Free: the complete Franchise System Blueprint. Get it →

Skip to content
The Lead Engine Playbook

Making the Phone Ring Is a System.
Here Is the Whole Build Order.

What a franchise actually sells you is lead flow. The systems behind it are not a secret and they are not proprietary — there are about a dozen pieces, and they only work if you build them in the right order. Most trade businesses that fail at marketing did not pick the wrong channel. They lit the channels before the foundation was under them, spent money they could not trace, and had no way to tell the difference between a bad channel and a badly run one. This is the whole engine, in the order to build it: foundation, then channels, then measurement.

Realistic budget: $150–$400/month in software and tracking, plus ad spend. A serious first month of paid lead generation in a mid-size metro runs $2,000–$5,000, and some of that is tuition. This page is largely about making sure the tuition buys you something.

One rule governs the entire build

Never spend a paid dollar you cannot attribute.

Not because tracking is virtuous, but because untracked spend teaches you nothing. Every channel below will lose you money before it makes you money. Instrumented, that loss is a lesson with a price tag. Uninstrumented, it is a donation.

Phase One

The foundation. Four things, before any spend.

Buying leads before this is in place is like hiring a dispatcher with no trucks. The leads will arrive, and they will land on a site that does not convert, a profile that does not rank, a business with four reviews, and a phone that cannot tell you where the call came from. Two to three weeks of work. Almost all of it free or close to it.

01

A website that gets calls, not compliments

Nobody browses for a plumber. A homeowner with water coming through the ceiling opens three tabs, scans them in about eight seconds each, and calls whichever business looks least likely to waste their afternoon. Your site has one job: make that decision easy and make the call one tap away. Everything else on it is decoration.

What good looks like

  • Phone number in the header on every page, tap-to-call on mobile — not buried on a contact page
  • The service and the city in your H1, because that is the search someone actually typed
  • Reviews and a Google rating visible before anyone scrolls
  • Real photos of your trucks, your crew, your finished work — stock photography reads as a middleman
  • A form with five fields at most: name, phone, ZIP, service, and a note
  • Loads in under two and a half seconds on a phone on cell data, which is where the emergency traffic is
What it costs
$16–$50/mo on a builder, or $2,000–$6,000 once for a built site

The mistake everyone makes: Designing the site for yourself instead of for someone in a hurry. You will be tempted by a full-screen hero video and an About Us story. Neither has ever booked a job. The businesses winning on the same street have an uglier site with the phone number bigger.

02

Google Business Profile, done all the way

For most local trades this is the single highest-return asset they own, and most of them fill in half of it. The map pack sits above the organic results and, on a phone, above almost everything. It is free. It takes an afternoon. Do it before you spend a dollar anywhere.

What good looks like

  • Primary category chosen precisely ("HVAC contractor" is not the same as "air conditioning repair service"), plus every honest secondary
  • Service area set by the towns you actually serve, not a lazy 50-mile radius
  • Hours that match reality, including a separate emergency-hours entry if you run one
  • Twenty-plus real photos, added on an ongoing basis — geotagged job shots, the van, the team
  • Services listed individually with price ranges, so you show up for the specific job
  • Messaging turned on only if someone will answer it, and Q&A seeded with the five questions you get every week
What it costs
Free. About four hours to set up properly, then fifteen minutes a week

The mistake everyone makes: Stuffing keywords into the business name — "Dave's Plumbing | Emergency Plumber Baltimore." It works, right up until a competitor reports you and you lose the profile generating half your calls. The suspension appeals process is not a place you want to spend a fortnight.

03

Review generation, automated from day one

Reviews are not a vanity metric in the trades. They are a ranking input for the map pack, a ranking input for Local Services Ads, and the thing a homeowner uses to break a tie between you and the other two tabs. Every channel further down this page gets cheaper as your review count and recency go up. Which is exactly why this belongs in the foundation and not in a "someday" list.

What good looks like

  • The ask fires automatically when the job is marked complete — not when you remember
  • Send it by text within the hour, while the tech is still in the driveway and the relief is fresh
  • One link, straight to the Google review form, no landing page in between
  • Target three to five new reviews a month, forever — velocity matters as much as total
  • Reply to every review, including the bad ones, in a tone you would be happy to have screenshotted
  • Never gate reviews behind a "were you happy?" filter. It violates Google's policy and it is the kind of shortcut that costs a profile
What it costs
Usually included in your job management software, or $50–$100/mo standalone

The mistake everyone makes: Doing one review push, collecting twenty in a fortnight, and stopping. Forty reviews with nothing since spring reads worse to a homeowner than fifteen with three from last month. Recency is the signal.

04

Tracking, installed before you spend a dollar

This is the step everyone skips and the only one that is genuinely non-negotiable. Advertising without attribution is not marketing, it is gambling with extra steps. You cannot cut what is not working if every lead arrives as an anonymous ring on the same phone.

What good looks like

  • A separate call tracking number for every channel, with dynamic number insertion on the website
  • UTM parameters on every paid link, and hidden fields on every form that carry them through
  • Job management software or a CRM that stamps a lead source on every single job — and a rule that no job gets closed without one
  • A "how did you hear about us?" question your office staff actually asks, as the backstop for word of mouth
  • Conversions defined and imported into Google Ads and Meta, so the platforms optimise toward booked work rather than form fills
What it costs
$50–$150/mo for call tracking; the rest is configuration time

The mistake everyone makes: Turning ads on Monday and planning to sort tracking out "once we see whether it works." You will never sort it out. You will have three months of spend, a vague sense that the phone rang more, and no way to tell which half of the money to keep.

That was the foundation. The channels are where the money goes.

Everything below this line has a price and a learning curve, and the learning curve gets paid for in ad spend. We have already paid ours across trade businesses in a dozen markets. Thirty minutes on a call and we will tell you which channel to light first in yours.

Phase Two

The channels, in the order to light them.

This order is not a ranking of which channel is best. It is the order that gets a new trade business to a working engine fastest — cheapest intent first, slowest compounding asset started early, demand creation last. Add one at a time and give each one long enough to produce a real number before you add the next.

01

Google Local Services Ads

Pay per lead

Light this one first. Local Services Ads sit at the very top of the page — above the search ads, above the map — carrying the Google Guaranteed badge, and you pay for a lead rather than a click. For a business that can answer the phone, it is the closest thing to a fair deal in local advertising.

  • Getting in means a licence check, an insurance check and a background check on the business and its techs. Start the paperwork before you need the leads; the queue is the slow part.
  • Ranking is driven by proximity, review count and score, your responsiveness, and whether your listed hours say you are open. Phase one feeds this directly.
  • You can dispute leads that were a wrong number, out of your area, spam, or a service you do not offer. Google credits the good disputes back.
  • Budgets are weekly rather than daily, and the system will spend ahead on a busy day and even out later.
What it costs
$25–$90 per lead, set by trade and metro
Time to results
1–3 weeks to verify, then leads within days

The mistake everyone makes: Not disputing. Most operators leave real money on the table every month because reviewing leads takes ten minutes on a Friday and nobody does it. The second mistake is letting calls ring out — Google is watching whether you pick up, and an unanswered lead is charged for and then quietly demotes you.

02

Google Search Ads

Pay per click

Search ads buy intent, not attention. Someone typing "AC not cooling" at 2pm in July is not researching — they are choosing. The upside over Local Services Ads is control: you decide the exact words, the exact hours, the exact towns. The downside is that the same control lets you burn a thousand dollars a week on people who were never going to call.

  • Start with exact and phrase match on a short list of emergency and replacement terms. Broad match is a graduate-level tool and it will happily spend your budget while you learn.
  • Build the negative keyword list on day one: jobs, salary, hiring, courses, DIY, how to, free, cheap, parts, wholesale, and every town you do not serve.
  • Send the click to a page about the thing they searched for. Furnace repair traffic landing on your homepage is money spent to make someone hunt.
  • Use call extensions and call-only campaigns for emergency intent, and schedule ads to the hours a human answers unless you genuinely run 24/7.
  • Let it gather data before you judge it. Fifteen clicks is not a verdict on a keyword.
What it costs
$8–$35 per click on emergency trade terms; $60–$200 per lead
Time to results
2–6 weeks of tuning before the numbers stabilise

The mistake everyone makes: Switching on broad match with "maximise conversions" bidding on a brand new account. You are asking an algorithm to optimise against no conversion history, and it will spend two thousand dollars finding that out on your behalf.

03

Local SEO

Time, or a monthly retainer

Third, because it pays late and pays longest. Paid channels stop the day you stop the card. Rankings you earn keep sending work for years, and the marginal cost of the hundredth lead is nothing. Start it early precisely because it is slow — the businesses that dominate a metro started this two years before you noticed them.

  • Post to your Google Business Profile weekly. Job photos, seasonal offers, a note about the storm. It is a small ranking signal and a strong trust signal.
  • Get your name, address and phone identical across every directory that lists you. Inconsistent citations quietly suppress the map pack.
  • Build one real page per service per city you seriously want — "furnace repair in Towson" with local photos, local job examples, the actual pricing conversation.
  • Review velocity from phase three of the foundation feeds this too. Local SEO and reviews are the same flywheel.
What it costs
Your hours, or $500–$1,500/mo; citation tools $100–$300/yr
Time to results
3–9 months before it moves, then it compounds

The mistake everyone makes: Spinning up forty thin city pages with the town name swapped and calling it coverage. Google has been catching that trick for fifteen years, and thin pages can drag the whole domain down. Ten pages you would be happy for a customer to read beat forty you would not.

04

Angi, HomeAdvisor and Thumbtack

Pay per shared lead

The honest framing: these are a volume tap, not a foundation. The same homeowner is sold to three or four contractors at once, so you are not buying a lead, you are buying a place in a race. Speed to call decides it. Useful while Local Services Ads verification grinds and SEO warms up, and useful for filling a slow week — but a business built on shared leads is renting its customer base at a rising price.

  • Call inside five minutes or do not bother buying. The contractor who dials first usually books it, regardless of who was cheaper.
  • Turn off the categories that send you jobs you do not want. Every platform defaults to sending you everything.
  • Pause the budget outside your answering hours. A lead bought at 2am is a lead your competitor calls at 8:01am.
  • Dispute the junk. The platforms have credit processes and, as with Local Services Ads, most operators never use them.
What it costs
$15–$100 per lead, plus an Angi membership
Time to results
Immediate — leads start the day you turn it on

The mistake everyone makes: Buying more leads than you can answer, then concluding "the leads are bad." Some are. But the gap between a 15% booking rate and a 45% booking rate on the same shared lead is almost always speed and script, not lead quality.

05

Meta Ads (Facebook & Instagram)

Pay per click or impression

Everything above captures demand that already exists. Meta creates it. Nobody scrolling Instagram is having a plumbing emergency — but plenty of them have been thinking about the bathroom for two years, or would book a $99 pre-season tune-up if reminded before the first cold night. Cheaper leads, colder leads, and a much harder follow-up.

  • Sell a specific offer with a date on it. "Pre-season furnace check, $99, through October 31" outperforms anything about your company values.
  • Best fit for planned work: remodels, replacements, water heaters, heat pumps, rebate-driven upgrades, seasonal maintenance plans.
  • Retarget every visitor your site already had. It is the cheapest audience you will ever buy and it closes far above cold traffic.
  • Native lead forms convert cheaply and convert cold. If nobody calls that lead within minutes, you have bought a spreadsheet, not a customer.
  • Refresh creative roughly monthly. Meta audiences fatigue faster than search terms do.
What it costs
$500–$1,500/mo minimum to learn anything; $20–$70 per lead
Time to results
2–4 weeks per test, with fresh creative every month

The mistake everyone makes: Running "We do plumbing — call today!" brand ads. That is a billboard with a worse view. Give them a reason and a deadline, or do not run it.

Phase Three

Measurement. The part that turns spend into knowledge.

You installed the plumbing for this back in phase one. This is what you do with it. Five numbers, one weekly habit, and a willingness to switch off a channel you were emotionally invested in.

Call tracking on every channel

One tracking number per channel, with dynamic number insertion swapping the number on your website based on how the visitor arrived. Record the calls. Then actually listen to ten a month — half of what looks like a lead problem turns out to be a booking problem, and you will only ever find that in the recordings.

$50–$150/mo for a small number pool

UTM discipline

One naming convention, written down once, used forever. Lowercase, hyphens, no spaces: utm_source=google, utm_medium=cpc, utm_campaign=emergency-plumbing. Hidden form fields carry those values into the CRM so a lead arrives already labelled. Two people inventing their own conventions produces data nobody trusts, which is the same as no data.

Free. Costs discipline, not money

Cost per booked job, not cost per lead

Cost per lead is the number every platform shows you, and it is the number that misleads you. What you actually pay for is a job on the calendar. Track spend, leads, booked jobs and revenue by channel, then judge on the cost of the booked job and what that job was worth.

Marketing under 10–15% of the revenue it produced

Speed to lead

Minutes between a lead arriving and your first contact attempt. Measure it, publish it, and hold the number. The odds of reaching a homeowner fall off a cliff after the first few minutes — they are calling down a list, and the list stops when someone answers. This one metric will move your close rate further than any bid adjustment you ever make.

Under 5 minutes, every lead, every channel

A weekly scorecard

Fifteen minutes every Monday with the same seven columns. Not a dashboard project, not a BI tool — a spreadsheet you fill in by hand at first, because filling it in by hand is what makes you notice things. Monthly is too slow to catch a channel going bad; daily is noise.

15 minutes, same time every week

The scorecard, and why cost per lead lies

Here is a month from a residential trade business running three channels. Read the cost-per-lead column, pick the winner, then read the last column and pick again.

ChannelSpendLeadsCost / leadBookedBook rateCost / job
Local Services Ads$2,00034$591956%$105
Google Search Ads$1,80022$82941%$200
Shared leads$90030$30517%$180

The cheapest leads on that table produce the second most expensive jobs. Shared leads at $30 look like the obvious winner right up until you divide by the five that booked. Cost per lead is the number the platform shows you. Cost per booked job is the number you are actually paying.

Add one more column once you have three months of history: revenue by channel. A channel can book cheaply and still lose, if everything it books is a $180 service call while another channel quietly delivers the replacements.

The honest part

You will lose money at first.
Every channel charges tuition.

The first fifteen hundred dollars you put into search ads mostly buys you a negative keyword list. Your first month of Local Services Ads teaches you which job types to switch off. Your first three Meta creatives will flop, the fourth will work, and you would not have guessed which. Your first two weeks of shared leads will convince you the leads are garbage, until you time your callbacks and discover the problem was that you called at four.

This is not a sign you are doing it wrong. It is the cost of information about your specific market, your specific trade and your specific price point — and nobody can hand it to you from a blog post, including this one. Anyone claiming their system prints money from week one is selling you something or has not opened the spreadsheet.

What matters is not avoiding the losses. It is making sure they buy you something. A tracked loss narrows the field: this keyword is dead, that town does not convert, this offer beats that one. An untracked loss teaches you nothing at all, and you will pay the same tuition again next quarter. That is the entire reason measurement comes before spending in this playbook.

So budget the tuition on purpose.

Pick a number you can genuinely afford to lose over ninety days — for most trade businesses that is somewhere between $3,000 and $10,000 — and spend it deliberately, one channel at a time, with every dollar reporting back. Then forge ahead. Instrumented.

Two doors

Somebody has to pay for the lessons.
The only question is whose money.

Door one — build it yourself

Take this playbook and pay the tuition in your own ad spend.

It genuinely works — that is why we published all of it rather than hiding it behind a sales call. You will spend a few months and a few thousand dollars finding out which keywords are junk in your market, which lead sources actually book, and which offer moves your phone. Plenty of good businesses have walked this path. It is slower and it is yours.

Door two — use ours

We have already lost thousands of dollars perfecting this exact system.

Across trade businesses, on real budgets, in real markets. The negative keyword lists, the channels that do not pay, the offers that do not convert, the follow-up cadence that does — all of it already found and already bullet-proofed. You skip the tuition and start from the version that works. On your brand, flat monthly, no franchise fee and no royalties.

Waste your own money learning, or use ours. We already spent it.

Free. 30 minutes. Bring your numbers and we will tell you which channel to light first — even if you go on to build it yourself.

Costs and per-lead figures are approximate as of mid-2026 and vary widely by trade, metro and season — treat them as a planning range, not a quote. The scorecard example is illustrative, built on typical residential trade numbers, not a specific client's results. FlexLever is not affiliated with any platform named on this page.