Jonathon's AI Wiki

AI Automation Client Acquisition Playbook

10 min read

Source: 7-day plan w/ Christian, Sav cold-email framework, 3-method playbook (cold + referrals + Trojan horse)

A consolidated playbook for landing AI automation clients without an audience or case studies — synthesized from three creator interviews on the same channel covering: a 7-day warm-outreach sprint that signed Christian his first 500K in pipeline as a full-time student in six months, and a no-content roadmap built around three acquisition channels (cold, referrals, Trojan horse partnerships). The pattern is directly applicable to any AI services business — including how WEO Marketly and similar AI marketing agencies land new dental and healthcare practice clients.

Key Takeaways

  • Sell outcomes, not workflows. Every source converged on the same mindset shift: position as a guide solving a business pain, never as a “template salesman.” Christian’s first close happened days after this reframe.
  • Trust precedes retainers. Don’t chase $10K/mo agreements before delivering one round of value. Get the foot in the door free or low-cost, deliver a measurable win, then expand or convert to retainer.
  • Volume negates luck — but only after the offer is right. Sav recommends 450 emails/day across 3 domains × 15 inboxes (~9K/mo) for beginners. Below 2% reply rate = lead list or deliverability problem. 2-5% = copy/offer problem. 5-10% = golden zone.
  • Niche databases beat Apollo for cold lists. Industry-specific directories (e.g., American Institute of Architects) come pre-filtered, eliminating the guessing game on filters. Ask Perplexity/Claude to find them.
  • The zero-risk offer is the whole game. “You don’t pay until results, no contract, only ask is permission to use you as a case study” doubled Sav’s reply rates and is the same offer pattern the 7-day plan recommends for the free pilot.
  • Three channels, in order: warm → cold → partnerships. Warm referrals first (statistically convert better, no proof needed). Cold outreach second (once you have one case study). Trojan horse partnerships third — partner with agencies/consultants who already have the trust, offer free AI audits to their clients on a 20% rev-share — this closes 46% faster than other deal types per the third source.

The Cold Email Framework

Volume + infrastructure. Don’t blast from one inbox. Stand up 3 domains × 5 inboxes each, send 30/inbox/day = 450/day = ~9,000/mo. Verify the list (Million Verifier or any mail-finder) before sending so you don’t burn deliverability on bounces.

Lead lists — niche databases first. Before Apollo or Sales Navigator, spend 10 minutes asking Perplexity or ChatGPT: “Where are some niche databases that hold a bunch of [target] companies?” Industry directories (AIA for architects, coaching directories, dental association rosters, etc.) are pre-filtered — there’s a 0% chance of off-target leads inside them. Then enrich with Apollo to pull decision-maker emails.

Personalization at scale. Sav’s exact stack: Google Sheet of leads → Make.com 4-module automation → Perplexity generates a short research report on each lead → ChatGPT writes a casual personal-sounding icebreaker → result writes back to the sheet. Four modules. Nothing fancy.

Subject line = “what someone on their team would say.” Avoid pitch-flavored subjects (instant unread). Use cliffhanger or insider phrasing. Examples Sav used:

  • Q is [company] taking on more clients?
  • Q from your neighbor
  • [firstName] Q

Body structure (verbatim from Sav’s word-for-word):

  1. Personalized icebreaker (top line — earns the open)
  2. Social proof in one line (“recently helped Sarah’s TNW scale to six figures with AI-driven automations”)
  3. The offer with desired outcome + timeframe + risk reversal (“23 booked discovery calls a month — you wouldn’t owe me anything unless it actually drives results”)
  4. Low-pressure ask: “Can I send over a 90-second Loom?” — never “30 minutes on your calendar”
  5. PS line that adds local/contextual color

Day-of-week & cadence. The sources don’t specify send day. They emphasize daily volume + tracking, with a weekly or bi-weekly iteration cycle on copy and offers based on reply data.

Closing With Proof

Every source said the same thing: the first question on every prospect call is “Who have you done this for?” No proof = uphill the entire conversation.

Build the proof before the cold campaign. Both Christian and Sav started with warm work — friends, family, a salon, a small medical-research helper. The work didn’t have to match the eventual offer. What mattered was being able to say “I helped X get Y result.” Sav explicitly traces the inflection point of his $500K pipeline to one free pilot for a video agency: he spent 3-4 hours on Zoom literally pointing at where to click their mouse, and that single case study line in his outreach nearly doubled reply rates.

The pilot offer that closes. Across all three videos:

“I’d love to build you a small automation that tackles [specific pain point] as a free pilot. My goal is just to prove these workflows save you time. In return I’d ask for honest feedback — and if it works, permission to reference your company.”

That phrasing removes every objection: no payment, no contract, no calendar block. The reply rate effect is mechanical — they’re not committing to anything.

The Loom video close. Instead of asking for a 30-minute call, send a 2-minute Loom walking through what you’d build for their specific business. People who watch the Loom and then book a call arrive nearly pre-sold — they already know your offer, your demeanor, and your competence.

The 7-Day Timeline

Christian executed this exact loop and closed his first $1,500 client on day 5:

  • Day 1 — Direction + trust map. Pick a loose hypothesis (“I help small businesses automate boring repetitive tasks with AI” — not “AI for dental clinics in Georgia”). Open a Google Sheet and list 20 people where trust already exists: friends with businesses, ex-coworkers, community members, second-degree connections.
  • Days 2-3 — 5-10 warm conversations. Not pitch calls. “I’m trying to start a business helping companies automate repetitive work with AI — could I ask a few questions about where things feel manual or annoying day-to-day?” If you have no business contacts, ask “do you know anyone this might benefit?” — softer wording, borrows trust without selling to friends.
  • Days 4-5 — Pick one prospect and propose the free pilot. Find the clearest, most painful repetitive task from your conversation notes. Send the zero-risk offer. Build the smallest possible MVP — the goal isn’t impressive tech, it’s “I saved you time on X.” Capture the exact language they use to describe the problem and the win.
  • Day 6 — Deliver and measure. Build the MVP. Track one simple before/after metric.
  • Day 7 — Decide the next step together. Two clean options: maintenance (“I’ll keep this running and handle small tweaks for $X/mo”) or expansion (“I noticed this related process — want me to scope it?”). If they don’t want either, ask for a short video testimonial and a referral. If the pilot didn’t land, ask why, fix it, run the loop again.

The loop is meant to be run multiple times. Each cycle yields better positioning, sharper language, and another case study — then cold outreach scales cleanly because you have proof.

Platform-Specific Prospecting: Google Maps, Kickstarter, Amazon/Shopify

Added 2026-07-16, from a motion-design-specific case study pairing this playbook’s cold-outreach mechanics with Higgsfield MCP video generation. Complements — doesn’t replace — the niche-database sourcing method above; this is a finer-grained targeting heuristic for three specific platforms.

A single-creator case study built a “client finder skill” for Claude that scrapes and filters three source platforms differently, then bulk-sends via Claude’s Gmail connector:

  • Google Maps. Local business categories (restaurants, boutiques, gyms, coffee shops, studios) filtered down to active emails belonging to owners/head managers, cross-checking other public platforms when an email isn’t listed directly on the Maps profile.
  • Kickstarter. Deliberately skips two extremes: zero-funded campaigns (no market validation, no working budget to pay a vendor) and multi-million-dollar campaigns (usually already locked into long-term agency contracts or an in-house team). The target band is campaigns that have raised 10-50% of their goal — enough validation and budget to spend, but still needing an explainer video or product animation and without an incumbent vendor relationship.
  • Amazon / Shopify sellers. Established storefronts with active sales and strong reviews, but still relying on basic product photos and text — motion content is a clear differentiation offer for exactly this segment.

The bulk-send mechanic. Once the skill compiles a list (200 emails across the three platforms in the case study — roughly 100 from Google Maps, 50 from Kickstarter, 50 from Amazon/Shopify), the creator connects Gmail via Claude’s connector (Settings → Connectors → Gmail, the same pattern as any other connector), pastes the full email list into the prompt, and asks Claude to “send this email to every contact on this list” — the Gmail connector personalizes and sends all of them in about two minutes.

Case-study economics (single creator, self-reported, not independently audited — treat as illustrative, not a benchmark): ^[single self-reported case study, not independently verified]

  • 200 cold emails sent; 16 replies within 14 hours; 7 leads moved forward after hearing price.
  • Of those 7: 1 refused to pay (expectation mismatch on the result), 1 required heavy revision but stayed, 5 paid in full (260 / 240 / 1,280 total revenue**), plus a free 5-image product-photo bonus thrown in for the largest client.
  • Total time invested across finding clients, outreach, and generating every deliverable: ~5 hours.
  • Total cost: 5,935 Higgsfield credits (1,030 in a single day** — roughly a 5x return on subscription spend and an effective **$256/hour**.
  • Framing tip for matching creative style to client type: “classic motion” for digital products, “hyper motion” for physical products — a quick vocabulary for scoping the pitch to what a given client actually sells.

This is one dated, self-reported case study from one creator — not a verified benchmark. It’s useful as a concrete existence proof of unit economics for a Higgsfield-MCP-based motion-design offer layered on top of this playbook’s general cold-outreach mechanics, not as a guaranteed result.

Try It

  1. List 20 warm contacts in a Google Sheet today. Note industry, relationship strength, whether they’re a potential client, intro source, or insight source.
  2. Send 5 low-pressure conversation requests this week with the curious-entrepreneur framing — no pitching, just listening for repetitive-work pain points.
  3. Pick the clearest pain and offer a free pilot with the zero-risk script: no payment, no contract, ask only for permission to reference if it works.
  4. Use Perplexity to find one niche directory for your target vertical before touching Apollo. Build a clean 500-lead list from it.
  5. Stand up minimum cold infrastructure when you have one case study: 3 domains, 15 inboxes, ~450 emails/day, track reply rate weekly. Below 2% = list/deliverability problem; 2-5% = copy/offer problem.
  6. If offering AI-generated motion/video content specifically, try platform-targeted sourcing instead of a generic niche database: Google Maps for local business categories, Kickstarter campaigns in the 10-50%-funded band, and Amazon/Shopify sellers still using only basic photos.

Graph View

Backlinks