# The first sale of a niche e-book comes from people who already know you: a warm-circle launch, before strangers

Written 31 Oct 2026, 22:3xZ (skills harvest). The root's priority: getting a niche e-book its first sale. Existing skills cover the page (first-sale-5-pound-ebook-mobile-shelf-*), the list emails (welcome-sequence-first-sale-*) and traffic. This one covers **who** buys first, and why waiting for strangers from search or social is the slow route at our numbers.

## The evidence (read this run unless marked)
- **Mollick 2014, "The dynamics of crowdfunding"** (Journal of Business Venturing; 48,526 Kickstarter projects; PDF read 31 Oct):
  - For an average Film project, a founder with **10 Facebook friends had a 9% chance of success; 100 friends 20%; 1,000 friends 40%**.
  - Projects "mostly succeed by narrow margins, or else fail by large amounts": the mean failed project reached 10.3% of its goal.
  - The size of the founder's own network is one of the strongest predictors of success.
- **Agrawal, Catalini & Goldfarb, "Some Simple Economics of Crowdfunding"** (NBER w19133; PDF read 31 Oct): "Friends and family disproportionately invest early in the funding cycle, generating a signal for later funders." The asymmetry "is strongest for the first investment decision" and fades once strangers can see progress. **The first buyer is a signal; it is almost never a stranger.**
- **Nathan Barry, practitioner testimony** (nathanbarry.com, read 31 Oct):
  - The App Design Handbook (2012) launched to a pre-launch list of **795 people**. "Sending to that list is what sold the first $1,000", within 10 minutes; it reached about $6,000 on day 1.
  - Later: **8,162 uniques → 321 sales = 3.9%**, and 14,436 → 421 = 2.9% when more of the traffic was cold (Hacker News). Cold traffic converts worse.
- **Sahil Lavingia (Gumroad founder), The Minimalist Entrepreneur** (summaries/snippets only, not read in full): sell to friends, then friends of friends, then strangers. "Launch" publicly only after the first 100 customers.

## What it means in arithmetic
- Cold page conversion for a well-made e-book page is roughly **3–4%** at best (Barry). Our cold traffic is crawler-heavy and unqualified, so assume ≤1%.
- At about 6 UK visitors a day to /ebooks/ (CF, 31 Oct), a 1% rate means **one sale every ~17 days**, and a 3% rate one every ~5–6 days. Search won't change this for weeks (see the impressions-zero-clicks-* skill).
- A warm circle converts in a different class. People who know the author and the place buy to support them, read it, and pass it on. Ten personal notes that produce 2–3 sales beat a month of cold traffic, and those buyers become the first reviews and shares (reviews-social-proof-direct-ebooks.md: 5 reviews means +270% purchase likelihood).

## The playbook: rings, in order
1. **Ring 0, the author's own circle** (owner action; Tab never contacts these people). These are people who know him and the farm's story: family who have asked, friends, neighbours, local people who have talked about the farm. Use **one personal message each**, in his own words, one to one, never a bulk send or BCC list:
   - one line on why he wrote it ("I finally put the farm's history into a book");
   - the free first edition link, for anyone who just wants to read;
   - the £4.99 edition link, framed as the way to support the research;
   - one ask: "if you know someone who'd want this, pass it on."
   - No pressure, no deadline and no "was/now" price. Prices exactly as listed.
2. **Ring 1, people who already raised their hand.** That's the free-chapter and Mailchimp subscribers (currently 1), anyone who replied to an email or comment about the book, and talk attendees. Covered by welcome-sequence-first-sale-* (E3 offer at the listed price).
3. **Ring 2, communities he already belongs to.** Local history and family-history groups he's a member of, posting under each group's self-promotion rules (family-history-society-journals-forums.md, reddit-forum-participation-heritage.md, civic-local-history-society-talks-penarth.md).
4. **Ring 3, strangers**: search, social reach, press. Only these channels scale, but they convert worst and slowest. Once Ring 0–1 have produced the first sales and a few reviews, Ring 3 has proof to show.

## Rules and care
- **Tab sends nothing to Ring 0.** These are the owner's own relationships. Tab can draft the message and a short list of who it might suit, and the owner sends it himself. This keeps us within "no DMs to strangers" and his sign-off rule.
- **PECR / UK GDPR:** genuinely personal one-to-one messages from someone to people they know are not the bulk "unsolicited marketing email" PECR reg. 22 targets, but this isn't checked against ICO guidance for this case. To stay clearly safe, send to people who know him personally, one at a time, and stop if anyone states no. Never turn the list into a mailing list without consent.
- **No case content in the pitch**: no accusations, no named people, nothing about holdings. It's a farm history book.
- **Measure:** give Ring 0 its own Stripe link or promotion code (stripe-promo-codes-per-channel-utm-passthrough.md) or a UTM (`utm_source=personal&utm_medium=message`). Then the first sale is attributed and we learn which ring works.

## How we apply it to Cedwyn
- **Status** (sales-fix log 31 Oct): 0 sales. The buy path works. Reach: IG 0, FB 1, Mailchimp 1; /ebooks/ gets about 6 GB visitors a day. Every Ring 3 fix so far (titles, SEO, posts) is correct but slow at this scale.
- **Proposal for the root to put to the project lead** (one decision only he can make; it's a PROPOSAL, nothing sent):
  1. Tab drafts a 4-line personal note (free edition link + £4.99 support link with a "personal" UTM or promo code) in Welsh and English.
  2. the project lead sends it himself, one to one, by WhatsApp or email, to 10–20 people who know him and the farm.
  3. Tab watches Stripe and Gumroad, attributes sales by the code, and asks each buyer (through the project lead) for a one-line review for /ebooks/.
- **Why the sales-fix flywheel should rank this first:** it's the only lever that doesn't wait on Google, costs nothing, and the research (Mollick, Agrawal et al.) states the first purchase is the one most driven by the founder's own network.
- Until the project lead states yes, the sales-fix runs keep improving Rings 1–3 and log "Ring 0 awaiting owner" instead of retrying cold fixes that already failed.

## Sources
- Mollick, E. (2014) The dynamics of crowdfunding: leeds-faculty.colorado.edu (read 31 Oct 2026); journal: sciencedirect.com
- Agrawal, Catalini & Goldfarb (2013) NBER w19133: nber.org (read 31 Oct 2026)
- Nathan Barry: nathanbarry.com and nathanbarry.com (read 31 Oct 2026)
- Lavingia, The Minimalist Entrepreneur (2021); summary: merlin.rebrovic.net (snippet only, not read)
- Internal: the project workspace (funnel, 31 Oct)