Operator’s manual / rev 0.5

The newsletter is the storefront. This is the machine.

Regional Newsletter is the complete operating system behind The Austin Newsletter: the content pipeline, the edition production runbook, the SEO guide engine, the outreach rails, the inbound reply automation, and the kill criteria that decide when to stop. Written while the machine runs, not reconstructed after an exit.

The rule the whole machine is built on: automate distribution and ops, keep voice and curation human.

Get the playbook

PRODUCT RECORD

WRITTEN COURSE

lessons, each ending in a checkpoint

chapters across six parts

of them stand between you and edition one

appendices, including the day-one guardrails

≈73h of hands-on work, by the lessons’ own estimates

Summed from the estimate each lesson carries in its own header. Our estimate of effort, not a schedule.

The source operation is live and pre-revenue, and every sheet here was written while it runs. theaustinnewsletter.com

RN-01 / The system

One machine, four rails

This is the Austin operating system as it runs today, not a reference architecture. Every box on the sheet has its own runbook, and the course chapter that builds it.

growth-system docs, one per tactic

setup runbooks, one per system

155

operational scripts wired to the rails

database migration files checked in

The working repo behind the diagram, counted on 2026-08-18. Nothing rounded up.

RN-02 / Curriculum

Every sheet in the manual

Six parts, 27 chapters, 83 lessons. Open a chapter to read its lesson titles and what each one costs you in reading and in work. These rows are generated from the shipped lesson files when this page is built, so the page cannot claim a lesson that is not there.

PART 0Start here

1 chapter · 1 lesson · 5m readON THE LAUNCH PATH

00How to run this course5m read · 5m work1 lesson
1. 0.1How to run this coursecore · 5m read · 5m work

PART 1Decide

3 chapters · 9 lessons · 29m readON THE LAUNCH PATH

01Qualify your city14m read · 3h 15m work4 lessons
1. 1.1The two numbers that kill most citiescore · 4m read · 45m work
2. 1.2The competition scancore · 3m read · 1h work
3. 1.3Search before you name anythingcore · 3m read · 45m work
4. 1.4Kill criteria: set the bar before you startcore · 4m read · 45m work
02What grew ours, measured10m read · 1h work3 lessons
1. 2.1The scoreboardcore · 4m read
2. 2.2The three speeds, and the attention budgetcore · 3m read
3. 2.3Your platform standingcore · 3m read · 1h work
03The honest economics5m read · 30m work2 lessons
1. 3.1What this actually costscore · 3m read · 15m work
2. 3.2What is optional at the startcore · 2m read · 15m work

PART 2Foundation

6 chapters · 20 lessons · 1h 2m readON THE LAUNCH PATH

04Voice and the persona system11m read · 4h work3 lessons
1. 4.1The persona policycore · 3m read · 1h work
2. 4.2The voice doc and the AI-tell killerscore · 4m read · 2h 30m work
3. 4.3Truth gatescore · 4m read · 30m work
05The operator's workbench13m read · 2h 45m work4 lessons
1. 5.1The repo is the operating systemcore · 3m read · 1h work
2. 5.2Claude Code: setup and delegationcore · 4m read · 45m work
3. 5.3MCP: the wiring patterncore · 3m read · 30m work
4. 5.4Browser automation: the agent's other handscore · 3m read · 30m work
06Domain, inbox, and DNS8m read · 1h 35m work3 lessons
1. 6.1Register the namecore · 2m read · 20m work
2. 6.2Google Workspace and your inbox-domain authcore · 3m read · 1h work
3. 6.3DNS disciplinecore · 3m read · 15m work
07beehiiv9m read · 2h 45m work3 lessons
1. 7.1Publication setup, and the address in the footercore · 3m read · 1h work
2. 7.2The referral programcore · 3m read · 45m work
3. 7.3The welcome emailcore · 3m read · 1h work
08Deliverability: placement over paperwork10m read · 45m work3 lessons
1. 8.1Sending-domain auth: what to know, what not to build yetcore · 3m read
2. 8.2The story: green checks, spam foldercore · 4m read
3. 8.3Your placement instrumentcore · 3m read · 45m work
09The minimal site11m read · 3h 30m work4 lessons
1. 9.1Why a site, and the minimum buildcore · 3m read · 1h 30m work
2. 9.2The two-surface architecturecore · 3m read · 30m work
3. 9.3Instrument on day onecore · 2m read · 45m work
4. 9.4Tag your links before anyone clicks onecore · 3m read · 45m work

PART 3Produce and launch

2 chapters · 7 lessons · 19m readON THE LAUNCH PATH

10The weekly edition12m read · 3h 45m work4 lessons
1. 10.1The edition template, and your send daycore · 3m read · 45m work
2. 10.2The content pipelinecore · 3m read · 1h 30m work
3. 10.3Production weekcore · 3m read · 1h work
4. 10.4Images, rights, and the AI linecore · 3m read · 30m work
11Launch7m read · 3h 30m work3 lessons
1. 11.1The seed listcore · 3m read · 1h work
2. 11.2Edition one: the first-send runbookcore · 2m read · 2h work
3. 11.3Read what you sentcore · 2m read · 30m work

PART 4Grow

13 chapters · 39 lessons · 1h 58m read

12The database6m read · 1h 30m work2 lessons
1. 12.1Why a newsletter needs a database, and the v1 schemastack · 3m read · 1h work
2. 12.2Migration and data disciplinestack · 3m read · 30m work
13Attribution before traffic4m read · 1h 30m work1 lesson
1. 13.1Attribution: the full plumbingstack · 4m read · 1h 30m work
14Schedulers, secrets, and switches9m read · 2h work3 lessons
1. 14.1Where jobs livestack · 3m read · 45m work
2. 14.2Secrets, and the two-sided secretstack · 3m read · 45m work
3. 14.3Kill switches and build disciplinestack · 3m read · 30m work
15Slack: the ops console12m read · 2h 30m work4 lessons
1. 15.1The console and the card patternstack · 3m read · 45m work
2. 15.2Veto windows done honestlystack · 3m read · 30m work
3. 15.3Panels over gatesstack · 3m read · 30m work
4. 15.4The listening feedengine · 3m read · 45m work
16Business outreach (our highest-yield outbound channel)23m read · 10h 30m work7 lessons
1. 16.1Why businesses, not readersengine · 3m read
2. 16.2The compliance floor for outboundengine · 3m read · 30m work
3. 16.3The listing page, and the verticals you must never rankengine · 3m read · 4h work
4. 16.4Census to inviteengine · 3m read · 3h work
5. 16.5Send-rail disciplineengine · 4m read · 2h work
6. 16.6The cold rail we lostengine · 4m read
7. 16.7Follow-ups, and this channel's honest verdictengine · 3m read · 1h work
17Awards, badges, and the credential8m read · 2h work3 lessons
1. 17.1The award as the wedgeengine · 3m read · 1h work
2. 17.2The naming lessonengine · 3m read · 30m work
3. 17.3Link hygieneengine · 2m read · 30m work
18The inbound reply engine and the all-mail CRM9m read · 4h 45m work3 lessons
1. 18.1The engine, and the CRM under itengine · 4m read · 4h work
2. 18.2Consent, preciselyengine · 3m read · 30m work
3. 18.3What stays humanengine · 2m read · 15m work
19Capture surfaces, and the signup moment7m read · 4h 30m work2 lessons
1. 19.1The gated tool playengine · 4m read · 4h work
2. 19.2The signup moment is yoursengine · 3m read · 30m work
20Community forums9m read · 1h 30m work3 lessons
1. 20.1The shape of the post that landedengine · 3m read · 1h work
2. 20.2The platform's physicsengine · 3m read · 30m work
3. 20.3Automation on a knife edgeengine · 3m read
21Groups on the big social platform8m read · 1h 15m work3 lessons
1. 21.1The reply-first playengine · 3m read · 45m work
2. 21.2Running it at cadenceengine · 3m read · 30m work
3. 21.3What never happens hereengine · 2m read
22UGC flywheels: the dish-of-the-week shape6m read · 1h 45m work2 lessons
1. 22.1One ask, four jobsengine · 3m read · 45m work
2. 22.2The pipeline, the choke point, and the ratio alarmengine · 3m read · 1h work
23SEO guides: the slow bet11m read · 3h 45m work4 lessons
1. 23.1The strategyengine · 3m read · 45m work
2. 23.2The publish gateengine · 3m read · 2h work
3. 23.3The third-party data policyengine · 2m read · 30m work
4. 23.4Honest expectations, and reading your toolsengine · 3m read · 30m work
24Measurement and honest verdicts6m read · 1h 45m work2 lessons
1. 24.1The verdict canoncore · 4m read · 45m work
2. 24.2The operator's weekly hourcore · 2m read · 1h work

PART 5Operate

2 chapters · 7 lessons · 18m read

25The guardrail canon8m read · 1h 30m work3 lessons
1. 25.1Verify the artifactcore · 2m read · 30m work
2. 25.2Alarms, queues, and stuck decisionscore · 3m read · 30m work
3. 25.3The process canoncore · 3m read · 30m work
26Deciding: reviews, parks, and city #210m read · 4h 30m work4 lessons
1. 26.1Running the kill-criteria reviewcore · 3m read · 1h work
2. 26.2Scale, park, or killcore · 2m read · 30m work
3. 26.3Entity, bank, and your first sponsor conversationcore · 2m read · 2h work
4. 26.4Cloning to city #2core · 3m read · 1h work

APPAppendices

Appendix A: the ESP automation boundary
Appendix B: the day-one guardrail checklist

Every one of the 83 lessons ends in a completion checkpoint · 22 carry copy-paste artifact blocks · 8 start a real-world wait and are ordered early because of it · 121 declared prerequisite links hold the sequence together

RN-03 / Launch path

How much of it comes before your first send

Most of a manual this size is not blocking. Every lesson declares whether it sits on the path to edition one, and the answer is 37 of 83. The other 46 are growth and operations engines. They are the reason the package is worth keeping after launch week, and none of them stand in front of it.

RN-04 / Inside a lesson

What one sheet actually contains

A contents list cannot tell you whether a lesson is four paragraphs or a working procedure, so here is the shape all of them hold, beside the real record of one.

EVERY LESSON, SAME FIVE BLOCKS

OUTCOME
The one thing that is true when the sheet is finished. Written before the lesson was.
TEACHING
Why the procedure is shaped this way, including what it cost us to learn it.
STEPS
Imperative, in order, with the decision points named rather than assumed.
ARTIFACTS
The real copy-paste block, in the lesson. If a lesson promises an artifact, the artifact is there.
CHECKPOINT
What you tick to say it is done. Your progress is tracked against these.

SPECIMEN RECORD

sheet: 6.2
chapter: Domain, inbox, and DNS
title: Google Workspace and your inbox-domain auth
track: core
effort: 3m read · 1h work
depends on: 1 earlier sheet
artifacts: 1 copy-paste block
checkpoint: 5 boxes to tick
waits on: domain verification (minutes to hours) + DKIM propagation (up to 48h)
launch path: yes, required before edition one

The waits are the reason the ordering exists. A lesson that starts a 48-hour propagation clock sits early on purpose, so the clock runs while you do the next thing rather than while you sit still.

RN-05 / Systems tour

The rails, one by one

Each of these is a document in the working repo and a chapter in the course. The path is the real filename, because the course teaches you to build the file, not to admire it.

setup-runbook/edition-production.md

Edition production

The full build sequence for a weekly edition: structure, data-dependent slots, card art, and the checks that run before a send. The same runbook produces the Austin edition every week.

growth-systems/guide-flywheel.md

SEO guide engine

Evergreen city guides built to a written page standard, with gated directory tools behind an email wall. The search rail is designed to compound while the send list is still small.

setup-runbook/outreach-engine-ops.md

Outreach rails

Local business outreach with a mailbox-wide daily send ledger, address verification before any send, deterministic follow-ups, and per-campaign instruments so a dead shape gets retired instead of repeated.

setup-runbook/inbound-reply-engine.md

Inbound reply engine

Every inbound thread lands in a CRM table, replies are drafted in the publication’s voice and reviewed before send, and warm repliers are subscribed only when the original invite carried consent language.

setup-runbook/referral-program.md

Referral program and welcome flow

Beehiiv’s native referral milestones wired into a welcome flow, with crediting built to never break a signup on a bad ref code.

growth-systems/reddit/ + facebook-groups/

Social rails

Drafting engines for Reddit and Facebook groups with review gates, veto windows, per-community cooldowns, and reconcile jobs that read the platform instead of trusting the database.

kill-criteria.md

Measurement and kill criteria

A morning growth report, retro gates on every tactic, and pre-registered pivot triggers written down before the results came in. The playbook says when to stop as clearly as when to ship.

RN-06 / No claims

What this page will not claim

No testimonials. The first cohort does not exist yet, so there is no one to quote.
No revenue screenshots. The Austin operation is pre-revenue and the playbook says so in writing.
No subscriber-count promises. Your city, your writing, and your consistency decide that.
No countdown timer. This page looks the same tomorrow.

What is real is the depth: the scripts, the runbooks, and the rails that run every night whether or not anyone is watching. That is the product.

RN-07 / Fit

Who this is for

BUILT FOR

An operator who will write every week and wants the machine around the writing handled.
Someone picking a growing mid-size city with real transplant inflow, or already living in one.
Someone who can hold a technical playbook: the stack is Beehiiv, Next.js, Supabase, and Vercel, and the runbooks are specific.
Anyone who would rather have kill criteria than a pep talk.

WRONG FOR

Anyone shopping for passive income. Nothing in here runs without an operator.
Anyone who needs revenue proof before doing the work. It does not exist yet, and we will not invent it.
Writers who want prompts and inspiration. This is plumbing, schedules, and checklists.
Agencies looking for a white-label content mill. The playbook’s growth rules forbid exactly that.

FREE INSTRUMENT

Find out whether your city qualifies, before you spend anything

Four questions, no payment, no account. An instrument that can say yes has to be able to say no, and this one will. The thresholds are ours, informed by the 80,000 to 150,000 city-population range the Naptown Scoop operator has published and the roughly 150 qualifying US markets a venture-funded multi-market operator’s screens identify. The test asking for 25 payable businesses and the 75,000 floor are ours, and they are unforgiving on purpose.

Run the city check

RN-08 / Founder

A note from Joe

I run the systems behind The Austin Newsletter. Carissa, my wife, is the byline and the voice; the machine that gets her edition out every week is mine.

The course this page sells is not a summary of that machine. It is the document we work from, rewritten so a stranger can run it: the same file gets edited when a rail changes, and the same kill criteria decide what we stop doing. When we clone to city #2, this is what we clone from.

I want to be plain about the state of it. Austin is live and pre-revenue. The lessons that depend on long-run data say where they stand, and the monetization material fills in from real sponsor invoices, not projections. That honesty makes the product less shiny. It is also the reason to trust the rest of it.

Joe Spisak

RN-09 / Price

Get the playbook

One purchase, once: the written course, the artifacts inside it, and a lifetime seat in the operator community. Organized to be run, not read once.

the course, 83 lessons across 27 chapters
2 appendices, including the day-one guardrails
the copy-paste artifacts, inside the lessons
the kill criteria, pre-registered
the operator community, lifetime, no renewal ever

$999

One-time. No subscription, no royalty, no upsell waiting behind the checkout.

Not sure yet? Start with the free city check and see whether your city clears the bar before you spend a dollar.

30-day money-back guarantee. If the playbook is not right for you, email joe@regionalnewsletter.com within 30 days of purchase and we will refund the full $999. No conditions, no forms. Refunds go back to your original payment method, issued within 5 business days.

Get the playbook

RN-10 / FAQ

Fair questions

Is this a course with videos?

No videos. It is a written operating system delivered as a gated web course: 83 lessons across 27 chapters plus 2 appendices, counted from the shipped files at build time rather than typed into this sentence. Every lesson ends in numbered steps and a completion checkpoint, with your progress tracked as you go; the launch-heavy chapters carry copy-paste artifact blocks where a template genuinely saves you time. Buying it also seats you in the operator community for life.

How long does it take?

The lessons carry their own hands-on effort estimates and they sum to roughly 73 hours of doing (the audit found the per-lesson reading estimates unreliable, so we do not quote one). Those are our estimates of effort, not a schedule and not a promise: 37 of the 83 lessons stand between you and your first edition, and the rest are engines you switch on afterwards, at whatever pace your week allows.

Will this work in my city?

Honest answer: not every city qualifies, and most sellers in this market refuse to say so. Two tests before you spend anything: can you name 25 local businesses that would plausibly pay to reach your audience, and does your metro have at least 80,000 people within a short drive, with 150,000 as the comfortable line. A venture-funded multi-market operator’s screens identify only about 150 qualifying US markets. A full market-qualification rubric ships inside the playbook, and it is built to be able to tell you no. A free four-question version lives at the city check on this site.

What is the refund policy?

A 30-day money-back guarantee: email joe@regionalnewsletter.com within 30 days of purchase and we will refund the full $999. No conditions, no exercises to show, no forms to fill. Refunds go back to your original payment method, issued within 5 business days. With no testimonials on this page, the guarantee is doing that job, so it is the strongest one we could write.

Do you promise revenue or subscriber numbers?

No. The Austin operation is pre-revenue and the playbook’s monetization section stays thin until real sponsor data exists. Anyone selling you certainty in this market is selling the certainty, not the market.

What stack does it assume?

Beehiiv for email, Next.js on Vercel for the site, and Supabase for the backend. You can substitute, but the runbooks are deliberately specific, and the specificity is most of their value.

Why sell the playbook instead of just launching more cities?

The plan is both. Each city is a separate publication on the same playbook, and the playbook is the parent product. Selling it forces it to stay clean enough for a stranger to run, which is exactly what city #2 needs.

How current is it?

The playbook is versioned (v0.5 today) and edited while the machine runs. The lessons that depend on long-run growth and revenue data say so where they stand, because writing them before the data exists would be fiction.

When does it open?

When the payment link goes live. The waitlist doubles as the Regional Newsletter operator letter: free tools and working pieces from the running machine while you wait, then the launch email. Unsubscribe anytime.

RN-11 / Field notes

The questions, answered in the open

The research everyone does before starting a local newsletter, answered from a running operation instead of a content mill. Free, no gate, and honest enough to talk you out of it.

How to start a local newsletter: the honest sequenceRN-NOTE-01

The nine-step sequence for starting a local newsletter, from an operator running one. Step one is qualifying the city, because most cities do not.

How much does it cost to start a local newsletter?RN-NOTE-02

Real, dated line items from a running city newsletter, plus a lean-start budget built from current public vendor pricing. Verified 2026-08-15.

Is a local newsletter profitable? The honest answerRN-NOTE-03

Some local newsletters are profitable, most are not, and nobody can prove yours will be. The structural drivers, the public record, and no income claims.

beehiiv vs Substack for local newslettersRN-NOTE-04

A local-operator comparison of beehiiv and Substack: sponsor tooling, referrals, automations, and what each costs. Facts verified 2026-08-15.

How the local newsletter business model worksRN-NOTE-05

The structure of the local newsletter business: free email product, sponsorship sold direct to local businesses, and the two constraints that decide it.