm. | 14 |
| 10 | Square envelope, a magnet, one stamp | Nonmachinable twice over: $1.31 at least, not 82¢. | 12 |
Six Pen Pass situations, all sample data. The four steps and the honest line decide most of them.
For each: pass it, fix it (say how), or don’t send it (say why).
Answer key: Appendix A, page 40.
| What you did | The true answer |
|---|---|
| The Pen Pass: your words, your hand | “Yes, I wrote it.” |
| Your words; AI fixed the spelling | “Yes. I had the spelling checked.” Or simply “Yes.” The words are yours. |
| A robot pen or font carried your words | “I wrote the words. A service printed or penned the card.” |
| AI drafted it and you edited | “I write with AI, but every word is mine before it ships” (Own Your Voice, Sep 2, 2026). |
Being exposed by someone else hurt trust more than disclosing it yourself (Schilke & Reimann, 2025). That is not a case for announcing every tool. It is a case for never needing to hide one.
Who gets what, and when it goes in the mail. One page, built from your own dates.
| When | Hand tier | Card tier | Route tier |
|---|---|---|---|
| Dec 21–22, 2026 | New Year notes | New Year cards, signed, one handwritten line | — |
| After each closing | A thank-you note | — | — |
| After each referral | A thank-you note. Nothing of value. | — | — |
| Each closing anniversary | A short note | A card | — |
| Quarterly (sample: Mar, Jun, Sep) | — | — | One identical piece to whole farm routes |
| When you list or sell (your deals only) | — | — | Just listed or just sold to the routes around it |
| Late Nov 2027 | Holiday notes, by USPS’s 2027 First-Class date | Holiday cards | — |
A sample plan, not a researched schedule. No study we found tests one calendar against another.
Practitioners differ. Tom Ferry’s 2007 plan gave each past client or center of influence a holiday card, a New Year letter, three personal notes, six newsletters and six branded postcards a year, and told agents to mail postcards to “200+ homes around the property after you list it, when it sells and after it closes,” followed up by phone or at the door. Buffini & Company’s system mails an “Item of Value” on the 1st of each month. Buffini’s claim that a phone call makes that mail “seven times” more effective has no source we could find.
USPS’s First-Class date for Christmas delivery is eight days after this session. For most of what you have not written yet, a New Year card is the practical choice.
| Date | What it is | Source |
|---|---|---|
| Wed, Dec 9 | This session | — |
| Thu, Dec 17 | Recommended First-Class send-by date for delivery by Dec 25, in the lower 48 and to Alaska, Hawaii and Puerto Rico. In the lower 48, Ground Advantage is also Dec 17. | USPS release, Sep 22, 2026 |
| Fri, Dec 18 · Sat, Dec 19 | Priority Mail · Priority Mail Express send-by dates | Same |
| Marketing Mail | No holiday date published. Do not count on a Marketing Mail card arriving by Christmas. | Same |
| Mon–Tue, Dec 21–22 | This book’s New Year mailing window, before the quiet week. Our practical choice, not a USPS date. | Ours |
| Wed, Dec 23 – Fri, Jan 1 | Holiday Networking’s quiet week: no outreach, and nothing sent on New Year’s Day. | Holiday Networking (Dec 2, 2026) |
| Mon, Jan 4 – Fri, Jan 15 | Holiday Networking’s January Callback. It follows the cards. | Holiday Networking (Dec 2, 2026) |
Military and international dates are on usps.com/holiday; we did not verify them. USPS delivery times are not promised here.
Why the New Year is the practical play. Eight days is not long to write real notes, and services that write and mail for you need production time on top. So write the warm part now and mail New Year notes on Mon–Tue, Dec 21–22, before Holiday Networking’s quiet week starts on Dec 23. Holiday Networking’s January Callback (Jan 4–15) follows. That timing is a practical choice. We found no study that compares a New Year card with a December card, or that tests when mail arrives.
What the data can say. In USPS’s quarterly results, October–December 2023 carried 12.4% more Marketing Mail than January–March 2024, but only 2.0% more First-Class Mail (our arithmetic on USPS releases of Feb 6 and May 9, 2025; we used 2023–24 because late 2024 was swollen by election mail). That is consistent with less advertising mail in the January quarter. It does not show December against January, or what reaches any one household. USPS does not publish monthly volume.
Fresh starts. Holiday Networking taught that goal-directed behavior rises after the New Year (Dai, Milkman & Riis, 2014). One study more: in a randomized mailing to 6,082 university employees in October 2013, framing a future savings date as a fresh start, such as a birthday or the first day of spring, raised later contributions; placebo dates did not; and 91% saw New Year’s as the beginning of a new cycle (Beshears, Dai, Milkman & Benartzi, Org. Behavior and Human Decision Processes, 2021). It tested wording, not arrival dates, and savings, not homes. On a Route piece, a line such as “If 2027 is the year you move, here is what sold on your street in 2026 (source: MLS)” borrows that fresh-start wording. Whether it raises response is untested.
Run it on every note, card and route piece before it goes in the mail. Each point names its rule and its page.
An AI tool can design and write a farm piece in minutes. This one fails the checklist on almost every point.
| # | The problem | Checklist point | Page |
|---|---|---|---|
| 1 | An AI-generated family in the photo | 6. Pictures show homes, not people (100.75(c)(1)) | 15 |
| 2 | “Perfect for growing families!” | 6. Wording tied to familial status | 15 |
| 3 | ZIP codes picked for “the most young families” | 6. Choosing locations by who lives there (100.75(c)(3)) | 15 |
| 4 | “Hi neighbor! — Sam” handwritten on an EDDM piece | 1. Handwritten matter may not go EDDM (DMM 143.3.4) | 11 |
| 5 | 50 extras slipped into mailboxes | 2. Nothing unstamped in a mailbox (18 U.S.C. 1725) | 11 |
| 6 | “JUST SOLD: 14 Oak Lane,” another brokerage’s deal | 5. SOP 12-7 | 14 |
| 7 | “Flying off the market!” | 8. No number, no source | 14 |
| 8 | “Best agent ever!” — The Smiths | 8. Real clients? Real words? Permission? (16 CFR 465) | 16 |
| 9 | “as low as $2,950 a month” | Stop: The Golden Handcuffs Conversation first | 16 |
| 10 | A team name and a cell number, no brokerage | 4. 61J2-10.025(1); SOP 12-5 | 14 |
A photo of the oak trees on Oak Lane, no people. “A home on Oak Lane closed this month (source: MLS),” if your MLS rules allow. One line about Sam’s own listing two streets over, with its real details. No testimonial unless a real client gave those words and written permission. No payment figure. “Sam Rivera, Sample Realty Group,” the licensed brokerage name in full. Sent by EDDM to every address on the routes around Oak Lane, chosen on a map. Nothing handwritten, nothing in a mailbox by hand.
Six pieces, all sample data. For each, run the ten points and give a verdict.
For each piece: mail it as is, fix it (say how), or stop and check first (say with whom).
Answer key: Appendix A, page 40.
Small, checkable steps. The pen work fits at your desk; the people work from Holiday Networking happens at the gatherings.
| When | Do this | Done means |
|---|---|---|
| Wed, Dec 9 today | Open your Monday Call List or CRM export. Draft your Keep List with Prompt 1, anonymized. Count your Hand tier against the hours you have. Buy rectangular cards and Forever stamps. | Four tiers, and a Hand tier you can actually write |
| Dec 10–16 | Write Hand-tier holiday notes, if you can mail them by Dec 17. Your words first; Prompt 3 only to tidy. Do Exercises 3 and 4. | Notes written in your hand, stamped |
| Thu, Dec 17 | USPS’s recommended First-Class send-by date for Dec 25. Anything not written yet becomes a New Year note. | Holiday notes in the mail |
| Fri–Sun, Dec 18–20 | Write the New Year notes and sign the Card-tier cards, one handwritten line each. Run the ten-point checklist on every piece. Do Exercises 5 and 6. | A checked stack, addressed and stamped |
| Mon–Tue, Dec 21–22 | Mail the New Year notes and cards, First-Class, before Holiday Networking’s quiet week begins. | In the mail |
| Dec 23 – Jan 1 | Holiday Networking’s quiet week: no outreach. Nothing to write, nothing to send. | — |
| Jan 4–8 | Holiday Networking’s January Callback (Jan 4–15) follows the cards. The card is not the Callback. Log replies to your cards. On Fri, Jan 8, sketch your 2027 Postmark Plan with Prompt 4 and date the first Route-tier drop. | A 2027 plan on one page |
Those are the only targets in this book. There is no response-rate target and no deal count. If you want a number of your own, set it, and call it a goal, not a forecast.
Every one of these is normal. Each has a fix.
| Symptom | Likely cause | Fix |
|---|---|---|
| You freeze at the blank card | Worry about wording | Three sentences. Your words land better than you fear (page 4). |
| The Hand tier is too long for your hand | Too many people in it | Move some to the Card tier. Count hours, not contacts. |
| The tidy added a memory or a feeling | It filled a gap | Delete it. Prompt 3’s rule is NOT KNOWN, never a guess. |
| Square cards came back for postage | Nonmachinable surcharge | $1.31 each, or rectangular cards next time. |
| A client asks whether a machine wrote it | A fair question | Tell the truth (page 29). |
| Someone asks you to stop mailing | Normal | Flag it that day, on every list, and thank them. |
| You missed Dec 17 | Normal | It is a New Year note now. |
| Nobody writes back | Normal. Cards rarely ask for a reply. | Nothing to fix. The card was a thank-you. |
The 2027 NAR Code of Ethics (effective Jan 1, 2027) was not posted on Nov 3, 2026. Check SOPs 12-5, 12-7, 10-3 and Article 16 against it. Checked on: ________
USPS Notice 123 before you quote a price. No January 2027 stamp increase had been filed as of Nov 3, 2026. Checked on: ________
Your MLS rules on sold data in printed mail. Checked on: ________
Vendor prices and terms, if you use a service. Every price here is as published on Nov 3, 2026. Checked on: ________
Nobody is marking this but you. Tick only what you could do tomorrow, at your desk or in front of your broker, without this book open.
| ✓ | I can… | Module |
|---|---|---|
| Explain what Kumar & Epley (2018) found about thank-you letters, and what it did not test | 1 | |
| Explain why the AI penalty attaches to belief, not to the quality of the words | 1 | |
| Say what the opera-house test found about handwritten cards, and why it matters | 1 | |
| Label a commissioned study, a vendor claim and stated intent whenever I quote one | 1 | |
| Keep anything unstamped out of mailboxes, and send handwritten notes First-Class | 2 | |
| Price a rectangular card, a postcard and a square card at 2026 postage, with the date | 2 | |
| Put the licensed brokerage name on advertising, and claim “sold” only for my own deals | 2 | |
| Choose farm routes by geography, and keep people out of farm imagery | 2 | |
| Sort my database into Hand, Card, Route and None by relationship alone | 3 | |
| Take active listings and stop requests out of every targeted tier | 3 | |
| Write a three-sentence warm note from a real memory, with no pitch | 4 | |
| Run the Pen Pass: my words, an optional tidy, my hand, my stamp | 5 | |
| Explain the honest line on robot pens and fonts, and answer truthfully if asked | 5 | |
| Plan December around Dec 17 and the New Year, and run the ten-point checklist | 6 |
| Item | My answer |
|---|---|
| My Keep List counts on Dec 9: Hand / Card / Route / None | |
| By Dec 17: Hand-tier holiday notes mailed, or moved to the New Year. Number mailed: | |
| By Dec 22: New Year notes and cards mailed, every one through the ten-point checklist. Number mailed: | |
| By Jan 8, 2027: my 2027 Postmark Plan on one page. The date I finished it: | |
| My broker’s answers on the brokerage name, robot pens and co-branded mail: the date I asked | |
| Who holds me to the Dec 22 and Jan 8 dates |
Your wording will differ. Check it for the idea, not the words. This is education, not legal advice.
| # | The rule | What you do instead |
|---|---|---|
| 1 | Handwritten matter may not be entered as EDDM-Retail (DMM 143.3.4), and a personal line makes it personal correspondence. | Send the EDDM piece clean. Mail handwritten notes First-Class to your Hand and Card tiers. |
| 2 | Florida: all real estate advertising must include the licensed brokerage name (61J2-10.025(1)); NAR: the firm name in any medium (SOP 12-5). | Add the brokerage’s licensed name. We found no Florida team-name rule, so ask your broker how the team name may appear beside it. |
| 3 | Nothing unstamped in a mailbox (18 U.S.C. 1725; DMM 508.3.1.3). | Leave the report at the door, or stamp it and mail it. |
| 4 | Only the listing or cooperating broker may claim “sold” (SOP 12-7). | Report it as a market fact, “A home on Oak Lane closed this month (source: MLS),” if your MLS rules allow, with your brokerage name. |
| 5 | A square envelope, a ribbon or a seal makes it nonmachinable: $0.82 + $0.49 = $1.31 (DMM 101.1.2; Notice 123). | Put $1.31 on each, or use rectangular cards. |
| 6 | “Fewest children” and “empty nesters” both point at familial status, a protected class (42 U.S.C. 3604(c)). Choosing locations on that basis (24 CFR 100.75(c)(3)) and wording that signals who the home is for (100.75(c)(1)) are both covered. | Choose whole routes on a map, for your own business reasons. Describe the homes, not the people. |
Where more than one answer is defensible, the key says what makes the difference.
| # | Tier | Why |
|---|---|---|
| 1 | Hand | A close, recent client. A note about something real from the deal. |
| 2 | Hand | A thank-you note. Nothing of value tied to the referrals: RESPA has no small-gift exception (Holiday Networking, Lesson 2.8). |
| 3 | Card | You cannot write a real memory, so do not invent one. One honest handwritten line: “I hope the house has been good to you.” |
| 4 | None, for that note | A targeted “thinking of selling?” note to an owner listed with another broker is what Article 16 and SOP 16-4 bar. Your general Route piece to the whole route is fine (SOP 16-2). |
| 5 | Route, but not that way | Choosing locations by “share of young families” is the pattern 24 CFR 100.75(c)(3) prohibits. Leave the filter off and pick whole routes on a map. |
| 6 | None | Flag it that day, on every list, your vendors’ included. |
| 7 | None | A stranger, and cards from strangers rarely land (Meier, 2016). If you use any “likely seller” list, ask how it was built, and take active listings off it first (Article 16). |
| 8 | Hand or Card | Both are right. It depends on the hours you have. Sphere either way. |
The Pen Pass and the ten points decide most of these. Your broker’s policy comes first.
| # | Verdict | Why |
|---|---|---|
| 1 | Pass | That is the Pen Pass: your words, a light tidy, your hand, First-Class with a stamp. |
| 2 | Fix | The ink is yours, the words are not. Write the warm part yourself first, then tidy. Check the AI version for details it invented. |
| 3 | Fix | Help from another person looked no better than AI help (Liu, Kang & Wei, 2024), and her hand signed with your name passes off someone else’s hand as yours. She can address and stamp. You write the warm part, or move those people to the Card tier and sign the cards yourself. |
| 4 | Pass, in the Card tier | Your words, your brokerage, no claim of your hand. Ask the vendor and USPS which mail class applies. If the service pens your signature, sign it yourself or print the name. |
| 5 | Don’t | Handwritten matter may not be entered as EDDM-Retail (DMM 143.3.4). Mail them First-Class. |
| 6 | Don’t | A business may not create a testimonial that misrepresents the client’s experience (16 CFR 465.2(a)), and the superlative has no source. Use “Sam made it easy!” word for word, with their written permission. |
| # | Verdict | Why |
|---|---|---|
| 1 | Mail it | Rectangular, stamped, First-Class, brokerage on the envelope, and before Dec 17. |
| 2 | Fix | A postcard is First-Class, so the handwritten line is fine, and 65¢ covers it if it is card size (3½–4¼ by 5–6 inches; a 4 × 6 fits). But there is no envelope, so put the brokerage’s licensed name on the card. |
| 3 | Mail it | Your own listing, whole routes, the house and not people, the brokerage named. Check EDDM size and the 200-piece minimum per mailing. |
| 4 | Fix | Square and ribboned: $1.31 each, not 82¢. |
| 5 | Fix | Delete “I wrote this by hand tonight.” A machine’s writing never claims your hand. |
| 6 | Stop | A partner paying the whole bill is a thing of value under RESPA. Each party pays its own fair-market share. Take it to your broker (page 16). |
The honest version is usually one small change away: write the words yourself, delete one false line, add the brokerage name, pay the right postage. The pieces to stop or not send are the ones that break a mail rule, or use someone else’s words, sale or money.
One page for your desk, next to the stamps.
The moment that really happened · the person, what it said about them · the wish, no pitch · your name, by hand.
Card 82¢ · postcard 65¢ (card size only) · square or lumpy $1.31 · EDDM flat $0.26.
Use an AI tool your broker approved, with training off. No real names or contact details. Every prompt produces a draft or a flag for you to check by hand. None of them writes the warm part, and none sends anything.
Below is an export of my contacts, anonymized: ID numbers or role labels (“Buyer A”) only, no names, phone numbers, e-mails or addresses. Your goal is to suggest a tier for each row, Hand, Card, Route or None, based only on my relationship with the person. It is not to rank people by value, predict who will sell, or add anyone.
MY EXPORT: [PASTE: ID, relationship (closed with me / referred me / sphere / farm route / met once), year of last transaction, last real contact, do-not-mail (Y/N), listed with another broker now (Y/N/NOT KNOWN)]
MY HAND-TIER LIMIT: [how many notes I can write by hand this month]
Do this:
1. Make a table: ID, suggested tier, one-line reason taken from my fields.
2. Anyone flagged do-not-mail goes to None. Anyone listed with another broker goes to None for anything about my services; flag them for my broker.
3. If more people qualify for Hand than my limit, mark the extras "Hand or Card: my choice."
RULES:
- Use only the relationship fields I gave. Never use or infer age, family, children, religion, health, disability, race, national origin, sex or income.
- Route-tier rows are whole routes I chose on a map. Never suggest routes by who lives there.
- If a field is missing, write NOT KNOWN. Never fill in a guess.
Then tell me to check every Hand-tier row by hand: can I name a real moment with this person? If not, it moves to Card.
I want to write a short thank-you note to one person and I am stuck on what to say. Your goal is to ask me questions that help me remember a specific, real moment with them. It is not to write the note, suggest memories, or guess what happened.
WHO (relationship only, no name): [e.g. "buyer, closed in August"]
WHAT I REMEMBER SO FAR: [anything, or "nothing yet"]
Ask me five short questions, one at a time, about what happened between us: a moment, a decision, something they did or said. Wait for my answer each time.
RULES:
- Ask only. Never offer a memory, an example from someone else, or a detail I did not give you.
- Never ask about health, family situation, religion, money or anything personal they did not share with me as part of our work.
- If an answer is vague, ask one follow-up. If I still cannot remember, write NOT KNOWN and say: "Write a shorter note. Don't invent one."
Then list my answers back to me, in my words, so I can pick one moment and write the three sentences myself.
Prompt 3 is the only prompt that touches a note, and it touches only spelling and clarity. Prompt 4 plans; it never schedules.
Below is a note I wrote myself, in my own words. Your goal is to tidy it: spelling, punctuation, and any sentence that is hard to read. It is not to make it warmer, longer or more polished, and not to change how I talk.
MY NOTE: [PASTE, with the person's name replaced by a role label such as "Buyer A"]
LENGTH LIMIT: [e.g. fits a 4 x 6 card, about 60 words]
RULES:
- Keep my words. Change only what is wrong or unclear.
- Never add a memory, a feeling, a detail, a compliment, a date or a fact. If a sentence seems to be missing something, write [NOT KNOWN] there. Never fill in a guess.
- Never add a pitch, a statistic, a prediction or a call to action.
- Keep slang and phrases that sound like a person talking.
Show the tidied note, then list: CHANGES (each one, before and after), ADDED (this should say "nothing"), NOT KNOWN (any gaps you marked).
Then tell me to read it out loud, put back anything that does not sound like me, and copy it out by hand.
Below are my dates for next year and my Keep List counts. Your goal is a one-page mail calendar: who gets what, and the date each piece goes in the mail. It is not to add mailings, set response targets or invent deadlines.
MY KEEP LIST COUNTS: Hand [ ], Card [ ], Route [routes / addresses]
MY DATES: [closings and anniversaries by month, no names; farm drops I can afford, by month]
USPS DATES I CHECKED MYSELF: [paste, with the date I checked, or leave blank]
Make a table: month, Hand tier, Card tier, Route tier, the date each piece goes in the mail.
RULES:
- Use only my dates and the USPS dates I pasted. If a date is missing, write NOT KNOWN. Never fill in a guess, and never state a USPS deadline I did not give you.
- Anything handwritten or personal is First-Class. Route-tier pieces carry nothing personal.
- No mail to anyone flagged do-not-mail.
- This is a plan only. Do not schedule, send or order anything.
Then tell me which three dates to check by hand against USPS and my own calendar.
Both prompts flag problems. Neither decides what is legal, and neither rewrites your piece as a stronger sales message.
Below is the text of a farm mail piece and a description of its images. Your goal is to flag anything that may need fixing before my broker reviews it. It is not to rewrite it as a stronger sales piece, and not to decide what is legal.
THE PIECE: [PASTE the text; describe each image]
HOW IT WILL BE MAILED: [EDDM / Marketing Mail / First-Class]
HOW I CHOSE THE AREA: [e.g. "the three routes around my listing, picked on a map"]
MY BROKERAGE'S LICENSED NAME: [ ]
Answer PASS, FLAG or NOT KNOWN for each, quoting the line:
1. Does any word, phrase or image suggest who the home is for, or show people?
2. Was the area chosen by geography only?
3. Does the licensed brokerage name appear?
4. Does it claim "sold"? If so, I must confirm I was the listing or cooperating broker.
5. Any number, claim or superlative without a source?
6. Any testimonial? If so, I must confirm a real client's real words, with permission.
7. Any payment or rate figure, or a partner's name or logo? Flag it for my broker.
8. Anything handwritten or personal on an EDDM or Marketing Mail piece?
RULES:
- If you cannot tell, write NOT KNOWN. Never fill in a guess.
- No legal conclusions. Write "ask your broker" instead.
- Never suggest targeting any group or adding people to the images.
Then remind me to run the ten-point checklist myself and send the piece to my broker.
Below is one note or card I plan to mail, and how it will travel. Your goal is to check it against my ten-point checklist. It is not to rewrite the message, and not to decide what is legal.
THE MESSAGE: [PASTE, name removed]
WHO WROTE THE WORDS: [me / me, AI tidied / AI drafted / a service]
HOW IT IS WRITTEN: [my hand / robot pen / font / printed]
SIZE, SHAPE, CONTENTS: [e.g. 5 x 7 rectangular, card only]
POSTAGE AND CLASS: [ ] HOW IT GETS THERE: [Post Office / collection box / by hand to the door]
BROKERAGE NAME: [where it appears, or "none"]
Answer PASS, FIX or NOT KNOWN for each: 1. Handwritten or personal content goes First-Class with a stamp. 2. Nothing unstamped goes in or on a mailbox. 3. Postage fits the size and shape (I will confirm prices). 4. The brokerage name appears. 5. No unconfirmed "sold" claim. 6. No wording or images about who a home is for. 7. Not a targeted piece to an owner listed elsewhere. 8. No unsourced number; any quote is real, with permission. 9. If a machine wrote it, nothing claims my hand. 10. The person is not flagged do-not-mail.
RULES:
- If you cannot tell, write NOT KNOWN. Never fill in a guess.
- No legal conclusions; write "ask your broker".
- Do not send, order or schedule anything.
Then remind me to read it once more myself before I stamp it.
Generic sample wording, not quotes from real notes. Each blank is for something that really happened between you and the person. If nothing fits, skip the line.
| # | The line | Fits | Fill the blank with |
|---|---|---|---|
| 1 | “You were so patient through ___.” | After a closing | The real hard part: three inspections, a long search |
| 2 | “Thank you for trusting me with ___.” | After a closing | The home, the sale, a first purchase |
| 3 | “I still smile thinking about ___.” | Any time | A moment you shared |
| 4 | “Thank you for thinking of me when ___ needed help.” | A referral | How they know the person: a coworker, a friend |
| 5 | “It has been a while, and I wanted to say ___.” | Reconnecting | A plain thank-you or a good wish |
| 6 | “I hope the first year in the house has been ___.” | A closing anniversary | Your honest wish |
| 7 | “Happy New Year. I hope 2027 brings ___.” | New Year | Something they told you they were looking forward to |
| 8 | “You made a hard week easier by ___.” | After a closing | What they did |
| 9 | “I drove past ___ and thought of you.” | Reconnecting | The street or the house, only if you did |
| 10 | “Thank you for your patience when ___.” | After a delay | The delay, in plain words |
| 11 | “Working with you reminded me that ___.” | After a closing | Something you learned |
| 12 | “I’m grateful for the way you ___.” | Any time | A habit you saw while working together |
| 13 | “I hope the ___ is everything you wanted.” | After a closing | The yard, the kitchen, the view |
| 14 | “Thank you for the kind words about ___.” | After a review or a compliment | What they said |
Practice the warm part by hand, not by prompt.
| # | The line | Fits | Fill the blank with |
|---|---|---|---|
| 15 | “It was good to see you at ___.” | After running into them | The place |
| 16 | “No business in this note. Just thank you for ___.” | Any time | One specific thing |
| 17 | “Here’s to a good year in the new place.” | New Year | Nothing. It stands alone. |
| 18 | “Thank you for coming back out to ___.” | An inspector or contractor | The re-check, the visit |
| 19 | “Thank you for picking up when ___.” | A lender or title rep | The moment they helped |
| 20 | “You asked a good question about ___. Here is what I found.” | A follow-up they asked for | Their question |
| 21 | “I pulled the sales on your street myself, and ___.” | A follow-up they asked for | What you found, with the source |
| 22 | “Congratulations on ___.” | Any time | Something they told you about their own plans: a new job, a finished project |
| 23 | “I learned something from you: ___.” | After a closing | The lesson |
| 24 | “Thank you for making room for me in a busy ___.” | After a closing | The season, the month |
| 25 | “See you around the neighborhood.” | Sphere | Nothing. It stands alone. |
Several of these have no settled answer we could find. That is the reason to ask. Write down who answered, and when.
| # | Question | Ask | Answer, who, date |
|---|---|---|---|
| 1 | Must the brokerage name appear on personal notes to past clients, or only on advertising? Where: card, envelope or both? | Broker | |
| 2 | How may our team name appear next to the brokerage name in print? | Broker | |
| 3 | May I mail “just sold” for a deal where I was the buyer’s agent? How do I get the listing broker’s consent before closing? | Broker | |
| 4 | What do our MLS rules allow on sold data in printed mail? | Broker, MLS | |
| 5 | What is our policy on robot-pen and handwriting-font services? Must recipients be told? | Broker, counsel | |
| 6 | Which mail class should a robot-pen or font card use? | Vendor, USPS | |
| 7 | Which AI tools are approved for client mail, on which plans, with training off? | Broker | |
| 8 | How do we record a do-not-mail request, including on vendor lists? Does our mail house suppress DMAchoice registrants? | Broker, mail house | |
| 9 | A lender or title company wants its name on our postcard. Who pays what, and how is it documented? | Broker, counsel | |
| 10 | May I send a purely personal holiday greeting to a past client whose home is now listed with another broker? | Broker | |
| 11 | Who reviews our farm routes and imagery for fair housing before we print? | Broker | |
| 12 | What written permission do we keep before using a client’s words on a mailer? | Broker, counsel | |
| 13 | What are the HOA and city rules on door hangers in my farm? | Broker, HOA, city | |
| 14 | How should we address Florida parcels that show no owner name? | Broker | |
| 15 | Does any state we mail into require a license number or other details on solicitations? | Broker, counsel | |
| 16 | (Outside Florida) What does my state’s advertising rule require on printed mail? | Broker, counsel |
Plain definitions for the terms in this book.
18 U.S.C. 1725. The federal statute against putting unstamped mailable matter in a mailbox to avoid postage. Upheld by the Supreme Court in 1981.
Aspect ratio. Length divided by height. Letters outside 1.3 to 2.5, a square envelope included, pay the nonmachinable surcharge.
Card tier. Keep List tier 2. Past clients and sphere: a card you sign by hand, with one handwritten line; or a service card in your words that makes no claim to be your hand.
Commissioned study. Research paid for by an organization with a stake in the answer, such as a postal operator. Quote it with who paid.
DMM. The USPS Domestic Mail Manual, the rulebook for what each mail class may carry.
EDDM. Every Door Direct Mail. Unaddressed flats to every address on chosen carrier routes. Nothing personal or handwritten.
Effort heuristic. The tendency to rate things higher when you know effort went in. Replications were mixed.
First-Class Mail. The everyday class for stamped letters and cards, and one of the few classes allowed to carry handwritten or personal mail. A Forever stamp covers a 1-oz letter.
Hand tier. Keep List tier 1. The few who get a real pen-and-ink note from you.
Handwriting font. A typeface that imitates handwriting, printed by machine.
Keep List. System 1. Hand, Card, Route or None, by relationship, with the rules attached.
Marketing Mail. USPS’s bulk advertising class. Needs a permit; may not carry handwritten matter.
Nonmachinable surcharge. $0.49 on a letter that is square, lumpy, rigid or has a clasp or string (as of Oct 4, 2026).
Notice 123. USPS’s current price list.
Pen Pass. System 2. Your words, an optional AI tidy, your hand, your stamp.
Postmark Plan. System 3. The year’s mail on one page, plus the ten-point pre-send checklist.
Preprint. A paper posted before peer review.
Preregistered. A study whose plan was filed publicly before the data came in.
Robot pen. A machine that holds a real pen and writes the card for you.
Route tier. Keep List tier 3. One identical piece to whole farm routes chosen by geography.
SOP 12-7. The NAR Standard of Practice that limits “sold” claims to the listing or cooperating broker.
Stated intent. What people say they will do on a survey, such as “very likely to respond.” Not measured behavior.
Vendor claim. A company’s statement about its own product or results, not independently checked.
Checked Nov 3, 2026, unless noted. “Via” means we read a report of the source, not the source itself. Publisher stakes are noted in the text where they apply.
Rules, prices and products change. Confirm the current text before relying on any of these.
To Your Next Listing,
Edmund Bogen
Because Mail That Goes Straight in the Bin is NOT an Option!