# Local Referral Engine > Builds referral flow and local partnerships for a local service business - the ask-after-five-star-moment rule, two-sided referral incentive design, complementary-trade partnerships, and neighborhood-domination tactics like job-site signage and just-worked-on-your-street cards. Use when an owner asks "how do I get more word-of-mouth", "should I pay customers for referrals", "how do I partner with other trades", or "how do I get more jobs in the same neighborhood". Do NOT use for gym referral and affiliate programs - use referral-and-affiliate-system instead. Do NOT use for formal B2B partnership deals - use partnership-strategy instead. - Skill: `skillmedev/local-referral-engine` (Agent Skill) - Install (CLI): `npx skillmds@latest add skillmedev/local-referral-engine` - Raw SKILL.md: https://api.skillmd.com/api/skills/skillmedev/local-referral-engine/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Docs & Writing - Author: SkillMedev (https://skillmd.com/u/skillmedev) - Updated: 2026-09-22 - Page: https://skillmd.com/skills/skillmedev/local-referral-engine --- # Local Referral Engine Referrals are the best customers a local service business gets: near-zero acquisition cost, pre-sold trust, and higher close rates than any paid channel. The costly mistake this skill prevents is leaving referrals passive - "we get word of mouth" - instead of building the three systems that manufacture them: a timed ask, a two-sided incentive, and a partner network. Run local-unit-economics first to see the prize: Summit Plumbing (4 techs, $450 average ticket, $75,000 a month) pays about $119 to acquire a paid-lead customer, so every referred customer is roughly $119 of margin that paid channels never touch. ## Operating procedure ### Step 1: Ask at the five-star moment - and only then The ask-after-five-star-moment rule: the referral ask fires when a customer has just expressed delight - a five-star review posted (review-engine detects exactly this), an effusive thank-you at job completion, a "you saved us" text. Never ask a merely-satisfied customer; a referral is the customer spending social capital with a friend, and they only spend it at an emotional peak. This is the concrete handoff from review-engine: every five-star review triggers the referral SMS within a day or two. ``` Referral ask SMS (1-2 days after a five-star moment): "Hi [FILL: name], [FILL: owner/tech] from Summit Plumbing - thanks again for the kind words! Small favor: most of our work comes from neighbors recommending us. If a friend ever needs a plumber, send them our way - and as a thank-you, you'll both get $50 off your next service. Just have them mention your name." ``` ### Step 2: Design the incentive two-sided Two-sided beats one-sided: reward the referrer and discount the friend. A one-sided reward makes the recommendation feel like a commission ("is he recommending them for the fifty bucks?"); two-sided lets the referrer open with a gift ("use these guys, you'll get $50 off"), which preserves the social framing that makes referrals work. Summit's design: $50 service credit to the referrer, $50 off the friend's first job. Against a $450 average ticket at 55 percent margin, a $100 total incentive still lands the customer far below the $119 paid CAC - and the referred customer closes better and refers onward. Size the incentive at roughly 10-25 percent of average ticket total across both sides. Keep redemption dead simple: "have them mention your name," tracked in a spreadsheet - no codes, no portal. ### Step 3: Build complementary-trade partnerships Trades see each other's future customers constantly: the electrician spots the corroded water heater, the GC needs a plumber on every remodel, and the plumber sees breaker panels. Build a bench of 3-5 non-competing partners - for Summit: an electrician, a GC, an HVAC company, a restoration company. Rules that keep it alive: - Reciprocity is the currency. Track referrals given and received per partner each quarter; a partner who receives five and sends zero gets a friendly conversation, then gets dropped. - Refer only businesses you would let into your own house - their bad job becomes your bad review. - Money between trades: a flat thank-you referral fee ($25-50) or pure reciprocity both work; keep it simple and consistent. Check local licensing rules - some jurisdictions and trades restrict paid referrals. - Make the intro warm: the partner texts the customer your name, not just the reverse. A warm handoff closes at multiples of a cold mention. - These relationships also earn the real local links local-seo-playbook wants - a "trusted partners" mention on each other's sites is prominence that money cannot buy cleanly. ### Step 4: Dominate the neighborhood around every job Every job is a beachhead; density is the goal, because five jobs on one street cuts drive time and makes the brand ambient. - Job-site signage: a small yard sign during multi-day jobs, with permission ("mind if we leave this up while we work?"). A truck wrapped with the name and number is signage that never comes down. - "We just worked on your street" cards: after each job, the tech drops cards at the 10-15 nearest doors. Script on the card: "We just fixed a water heater for one of your neighbors on [FILL: street]. If your plumbing ever acts up, we're already in the neighborhood - $25 off for [FILL: street] residents this month." Neighbor proof plus a local-only offer outperforms any cold flyer. - Neighborhood groups: when a customer offers to post in the neighborhood Facebook group or on Nextdoor, say yes and make it easy - a recommendation there is a referral to three hundred neighbors at once. ### Step 5: Track referral share monthly Log every new customer's source at booking ("How did you hear about us?" is mandatory, not optional). Track referral plus partner share of new customers monthly and feed the repeat/referral rate into local-unit-economics. Healthy mature share for a residential trade: 30-50 percent of new customers from referral, partner, and neighborhood sources. Below 15 percent after six months of running this engine means the asks are not actually firing - audit whether the five-star trigger is being executed, not whether customers "just don't refer." ## Inputs to elicit Collect: current referral share of new customers (label the usual guess as a guess - booking-source logging is step zero), average ticket and margin (from local-unit-economics), review flow (is review-engine live - this skill's trigger depends on it), candidate partner trades already known, and jobs per month per neighborhood. Defaults: $50/$50 two-sided credit, cards at 10-15 doors per job, 3-5 partners. ## Concrete thresholds - Ask timing: within 1-2 days of a five-star moment; never cold, never in the same message as the review ask (one ask per message, always). - Incentive: two-sided, totaling 10-25 percent of average ticket; credits beat cash for repeat-visit trades because they guarantee a next booking. - Partners: 3-5 active; reciprocity reviewed quarterly; give-receive imbalance beyond 3:1 for two quarters means drop or renegotiate. - Neighborhood cards: 10-15 doors per completed job, same day. - Referral share target: 30-50 percent of new customers within a year; under 15 percent after six months means the trigger is not firing. ## Worked artifact: Summit's referral math ``` Program: $50 credit to referrer + $50 off friend's first job Five-star moments/mo (from review-engine, at its 12/mo target plus in-person delight): ~18 asks sent Referrals landed (25% of asks): ~4-5 new customers/mo Partner bench (electrician, GC, HVAC, restoration): referrals received: ~4/mo given: ~5/mo (healthy) Street cards: ~55 jobs/mo x 12 doors = ~660 cards -> ~3 jobs/mo New referred/partner/neighbor customers: ~11-12/mo of ~35 new customers = ~32% referral share (target band: 30-50%) Cost check: ~$100 incentive per referred customer vs $119 paid CAC, with a customer who closes better and refers onward. Verdict: scale. ``` ## Deliverable A referral operating system on one page: the five-star-trigger ask SMS with fields filled, the two-sided incentive spec and its tracking sheet, a partner bench list with the reciprocity ledger, the street-card copy and drop routine, and the monthly referral-share line that feeds local-unit-economics. ## Do NOT - Do not ask merely-satisfied customers for referrals; a lukewarm referral spends trust the customer does not have, and the silence teaches staff the program "doesn't work." - Do not run one-sided incentives; paying the referrer while the friend pays full price converts a recommendation into a commission. - Do not reward referrals for leaving reviews or blend the two asks - review incentives violate platform policy (review-engine owns that boundary); referral incentives are fine, so keep the programs separate. - Do not partner with a trade you have not seen work; their callback becomes your one-star review. - Do not canvass streets cold with generic flyers; the card works because of the neighbor-proof line, which requires an actual nearby job. ## Quality bar - Every ask traces to a logged five-star moment; no cold blasts. - Both sides of the incentive are stated in every ask message. - The partner ledger shows given and received per partner per quarter. - Booking staff log a source for 100 percent of new customers - the share number is only as real as the logging. ## Escalation and routing Check local licensing and consumer rules before paying inter-trade referral fees - some trades and jurisdictions restrict them; this is not legal advice. Route the five-star detection and review asks to review-engine, partner-site links into local-seo-playbook, gym-specific referral programs to referral-and-affiliate-system, and the monthly referral-share and incentive-cost math into local-unit-economics to confirm the program stays cheaper than paid acquisition.