Ask ten owners where their last twenty jobs came from and you will hear the same three names in a different order. The order is not random, it follows what the customer was doing when they decided to buy.
The short answer: if the job is urgent and local (a burst pipe, a lockout, a toothache), leads come from Google search and the map pack. If the job is visual and optional (kitchen remodel, gym membership, salon), paid social does the work. Referrals win for anything expensive and personal. Most small businesses should spend their first dollar on Google, not Meta.
The three channels do different jobs
Google answers demand that already exists. Someone types "emergency plumber near me" at 9pm with water on the floor. There is no persuasion happening; the customer has decided and is choosing between the three results in front of them.
Meta creates demand. Nobody scrolls Instagram hoping to find a landscaper. The ad interrupts a person who was not shopping, so the creative has to do the persuading.
Referrals skip both. Somebody the customer trusts already did the filtering, so the lead arrives pre-sold and usually closes at a higher price.
What each channel costs per lead
Published averages hide the only number that matters: cost per job, not cost per lead. The arithmetic is simple and worth doing on paper before spending anything.
- Google search: cost per click varies wildly by trade and city, and home services are among the most expensive clicks in US advertising. If a click costs $6 and one in ten clicks becomes a lead, the lead costs $60. If one in four of those leads becomes a job, the job costs $240 in media.
- Local services / map pack: the searcher is often ready to call, so the conversion rate from click to lead is far higher than a display click, and you pay in ranking work rather than per click.
- Meta: cheaper clicks, lower intent. A $1.50 click that converts at 1.5% is a $100 lead, and the job only pays for itself if the average ticket is well above that.
The point is not the exact numbers, which differ by market. The point is that a cheap click is not a cheap lead, and a cheap lead is not a cheap customer.
The order to spend in
- Google Business Profile, properly filled. Categories, service area, photos, hours, and a habit of asking every happy customer for a review. This is the highest-return hour a local business can spend, and it is free.
- A page per service, not one page for everything. Someone searching for "water heater install" should land on a page about that, not the homepage.
- A phone number that a human answers. Most local leads are lost in the first sixty seconds, not in the ad.
- Paid Google on the two or three services with the best margin. Not on everything.
- Paid social once the offer converts on the site. Paying to send traffic to a page that does not convert just buys expensive proof.
Where businesses waste the most money
Not in the ad platform. In the follow-up. A quote sent and never chased, a form submission nobody called back, an estimate that took four days. In most local trades the second call wins the job, and most businesses never make it.
Before increasing ad spend, count how many leads from last month were never contacted twice. That number is usually the cheapest growth available.
What a lead actually costs you
Cost per lead is a middle number. The one that decides whether the business grows is cost per customer, and it is a chain of three rates.
Ad spend divided by clicks gives cost per click. Clicks divided by enquiries gives conversion rate. Enquiries divided by jobs won gives close rate. Multiply the three and you have cost per job. A $1,000 ad budget that produces 500 clicks at 2% enquiry rate is 10 enquiries; if half become jobs, each job cost $200 in media before labour.
That number is either comfortable or fatal, and it depends entirely on the average ticket. A $200 media cost is fine on a $4,000 kitchen and ruinous on a $150 call-out.
Speed to lead is the cheapest upgrade
Callers who reach voicemail keep calling the next number in the map pack. The business that answers wins work it never had to advertise for. Two changes are worth more than any campaign: a phone that a human answers during business hours, and a form submission that gets a reply within minutes rather than days.
You can test this in a week. Take last month's enquiries, count how many were contacted within an hour, and compare their close rate with the ones contacted the next day. Most owners find the gap is bigger than any ad optimization they could run.
The review engine
Reviews move map rankings and they remove hesitation at the moment of choosing. The system that produces them is boring: ask every customer, on the day the job finishes, with a direct link. Ask in person first, then by text the same evening.
Two rules keep it honest. Never buy reviews, and never filter out unhappy customers by only asking people you know are happy. A reply to a bad review, written calmly and without excuses, does more for a local business than twenty five-star lines.
What to track every week
- Calls received, and how many were answered
- Form enquiries, and hours to first reply
- Jobs booked per channel
- Average ticket by channel
- Reviews requested versus reviews received
Five lines in a spreadsheet, ten minutes on a Friday. After six weeks you can see which channel produces profitable work and which one only produces activity.
When to add paid channels
Add Google Ads when the map pack is full and the calendar still has gaps. Add Meta when you have a visual offer, a clear price and a page that converts on a phone. Do not add either while enquiries are going unanswered, because you will pay for demand you cannot serve.
What a small budget should look like
Under $500 a month: profile, reviews, one service page per service, and follow-up. Between $500 and $1,500: one paid channel pointed at the two most profitable services, with call tracking so the spend can be judged. Above $1,500: split by service, not by platform, and kill anything that has not paid for itself in ninety days.
Why the third result wins more than the first
Position matters, but not as much as people assume once the searcher is comparing. A business in third place with forty recent reviews and a clear price range often beats the first result with six reviews and no prices.
That is good news for small operators: the map pack rewards signals you can control, not just budget. Photos, categories, service area, response time and review count are all things a competitor cannot buy their way past.
The offer problem, not the traffic problem
Most businesses that say "advertising does not work" have a page that gives the visitor nothing to act on: no price range, no timeframe, no next step. The click arrives and leaves in four seconds.
Fix the page first. One line saying what you do and where, a price starting point, three reviews, and a form that asks for two fields instead of seven. On a phone, that is the whole page.
Handling the leads you already have
The cheapest growth in any local business is calling back the people who already asked. Take last month's enquiries and mark what happened: quoted, won, lost, never contacted. The "never contacted" column is normally the largest and the most embarrassing, and it costs nothing to fix.
Set a rule: every enquiry gets a reply within one business hour, and every quote gets one follow-up at day three. That is it. No software required, though a simple reminder helps.
When a channel stops working
Channels decay. Costs rise as competitors arrive, and the same budget buys fewer calls each quarter. Watch cost per job monthly, not yearly, and be willing to move the money rather than defend the habit.
The businesses that survive a channel shift are the ones with a second source already running: referrals, a review engine, or an email list of past customers. Never let one platform be the only reason the phone rings.
The forty-eight hour rule
Every enquiry that is not answered within two days is a lead you paid for and gave away. Put a number on it: if ten enquiries arrive a week and three go unanswered for two days, that is fifteen lost enquiries a month at your average close rate.
Fix it with a rota, not software. One person owns the inbox each day, the backup is named, and anything unanswered by lunchtime gets a call. That single change usually beats the next thousand dollars of ad spend.
Verdict: Google first for anything urgent, local, or high-ticket. Meta second, for visual work and repeat customers. Referrals treated as a system, not luck: ask every customer, every time, on the day the work is finished. If you only have a few hundred dollars a month, the map pack plus follow-up beats any ad campaign.
Questions people ask
How long does the map pack take?
Reviews and complete profiles move faster than most people expect, usually weeks rather than months, but rankings are local and competitive. It is the cheapest channel and the one most businesses neglect.
Should a local business run ads at all?
Yes, once the basics convert: a clear offer, a phone that gets answered, and a page that loads fast on a phone. Advertising multiplies whatever is already there, including the problems.
Related: cold email that gets replies, and our website plans if the landing page is the weak link.