The Complete Non-BS Guide to Creating Your Business With AI

Do you own a company, or a job you cannot quit?

There is a hard difference between the two, and most owners never name it. A job you own is a business that only produces money when you personally swing the wrench. A company is a system that produces money whether or not you are on the truck today. You do not have to choose the second one. Plenty of great livings come from being a skilled solo operator. But you must decide on purpose, because the answer changes everything downstream: how many zip codes you target, how you price, whether you hire, and how much you can afford to spend to get a customer.

Write one sentence at the top of your plan: "In three years, I want a business that ___." Everything in Part I either serves that sentence or fights it.

Define your service area with data, not a gut feeling

Most owners draw their service area as "wherever I feel like driving." That is how you end up burning two hours of windshield time for a 90-dollar job in a neighborhood that will never call you again. Your service area is a business decision, and you can make it with free public data.

Start with income per zip code

Pull the median household income for every zip code within a reasonable drive. The US Census Bureau publishes this for free through the American Community Survey (ACS). Go to data.census.gov, search a zip code, and look at median household income and median home value. It is free, it is authoritative, and it is more honest than any hunch.

Owner-occupied versus rental, and why it matters

The ACS also reports the share of homes that are owner-occupied versus rented. This single number quietly decides whether a zip code is worth your time, because renters rarely buy the work you sell. A renter with a dirty carpet calls the landlord. A renter with a leaking water heater calls the landlord. The homeowner is the one who opens the wallet.

Read the housing stock for hidden demand signals

The homes themselves tell you how much work exists. This is where you separate a busy area from a profitable one.

A 20-year-old subdivision of owner-occupied homes at a solid income level is, for a plumber, a slow-motion pile of failing water heaters and worn-out fixtures. You now know that from data before you have knocked on a single door.

Demand is seasonal; plan for the valleys

Every home service has a rhythm. HVAC spikes in the first heat wave and the first cold snap and goes quiet in the shoulder seasons. Carpet cleaning jumps before the holidays and around move-out season. Plumbing is steadier but has emergency spikes in a hard freeze. Map your own year in twelve rows: which months are feast, which are famine. You will use this later to decide when to push budget (Chapter 5) and which services to promote when demand for your bread-and-butter dips.

Map your competitors before you pick a fight

You are not entering an empty market. Someone already services these zips. Your job is to see them clearly, then decide whether to meet them head-on or go around them.

• Who shows up. Search your money keywords and note who ranks in the map pack and the top organic results for each zip. These are your real competitors, not the ones you hear about at the supply house.
• Review count and rating. A competitor with 600 reviews at 4.9 owns the trust in that area. One with 40 reviews at 4.2 is beatable. Write down both numbers for each.
• Pricing visibility. Do they publish prices, "starting at" ranges, or nothing? Silence on price is an opening. So is a competitor who only competes on being the cheapest.

Compete head-on, or flank

You have two honest choices, and both work.

Fill out the Market Snapshot worksheet at the end of this chapter before you move on. It turns everything above into the short list of zips you will actually build for.

Your market tells you where the money is. Your consumer tells you how it moves. A water heater replacement is one job, but the person buying it might be terrified and typing with wet feet in a flooded garage, or calmly researching options three weeks before their unit finally dies. Sell to both the same way and you lose both. Here are the three buyers you will meet, over and over, in every home service.

The three buyers

1. The panic buyer

Something broke, and it is an emergency. Water on the floor. No AC in a heat wave. A sewage smell. This person is not comparing five quotes. They want someone competent to answer the phone right now and show up fast. Price is almost irrelevant in the moment; speed and trust win.

2. The planner

Nothing has broken yet, or it is a want, not an emergency: an aging furnace they know they should replace, a whole-home repipe, a deep carpet clean before hosting. This person researches, reads reviews carefully, and wants to feel like they made a smart, informed decision. They will wait days or weeks and talk to two or three companies.

3. The price shopper

They have decided what they want and now they are hunting the best number. Sometimes that is a savvy homeowner; sometimes it is a landlord or property manager buying in volume. They are useful business if your model is built for volume and efficiency, and a trap if it is not.

Match the buyer to the search, then to the page

This is the whole point of naming them. Each buyer maps to a different kind of keyword and a different kind of page, and that mapping is exactly what you will act on in Part II.

Map the consumer's perception journey

Knowing who the customer is gets you started. It is not enough. The panic buyer, the planner, and the price shopper are types, but a type does not book a job. A person does, and that person moves through a series of moments between the instant their problem appears and the instant they hand you money, then keeps moving after the job is done. At every one of those moments they are thinking something, fearing something, comparing you to someone, and waiting to believe one specific thing before they take the next step.

Your job is to know what that thing is at each stage and to put the proof of it directly in their path. That is the whole game, and it is worth saying plainly.

Here is the journey, start to finish. Ten stages. Read them as a chain, because a customer who cannot get past stage four never reaches stage seven, no matter how good your estimate is.

For each stage, ask the same seven questions: what is happening to the customer, what are they thinking, what are they afraid of, what could stop them from advancing, what do they need to believe next, what proof creates that belief, and what asset delivers that proof. Work through them one at a time.

1. Problem / Trigger

What is happening: something changed. A pipe burst, the AC quit in July, the carpet got a stain guests will see Saturday. Or a slow realization: the furnace is fifteen years old and groaning. What they are thinking: "How bad is this? How much is this going to cost me? Who do I even call?" What they fear: the problem getting worse, being taken advantage of because they do not know the trade. What stalls them: nothing yet, the trigger itself pushes them forward, but panic can freeze a person or send them to the wrong first move (a cheap patch, a handyman). What they must believe next: "This is solvable and there is a right kind of company for it." Proof that creates belief: plain-language content that names their exact situation. Asset: problem-focused service pages and blog answers that show up the moment they search (Chapters 22 and 24).

2. Search

What is happening: they type the problem into Google, usually on a phone. What they are thinking: "Who comes up? Who looks legit and close?" What they fear: picking a scammer, a no-show, an out-of-area outfit. What stalls them: you are not on the page they see, or you are buried below ads and the map pack. What they must believe next: "There are real, local, credible options for me here." Proof: a strong map pack presence, ranking on the terms they actually type, a visible star rating and review count. Asset: your Google Business Profile and rankings (Chapters 17 to 19), and the keyword-matched pages behind them (Chapters 6 to 8).

3. Discovery

What is happening: they land on your listing or your site for the first time. You have a few seconds. What they are thinking: "Is this the kind of company I want? Do they do my exact thing? Are they close?" What they fear: wasting time on a company that does not really handle their problem. What stalls them: a vague homepage, no clear service or area, a slow or ugly site, no phone number in sight. What they must believe next: "These people do exactly what I need, near me." Proof: a headline that names the service and the city, an above-the-fold phone number, a clean fast page. Asset: your homepage and service-page messaging and the conversion layout (Chapters 11 and 12).

4. Evaluation

What is happening: they are sizing you up on purpose now, reading. What they are thinking: "Are they any good? Have they done this before? Can I trust them in my home?" What they fear: incompetence, hidden damage, a stranger who is not vetted. What stalls them: no reviews, no proof of work, no licensing or guarantee, no faces. What they must believe next: "These people are competent and safe to let inside." Proof: a high review count with a strong rating, before-and-after photos, license and insurance, a written guarantee, real team photos. Asset: reviews and the proof assets on your pages (Chapters 17 and 18 for review velocity; conversion proof blocks in Chapters 11 and 12).

5. Comparison

What is happening: you are now in a shortlist of two or three, being compared side by side. What they are thinking: "Which of these is the best choice for me?" What they fear: picking the wrong one, overpaying, missing an obvious better option. What stalls them: you look interchangeable with the others, or worse on the one thing they weigh most, usually reviews or response speed. What they must believe next: "This one is clearly the better pick." Proof: a distinct positioning and a reason-why, more and better reviews than the alternatives, a stronger or clearer offer. Asset: your positioning and USP (Chapter 3) and your offer (Chapter 5), carried through your messaging.

6. Contact

What is happening: they reach out. Call, form, text, or booking. What they are thinking: "Will a real person help me? Will this be easy?" What they fear: voicemail, a runaround, feeling dumb for asking, pressure. What stalls them: no answer, a slow callback, a clumsy form, a rude or scripted-sounding rep. What they must believe next: "These people are responsive and easy to deal with." Proof: a fast human answer, a friendly competent first exchange, easy scheduling. Asset: your lead handling and phone process (Chapter 32), plus the click-to-call and forms on the site (Chapter 11).

7. Estimate / Quote

What is happening: they get a number, on the phone, on site, or in writing. What they are thinking: "Is this fair? Is it the final number? What am I actually getting?" What they fear: the classic one, the quote changing, surprise charges, being upsold. What stalls them: a vague verbal number, no breakdown, a price with no explanation of value, a pushy close. What they must believe next: "This price is fair and it is what I will actually pay." Proof: a clear written estimate, a stated price guarantee or "no surprise" promise, an explanation of what is included. Asset: your pricing structure and offer (Chapters 4 and 5) and how the estimate is delivered (Chapter 32).

8. Purchase

What is happening: they say yes and the work gets scheduled or done. What they are thinking: "Did I make the right call? Now do not let me down." What they fear: buyer's remorse, being forgotten, the company vanishing after the deposit. What stalls them: a hard yes that never turns into a scheduled visit, silence after they commit. What they must believe next: "I chose well and they have this handled." Proof: a confirmation, a clear next step, an on-time arrival, a professional who looks the part. Asset: your booking and dispatch process and the first touch of the job experience (Chapters 32 and 33).

9. Job Experience

What is happening: you are in their home doing the work. This is the product. What they are thinking: "Are they careful, clean, honest, and good at this?" What they fear: mess, damage, a job done badly, being talked down to. What stalls them: nothing stalls the job, but a bad experience kills the review, the referral, and the repeat, which is where the real money is. What they must believe next: "That was a great decision and I would use them again." Proof: the work itself, done clean and right, plus small professional touches, shoe covers, a walk-through, a tidy exit. Asset: your service delivery and retention process (Chapter 33).

10. Review / Referral / Repeat

What is happening: the customer becomes marketing, or does not. What they are thinking: "Would I recommend them? Will they remember me?" What they fear: being spammed, or on the other side, forgetting a company they liked. What stalls them: you never ask for the review, never follow up, never give them a reason to come back. What they must believe next: "These are my people for this now." Proof: a timely, easy review ask, a genuine thank-you, a reason to return or refer. Asset: your review-ask system (Chapter 18) and your retention and referral program (Chapter 33), which feed the very first stage for the next customer by growing your review count and reputation.

The journey in three trades

The stages are the same for every home service. What changes is the fear at each one and the proof that clears it. Here is where the same journey diverges across three trades, at the stages where the trade matters most.

Notice the pattern. The trade changes the words, but the barrier structure is identical: a fear at evaluation about competence and safety, a fear at estimate about the price changing, a job experience that decides whether they ever come back. Map your own trade this way and you stop guessing what to say.

How the journey feeds the rest of this manual

This is not a standalone exercise. The perception journey is the spine the rest of the manual hangs on. A fear you name here becomes an asset you build later. Follow the chain:

Read it as cause and effect. The fear you uncover at the evaluation and comparison stages tells you what to position around (Chapter 3) and what your USP must promise. That promise shapes the offer you make (Chapter 5) and the messaging on your site (Chapters 11 and 12). The fear about competence and safety is answered by reviews and proof (Chapters 17 and 18). The fear about difficulty and pressure is answered by an easy CTA and a good phone process (Chapters 11 and 32). The fear that the job goes badly is answered by the job experience and retention work (Chapter 33). Every downstream chapter is removing a barrier you can name right now.

Now put it on paper. The two worksheets that follow turn this chapter into a decision framework: one maps the full journey for your business, the other forces you to name the three specific beliefs standing between your ideal customer and a booked job.

Positioning is not your logo or your slogan. It is the one thing a customer thinks when your name comes up, before they have called you. Marketing does not create that thought so much as it decides which thought you aim for and then earns it, relentlessly, with everything you do. This chapter is the strategic core of Part I. It borrows the classic thinking of Ries and Trout and applies it to a plumbing truck.

First, assess your brand honestly

Before you claim a position, measure where you stand. The single cleanest test: do people search for your name? Open Google Keyword Planner or Search Console and look for branded searches like "[your company] plumbing" or "[your company] reviews." Volume there means you have real brand equity and people are choosing you on purpose. Near zero means you are still an unknown competing on category terms, and your positioning has to do more work.

Your USP must be provable, not a slogan

"Quality service and fair prices" is not positioning. Every competitor says it, so it says nothing, and worse, nobody can check it. That is the tell.

The test: could a competitor put your USP on their own truck with a straight face, and would a customer have any way to hold either of you to it? If the answer is yes and no, throw it out.

The kinds of USP you can own

There is no single right USP. There are several distinct types, and a strong home-service company owns one or two of them loudly rather than mumbling all of them. Here is the taxonomy, each with what makes it measurable and a weak-versus-bold rewrite.

1. Service-based USP

You do more, or do something extra, that competitors skip. It is measurable because it is a concrete, checkable deliverable.

2. Speed and response USP

You show up faster, guaranteed. Measurable because it is a clock the customer can start.

3. Guarantee and risk-reversal USP

You take the customer's risk onto yourself. Measurable because the guarantee has a defined trigger and remedy.

Most owners price by feel: they glance at what a competitor charges, shave a little off, and hope. That is how you end up working sixty-hour weeks at a margin that never lets you hire. Pricing is one of the highest-leverage decisions in the whole book. Do it in two moves: pick a model to enter the market with, then walk that price toward equilibrium, the highest number your target customer still says yes to. This chapter gives you both: five entry models to choose from, then the method for tuning your way to the sweet spot.

Five ways to set your entry price

There is no single correct pricing model. There is the right one for your trade, your market, and where your business sits today. Here are the five that matter for home services. Read all five, then use the chooser to pick your starting point.

Which model, which situation

Real businesses blend these per service line. It is normal to price emergency work value-based (Model C), routine maintenance at market-rate (Model B), and a brand-new add-on with penetration (Model D), all under one roof. The model is a per-service decision, not a company-wide identity.

Enter below median, then climb: the price matrix

Penetration and market-rate work together in a single planned climb. Write the raises down in advance and tie them to review milestones, not the calendar alone. This is a price matrix worked example for a carpet cleaner:

You launch below the median (Model D) to win the first 25 reviews fast, climb back to market-rate (Model B) once you have social proof, then price slightly above it once your reviews and positioning justify a premium (Model E). The same table works for a plumber: swap the rows for drain cleaning, water heater install, faucet replacement, and a service-call fee.

When NOT to discount

Discounting is a tool for buying trust, not a permanent identity. There is one situation where you should never lead with price: emergency work. When a panic buyer (Chapter 2) has water pouring into the garage, speed and certainty beat price by a mile. Discounting an emergency call leaves money on the table and signals that you compete on cheapness in exactly the moment the customer would happily pay a premium for a fast, sure fix.

Price equilibrium: the number you are actually hunting for

Every model above is a way to enter the market. Equilibrium is where you want to end up. Your price equilibrium is the highest price at which most of your target customers still say yes. It is a real number for your business, and finding it is the whole point of pricing.

Sit below equilibrium and two things go wrong at once: you leave margin on the table on every single job, and you fill the schedule with the wrong customers, the price shoppers who never come back and never refer. Push above equilibrium and your close rate falls off a cliff. The estimates keep coming; the yeses stop. The skill is landing right at the top of the band where the yes still happens.

How to find it: raise in small steps and watch your close rate

You cannot calculate equilibrium from a spreadsheet; you find it by moving the price and watching what happens. Track one number: your close rate, the share of estimates that turn into booked jobs. Then use this practical rule.

The growth logic: revenue per crew-hour, not jobs closed

Here is the counterintuitive part most owners miss. The goal is not to close the most jobs. It is to earn the most revenue per available crew-hour, because that is the number that funds everything downstream: the marketing budget in Chapter 5, and the review velocity in Chapters 17 and 18. When every job clears real margin instead of just covering costs, the surplus buys your growth. Watch the middle row win:

The 189 row closes the most jobs and earns the least. The 329 row prices out too many buyers. The 249 row closes fewer jobs than the cheap option yet earns the most revenue from the same ten estimates, and it does it with fewer crew-hours, leaving capacity to take on more leads. Fewer, better-priced jobs beat a full schedule of underpriced ones. That surplus per hour is what you reinvest.

Fill out both worksheets at the end of this chapter before you move on. The Price Matrix Builder turns the models above into an entry number for every service; the Find Your Equilibrium tracker walks that number to the sweet spot once the jobs start coming in.

Positioning is what you stand for. Pricing is what you charge. The offer is the specific, packaged thing you put in front of a customer and ask them to buy today. This is the last chapter of Part I because it is where strategy becomes something you can actually run an ad for and build a page around.

Build the offer

A strong offer is not just a price. It bundles the service, removes the customer's risk, and gives them a reason to act now. The parts:

• The core service, described in the customer's words, priced from your matrix in Chapter 4.
• A risk reversal that reflects your positioning: a written flat-rate quote, a satisfaction guarantee, a two-hour arrival window. This is where Chapter 3 becomes a promise.
• A reason to act now, honest and specific: a seasonal tune-up before the first freeze, a limited number of new-customer slots, a same-day booking bonus.

Now decide the budget, on purpose

Here is where most owners either spend nothing and stay invisible, or spend randomly and cannot tell what worked. You will pick one budgeting model and one number, and you will pick them deliberately. There are three sound models.

Model 1: percentage of revenue

The simplest. You spend a fixed share of revenue on marketing. The right percentage depends on your mode:

If you are new and following the market-entry plan in Chapter 4, you are in growth mode. Budget accordingly and do not flinch; you are buying a customer base and a review base that will pay back for years.

This percentage is not just an SEO budget; it is your total surface-area budget. Part III shows exactly where these dollars buy surface area: free organic placement, and paid placement across Google Ads, Local Services Ads, and the marketplaces (Chapters 25 to 27). Decide the number here; allocate it there.

Model 2: fixed cost per lead

Instead of a percentage, you decide the most you will pay for one qualified lead, then spend up to that on any channel that beats it. If a booked plumbing job is worth 400 dollars in gross profit and you close one in three leads, you can pay up to roughly 130 dollars per lead and still profit. This model keeps you honest channel by channel and connects directly to the attribution work in Chapter 31.

Model 3: lifetime-value-informed budgeting

The most powerful model, and the one most owners never use. You budget based on what a customer is worth over years, not on the first job. A carpet-cleaning customer who comes back twice a year for five years is not a 216-dollar sale; they are a 2,000-dollar-plus relationship. That completely changes what you can afford to pay to acquire them.

Fill out the Budget Picker worksheet at the end of this chapter before you move on. It commits you to a monthly number and checks it against what a customer is truly worth.

You finished Chapter 6 with a spreadsheet of keyword candidates. Now you face the question that decides your entire site structure: how many pages do I build, and which keywords go on which page?

Most owners guess. They build one page per keyword and end up with ten near-identical pages fighting each other, or they cram everything onto one page and rank for nothing. There is a precise, free method that removes the guessing. It is called SERP overlap analysis, and it works because you let Google tell you the answer.

The core idea

SERP means Search Engine Results Page: the ten blue links you see after a search. Google has already decided, for every phrase, which pages best answer it. So you compare the results for two candidate keywords and read Google's mind:

The exact method, step by step

1. Open an incognito browser window so your history does not skew results. Set your location if needed.
2. Search the first candidate, for example water heater repair phoenix.
3. Write down the top 10 organic URLs (skip ads and the map pack for now).
4. Search the second candidate, for example hot water heater fix phoenix.
5. Write down its top 10 URLs.
6. Compare the two lists. Count how many URLs appear in both.

A worked example

Here are the real top-10 URLs for two candidate phrases. Shared URLs are highlighted.

Seven of ten URLs are shared. Verdict: "water heater repair phoenix" and "hot water heater fix phoenix" are the same intent to Google. You build one page and target both phrases. You do not build a second "hot water heater fix" page. It would only cannibalize the first.

Now run the same test on a pair Google might see differently:

Only one shared URL, and even that is a different page on the same domain. Verdict: different intent, two separate pages. "Repair" buyers and "tankless installation" buyers want different things, so Google ranks different pages, so you build both.

This decides your entire page list

Run overlap analysis across your whole candidate spreadsheet. Group synonyms together. Each surviving group becomes exactly one page. When you finish, you no longer have a keyword list. You have a page list: the precise set of pages your site needs, each with its cluster of target phrases. That list feeds Chapter 8, where it becomes a URL map.

Bonus: mine the words that go ON the page

While you have those top-10 results open, read the page titles and headings of the sites that rank. The synonyms and adjacent terms they use are the vocabulary Google expects on this topic. This is called semantic coverage.

Fill out the SERP Overlap Grid worksheet at the end of this chapter before you move on. Run it on any two keywords you are unsure about and let the results decide one page or two.

A URL is the address of a page. It looks small, but it is one of the few things Google reads before it even loads your content, and it is nearly impossible to change safely later. So you decide the pattern now, once, for the whole site.

You have two real choices for a local service site.

Why flat wins for most local sites

For a business serving one metro or a handful of cities, flat is the right default. Three reasons, in plain terms.

First, clear up a common myth. Folder depth in the URL string is not itself a ranking penalty. Google does not dock you for slashes. What actually matters is link architecture and click depth: how many clicks it takes to reach a page from the homepage, and how your internal links pass authority around. Flat URLs matter because of the habits they encourage, not because of the slashes themselves.

1. They keep you honest about click depth

Click depth is how many clicks it takes to reach a page from the homepage. A flat URL paired with a homepage nav link naturally puts every money page one click from the root. Nested paths tend to come with deeper click paths, because owners build intermediate state and city landing pages and bury the service pages beneath them. It is that deeper click path, not the URL string, that slows discovery and dilutes signals.

2. Internal link equity concentrates near the homepage

When another site links to your homepage, that authority (SEO folks call it PageRank) flows through your internal links to your other pages. Pages linked directly from the homepage sit where that equity concentrates. Flat sites tend to link every money page straight from the main nav, so the equity lands on the pages that earn revenue. Deep nested structures tend to leave mid-level folder pages that exist only as pass-through directories, and their child pages can end up effectively orphaned, reachable only by a long chain of clicks.

3. Crawlers find shallow pages fastest

Crawlers discover pages by following links from pages they already know. Pages one click from the homepage get found on the first pass. Pages buried several clicks deep get found later, if at all. A flat site with a full nav plus a sitemap exposes every page immediately.

When nested actually earns its complexity

Nesting only earns its added complexity when scale forces organization on you:

• Multi-state operations with the same services in dozens of markets, where /tx/austin/ and /az/phoenix/ genuinely help humans and search engines navigate.
• Hundreds of city pages where a flat root would become an unmanageable pile of thousands of top-level URLs.

If that is not you, stay flat. For a business in one or a few cities, flat keeps your click depth honest with the least effort.

The five URL rules you never break

Build your URL map now

Take the page list from Chapter 7. Write one flat URL per page. This table is your URL map, and it goes straight to the AI in Chapter 9.

The city-template trap: doorway pages

Once you have a template, it is tempting to spin up 40 city pages by swapping the city name and nothing else. Do not. Google calls scaled, near-duplicate pages that exist only to catch searches for many locations doorway pages, and they are against the guidelines. They can get your whole site demoted.

Pick one version of your site and stick to it

A page can be reachable at several addresses that are technically different URLs but the same content. Left unmanaged, that splits your signals across duplicates. Consolidate to one canonical version on every axis:

• Canonical tags. Put a <link rel="canonical" href="..."> on every page pointing to its own preferred URL. This tells Google which version to treat as the real one when duplicates or URL parameters appear.
• www vs non-www. Choose one, for example acmeplumbing.com, and 301-redirect the other to it. Never let both resolve independently.
• HTTP to HTTPS. Redirect all http:// requests to https:// so there is a single secure version.
• Trailing slash consistency. Decide whether URLs end in a slash or not, and make the site consistent. Redirect the other form.
• Useful 404s. When a page truly does not exist, return a real 404 with a helpful page: a short apology, your nav, and a phone number, so a lost visitor still calls you.

Fill out the URL Map worksheet at the end of this chapter before you move on. Lock every slug now and you will never have to set up a redirect later.

You spent the last several chapters making real decisions: who you serve, what you sell, what you charge, and the exact URL map your site needs. That work was not busywork. It is the context that turns a generic AI into a tool that builds your business instead of a template.

Here is the principle that governs this entire chapter: a prompt is only as good as the context you load into it. Ask an AI to "build me a plumbing website" and you get the same beige site every other plumber gets. Paste in your Market Snapshot, your consumer ranking, your USP with its proof mechanism, your price matrix, and your URL map, and the same AI builds something specific, defensible, and correct from the first draft. The difference is not the model. The difference is you.

What Replit is

Replit is a website in your browser where you build and publish real software by talking to an AI agent. There is nothing to install. You type what you want, the AI builds it, you preview it live, and when you are happy you publish it with your own domain name. For a home service site, it is the fastest path from your plan to a live page that exists.

Start a project

1. Go to replit.com and create a free account.
2. Click Create and choose the AI app builder (often called Replit Agent).
3. Your very first message will be the context-loading master prompt below. Everything else builds on it.

The rule that makes every prompt work

Sophisticated prompts do four things that "make me a website" never does. Every prompt that follows uses all four:

• Role framing: you tell the AI what expert to act as, so it reasons the way that expert would.
• A context block: you paste your actual business facts from your worksheets, marked as the source of truth.
• Explicit constraints: what to do, what never to do, and the standards to hold (semantic HTML, real prices, no invented claims).
• Output format: exactly what you want back, in what shape, so you can use it without cleanup.

HTML gives you six heading levels, H1 through H6. They are not sizes. They are an outline. H1 is the title of the page, H2s are the major sections, H3s are points inside those sections, and so on. Headings are primarily a semantic and accessibility structure: they describe how your content is organized, and screen readers and AI systems use that outline to navigate and extract meaning. They are not a weighting dial you turn up to rank higher.

The rules that matter

• One H1 per page, containing your primary keyword and city. On your water heater page: "Water Heater Repair in Phoenix, AZ."
• H2s for major sections: the service overview, pricing, service area, FAQs.
• H3s for points inside an H2 section. Under a "Common Water Heater Problems" H2, each problem can be an H3.
• Do not skip levels. Do not jump from H2 to H4. Go in order.

The point almost every site misses

Real hierarchies rarely need H4, and almost never need H5 or H6. A local service page is not a legal textbook. If you find yourself reaching for H4 and beyond, that is a warning sign: your page structure is probably too deep or too tangled. Flatten it. Turn some of those buried H4s into their own H2 sections, or fold them into paragraphs. A clean page is mostly one H1, several H2s, and the occasional H3.

Make your headings do SEO work

Here is the upgrade most sites never make. Do not waste headings on generic labels. Load them with your service and your city, truthfully. The reason localized descriptive headings win is not that Google "weighs the tag" more heavily. It is that they match the language people actually search. A heading that reads like the query a customer typed lines up cleanly with that query, and it hands AI systems a self-contained, on-topic passage to lift. "Welcome" matches nothing anyone searches. "Water Heater Repair in Phoenix" matches the search exactly.

Both pages have the same sections. The right one tells Google, in its own outline, that this page is about water heater repair for plumbing customers in Phoenix and Scottsdale. The left one tells Google nothing.

A correct outline for a service page

H1  Water Heater Repair in Phoenix, AZ
H2  Signs You Need Water Heater Repair in Phoenix
   H3  No Hot Water
   H3  Leaking Tank
   H3  Rusty or Smelly Water
H2  Our Water Heater Repair Pricing
H2  How Fast We Respond in the Phoenix Area
H2  Water Heater Repair FAQs
   H3  How much does water heater repair cost in Phoenix?
   H3  Can you come out the same day?

Every chapter before this one exists to get the right person onto the right page. This chapter is about what happens in the next ten seconds. The panic buyer from Chapter 2 will not fill out a five-field form; the planner will not call before they trust you. Your pages have to make the next step obvious and frictionless for the buyer who landed on them.

Put the call button where the thumb already is

For home services, the phone is still the highest-intent action. A large share of your visitors are on a phone, often mid-emergency. Make calling a reflex, not a hunt.

• Click-to-call everywhere. Your phone number is a real tel: link so one tap dials. Put it in the header, pinned so it stays visible as they scroll, and repeated near the bottom of every page.
• One primary action per page. The main call button should be the loudest thing on the screen. Do not make "Call Now" compete with a newsletter box and three social icons for attention.
• Tap-friendly targets. Buttons need enough size and spacing that a thumb hits them cleanly, and body text needs to be readable without pinch-zooming. Cramped, tiny controls cost you bookings and get flagged in the usability audit in Chapter 16.

Forms that ask only what a booking requires

Every extra field is a reason to abandon. For the planner who prefers to type rather than call, the booking form should feel like thirty seconds of work, not a job application.

Ask for a phone number over an email when you can; for home services, a callback closes far more jobs than an email thread. And a hidden source field on the form (captured automatically) is what lets you attribute the lead later in Chapter 31.

Trust blocks belong next to the button

People decide to act at the exact moment they see the call button, so that is where the proof has to be. Do not bury your credibility on an "About" page nobody visits. Stack it right beside the thing you want them to click.

Speed is a conversion factor, not just an SEO one

A slow page loses bookings before the visitor ever reads a word. The panic buyer will tap the back button and call your competitor while your hero image is still loading. The technical work in Chapters 12 and 15 (compressed images, no junk scripts, a lean static site) is not only for ranking; it directly protects the booking. Treat every second of load time as money walking out the door.

Fill out the Conversion Audit worksheet at the end of this chapter before you move on. Walk your own site like a stranger and every unticked box is a leak worth plugging.

The mechanics from the last chapter (fast load, call buttons, short forms, trust blocks) are the ingredients. The order you serve them in is the recipe. A page that converts is not a pile of good elements; it is a deliberate journey down the screen. Use the classic awareness-to-action arc, adapted for a home-service buyer who often needs help today.

The five bands, top to bottom

Every service page moves the visitor through the same five stages, in order:

1. Discovery (the hero). Who you are, what you do, and where, with your USP bold and immediate. This is also where you echo the promise that earned the click (more on that below).
2. Education. What the service includes, your process, timelines, and real pricing. Answer the practical questions before they have to ask.
3. Relevance ("this is for me"). Localization: name the neighborhoods you serve, the local problems you solve, and the this-is-you scenarios that make a visitor feel understood.
4. Social proof. Reviews that name the service and the city, real job photos, and badges or licenses. Borrowed trust, right when doubt peaks.
5. Commitment (the CTA). A clear call button and a short form, plus a plain statement of what happens after they call. Remove the last shred of friction.

When every page follows this arc, every page is a funnel. The visitor is never left hunting for the next step; the page hands it to them. This is a proven default, not the only valid layout, but if you are unsure how to structure a page, start here and earn your way to exceptions.

Your real job photos are some of your best content (Chapter 22). But an image the search engine cannot read, or one that takes four seconds to load, works against you. Three things turn images from dead weight into an asset: names, alt text, and speed.

Name files like a human, not a camera

Your phone saves photos as IMG_4023.jpg. That tells Google nothing. Rename every image to describe what it shows, using your service and city words when they are true.

Use the same rules as URLs: lowercase, hyphens between words, no spaces, no underscores.

Alt text: describe the image truthfully

Alt text is a short written description of an image. It helps blind visitors using screen readers, and it tells Google what the image shows. Describe what is actually in the photo, and naturally include location and service only where it is true.

<!-- Good: describes the real image, location is truthful -->
<img src="water-heater-repair-phoenix-tech.jpg"
     alt="Acme technician replacing a water heater in a Phoenix home">

<!-- Decorative image that adds no info: empty alt -->
<img src="blue-divider-line.png" alt="">

Speed: the part that quietly kills rankings

A giant image is the number one reason home service sites load slowly, and slow pages rank worse and lose customers. Four fixes:

<img src="water-heater-repair-phoenix-tech.webp"
     alt="Acme technician replacing a water heater in a Phoenix home"
     width="800" height="600" loading="lazy">

Robots.txt is a plain text file at the root of your site, at yoursite.com/robots.txt, that tells search engine crawlers which parts of your site they may crawl. It is a set of polite instructions to well-behaved bots like Googlebot.

What it does and does not do

A correct file for a local business

# Allow all crawlers to access the whole public site
User-agent: *
Allow: /

# Block back-office and junk paths that should never rank
Disallow: /admin/
Disallow: /cart/
Disallow: /checkout/
Disallow: /search-results/
Disallow: /*?sort=

# Point crawlers to your sitemap
Sitemap: https://acmeplumbing.com/sitemap.xml

Line by line: allow every crawler to reach everything, then block a few paths that would waste crawl budget or embarrass you if indexed (an admin login, a shopping cart, internal search result pages, sorted duplicate URLs), then hand over the sitemap. That is all a home service site needs.

The one-character catastrophe

User-agent: *
Disallow: /

Test it before and after launch

1. Visit yoursite.com/robots.txt in your browser and read it with your own eyes. Confirm it does not say Disallow: /.
2. Run a crawler against your site to confirm it can reach every page. Screaming Frog's free tier (up to 500 URLs) is plenty for a home service site and will flag any URL blocked by robots.txt. Google's Rich Results Test and Lighthouse will also fetch a page and tell you if robots.txt is blocking it. (Note: Google retired the old standalone robots.txt tester inside Search Console, so do not go looking for it there.)
3. Recheck it the moment you go live. This is the number one thing to verify at launch.

Your pages are planned, built, written, and marked up. This chapter is the pass that catches the technical details that quietly cost rankings and customers. Work through it once and hand any misses to the AI.

Mobile readability

A word on what these numbers are. The 16px text and 48px tap-target rules are not direct ranking factors. They are Lighthouse and mobile-usability audit flags, drawn from WCAG accessibility defaults. Google will not necessarily rank you lower for missing them by a hair. The real reason to hit them is simpler: tiny text and cramped buttons lose customers first, and they fail the usability audits your competitors are passing. That is reason enough.

• 16px minimum body text on mobile. Smaller text gets flagged as "text too small to read" in mobile usability reports, and it is genuinely hard to read on a phone. Bigger is fine.
• Tap targets at least 48 by 48 CSS pixels, with 8px of space between them. A thumb needs room. Real buttons need real height and line-height, not a 20px text link crammed next to another. Cramped taps get flagged and cost you calls. Make your call and form buttons big.
• Viewport meta tag on every page. Without <meta name="viewport" content="width=device-width, initial-scale=1"> your site renders like a shrunk desktop page on phones. This one line is non-negotiable.

Core Web Vitals in plain English

Google measures three real-world speed and stability numbers, called Core Web Vitals. Here they are without the jargon.

The good news: three fixes handle nearly all of it, and you have already met most of them.

• Compress images (Chapter 13). Heavy images are the top cause of slow LCP.
• Set width and height on every image (Chapter 13). This kills layout shift and fixes CLS.
• Do not load junk scripts. Every tracker, chat widget, and pop-up plugin slows the page and hurts INP. Keep the static site you asked for in Chapter 9 lean.

The rest of the checklist

• HTTPS. Your site must load over https:// with a padlock. Replit handles this on publish. Confirm the padlock is there.
• Mobile-first indexing. Google ranks the phone version of your site, not the desktop version. The phone view is the site. Test every page on an actual phone.
• NAP consistency. Your Name, Address, and Phone must match exactly across your website, Google Business Profile, Facebook, and Yelp. Even "St" vs "Street" differences can confuse Google.
• Google Business Profile connected. Claim and complete your free Google Business Profile and link it to your site. This gets its own chapter next, because for local search it is as important as the website itself.

Titles, descriptions, and social preview

• Unique title tag per page, front-loaded. Google displays roughly the first 580 pixels of a title, about 55 to 60 characters, before it cuts it off. Put your city and service first so they never get truncated. Good: "Water Heater Repair Phoenix, AZ | Acme Plumbing." Bad: "Acme Plumbing, Your Trusted Local Family-Owned Water Heater..."
• Meta description per page. The description is not a ranking factor. Its job is to earn the click (CTR). Write a compelling one to two sentences with the service, city, and a reason to call. Know that Google rewrites descriptions often, pulling its own snippet from your page, so do not obsess over it.
• Open Graph tags and a favicon. Add og:title, og:description, and og:image so your links look right when shared on Facebook or texted. Add a favicon (the little icon in the browser tab) so the site looks finished, not broken.
• font-display: swap. If you use a web font, add font-display: swap so text shows immediately in a fallback font instead of staying invisible while the font loads. This protects your LCP.

Internal linking done right

• Link between related pages. Link city pages to related service hubs and back. Your drain cleaning page should link to your emergency plumber page. This spreads ranking power and helps visitors.
• Descriptive anchor text. The clickable words in a link should describe the destination. Link the words "emergency plumber in Phoenix," not "click here." Anchor text tells Google and visitors what is on the other end.
• Breadcrumbs. Add a simple breadcrumb trail (Home › Services › Water Heater Repair) with BreadcrumbList schema. It helps visitors and can show in search results.
• Orphan page check. An orphan page is one that no other page links to, so visitors and crawlers cannot reach it through your site. Crawl your site (Screaming Frog free tier) and confirm every page is linked from at least one other page. Fix any orphans by adding a link from the nav or a related page.

When someone searches "plumber near me," Google usually shows a map with three business listings above the regular results. That is the map pack, and it is powered by Google Business Profile, the free listing you claim for your business. For a home service company, showing up here is often worth more than any single web page. Two things drive it: a complete, accurate profile, and a steady flow of real reviews.

Claim and verify

Go to business.google.com, search for your business, and claim it (or create it if it does not exist). Google then verifies you actually run the business, usually by postcard, phone, email, or video. You cannot rank in the map pack until you are verified, so do this first.

Set the primary category correctly, once

Your primary category is the single biggest lever in your profile. Choose the most specific category that fits, for example "Plumber" or "HVAC contractor," not a vague catch-all. Add secondary categories for other services you offer, but get the primary one exactly right. It shapes which searches you appear for more than anything else in the profile. Chapter 18 goes deep on how to choose it like a headline keyword; for now, just pick the closest, most specific match.

Your GBP is your canonical NAP

Before you fill in anything else, decide your business Name, Address, and Phone and write it down exactly. This is your canonical NAP, the master record. Every citation, directory listing, and schema block in this book must match it character for character, down to "Ste" versus "Suite." The profile you set up here is the source of truth. Citations (Chapter 25) and your schema markup (Chapter 21) both copy from it.

Complete the essentials

• Storefront vs service-area. If customers come to you, show your address. If you go to them (most home service businesses), set it up as a service-area business: hide the street address and list the cities you cover. Do not fake a storefront you do not have. Google can suspend you for it. Chapter 20 covers why a real address still helps even for service-area businesses.
• Services and products. Add your services with your real service names. Chapter 18 shows how to fill these out completely, in natural language, using terms adjacent to your website copy.
• Photos to start. Upload real photos of your trucks, crew, and completed jobs. Profiles with genuine photos get more calls and clicks. Chapter 18 turns this into a standing field SOP that also helps you rank.
• Q&A: seed it yourself. The profile has a public questions section anyone can post to. Get ahead of it: post your own most common questions and answer them from your business account. This controls the narrative and answers real buyers.
• Website link with UTM tags. Point the website link to your homepage with a tracking tag like ?utm_source=google&utm_medium=gbp so you can see in analytics (Chapter 30) how much traffic your profile sends.

Reviews: the engine of local trust

Review count, rating, and recency all feed both your ranking and your click rate. The businesses in the map pack almost always have more and fresher reviews than the ones below. Reviews are not a nice-to-have. They are the game, and Chapter 18 shows how the words inside your reviews quietly teach Google what you do.

Respond to every review

Reply to every review, good and bad. Thank the happy ones by name and mention the service and city ("Thanks, Maria, glad we got your water heater sorted in Scottsdale"). Answer the unhappy ones calmly, take it offline, and show future customers you handle problems well. Responding signals an active, real business, and it is a small SEO and trust win on every reply. Chapter 18 explains why naming the service and city in your replies is not just courtesy, it is entity-building.

Most owners treat Google Business Profile like a form: fill it out, verify it, forget it. Then they wonder why a competitor with a worse website sits above them in the map pack. The difference is almost never the setup. It is the growth cadence, the steady weekly rhythm of real activity that tells Google you are a living, working business in this area right now. This chapter is that rhythm, plus the two levers most owners get wrong: categories and the words inside your reviews.

The growth cadence is an operating rhythm

A profile does not grow because you optimized it once. It grows because, week after week, real signals of real work keep arriving: new photos from real job sites, fresh posts about real jobs, new reviews using real service language. Google rewards sustained, current activity and quietly discounts profiles that went silent. Build the rhythm into how the business runs, not into a one-time launch checklist.

The tech photo SOP

The single most underused growth lever is your technicians' phones. Make on-site photo uploads a standing part of closing out every job, the same as collecting payment or cleaning up.

• Real jobs, on location. Photos taken at the customer's home or site, not staged at the shop and definitely not pulled from stock.
• Phone GPS on. Location services enabled so the photo carries the metadata of where the work actually happened.
• Two or three per job, before they leave. Before, after, and the finished work. Upload from the job site as the last step, not batched a week later at the office.

Written into the job-close routine, this produces a constant stream of geo-relevant, authentic photos with zero extra staff. It is the highest-leverage habit in this entire cluster, and Chapter 19 shows how it may help you rank in specific zones.

Posting cadence: at least twice a week, and do not overdo it

Post to your profile at least twice a week. Consistency beats volume. Do not post daily: a flood of near-identical posts looks automated, dilutes engagement, and trains Google and customers to ignore you. Two to three genuine posts a week is the sweet spot.

A post outline you can reuse

1. What job and what outcome (with a number if you have one)
2. Where: neighborhood and city
3. One keyword worked in naturally, never stuffed
4. A real on-site photo
5. A clear next step (call, book, get a quote)

Open Google on your phone at the shop, search "plumber near me," and there you are at the top. Reassuring, and almost meaningless. You are standing on top of your own pin, so proximity hands you the result. The real question is what a homeowner sees from three miles away, or five, or across the county line. That is your Map Pack radius, your true circle of local visibility, and you cannot judge it from where you sit.

What a geo-grid scan is

A grid scan (also called a geo-grid or map scan) uses a tool to check your Map Pack ranking from a grid of simulated points spread across your city, not just from your location. Instead of one ranking, you get a lattice of them: your position at each point, laid out on a map. Tools like Local Falcon, BrightLocal, or Places Scout run these for a small monthly fee.

How to read the grid

The output is a grid of colored dots over a map of your service area.

• Green in the center means you rank in the top few at points near your location. Expected, and not where the money hides.
• Fading to yellow, then red toward the edges shows exactly where your visibility dies. The line where green turns to red is your effective radius.
• A red zone over a profitable neighborhood is a target. Someone else owns the pack there, and grid scans show you precisely where to fight or where to plant a new location (Chapter 20).

Why the radius expands and contracts

Proximity is the anchor. All else equal, Google shows the closest qualified business, so your radius is naturally larger where few competitors sit near a searcher and smaller where many do. In dense, competitive areas, Google effectively caps how far your pack visibility reaches, because a closer competitor wins each of those far-off grid points no matter how good you are. You cannot out-optimize physics; a shop ten miles away rarely beats the equally good shop around the corner.

What you can do is push your radius outward as far as competition allows, and hold it there. These are the levers that grow and defend the radius:

Scans are your measurement loop

Do not run a grid scan once and forget it. It is the recurring instrument that tells you whether everything else is working.

• Monthly, to watch the radius grow or shrink and catch problems early.
• After any GBP change, a new category, a rush of reviews, a fresh photo habit, to see what actually moved the map.
• To pick the next expansion target, feeding the war game in Chapter 20 with hard data on where you are invisible but the work is worth having.

Your grid scan from Chapter 19 shows a wall of red over a neighborhood you would love to work in. Your existing profile cannot reach it, because a competitor sitting closer wins every grid point out there. There is a legitimate way to plant your flag in that red zone: a second, real Google Business Profile. This chapter is the strategy, and the guardrails, for doing it without torching your whole account.

The requirements come first, because Google is ruthless

Before any strategy, understand the hard rule: every Google Business Profile must represent a real, staffed, verifiable business presence. Google actively hunts for fake locations and suspends them, and a suspension can take your good profiles down with the bad one.

Storefront versus service-area, and which you qualify for

There are two legitimate profile types, and the rules differ.

A second profile has to be a genuinely separate operation, not the same business wearing a mask. That means a real place with staff or trucks based there, a distinct phone number, and real work happening from it. Google's own guidance is strict about this, so treat the requirement as absolute, not a hurdle to game.

The war game: where each profile plants a circle

Here is the strategy the requirements make possible. Each profile has an effective radius (Chapter 19). Your first profile owns a circle of visibility around its location. A second legitimate location plants a new circle, centered where your old one faded to red, precisely where competitors currently own the pack. Two real locations, two overlapping circles, and suddenly you appear in packs you could never reach from one pin.

Target with data, not a dartboard

Do not open an office where it feels nice. Let your own numbers pick the spot. This is where Parts I and IV pay off.

1. Open your CRM and run revenue by zip code. You are looking for zips where you already win real, profitable work despite thin map visibility.
2. Cross-reference those zips with your Part I demographic data: median income, owner-occupancy, and competitor density. You want high income, high owner-occupancy, and a gap in strong competitors.
3. Overlay your grid scan (Chapter 19). The sweet spot is a zip that is red on the map but green in your bank account: underserved in visibility, already profitable, and demographically strong.

Satellite office economics

A second location costs real money: rent or a yard, a dedicated phone line, someone staffing it, and signage. Justify it the way you justify any spend in this book, with the lifetime-value math from Chapter 5. If a new circle of map visibility over a proven-profitable zip is expected to book enough jobs to clear that overhead with margin to spare, it is an investment. If you are reaching to fill the office rather than serving demand you already have, it is a liability. Let the CRM revenue you are already earning there make the case before you sign a lease.

How not to get suspended, and how to stage it

• Real presence only. Staffed location or trucks genuinely based there, a distinct local phone number, and consistent NAP for that specific location (Chapter 25).
• One profile at a time. Stand up a new location, verify it, build its reviews and photos, and confirm it is stable on grid scans before you even consider the next. A slow, real rollout does not trip Google's fraud signals.
• Treat each profile as its own business for reviews, posts, and field photos. A location with no reviews and no activity looks fake even when it is real.
• Never spin up multiple profiles at once at questionable addresses to blanket a metro. That is the fastest way to lose everything.

Schema markup is structured data you add to a page, written in a format called JSON-LD, that spells out facts about your business in a way search engines read perfectly: your name, address, phone, hours, service area, and services. It does not change how the page looks to visitors. It sits in the code and removes all guesswork for Google.

Start with LocalBusiness

Every page of your site should carry LocalBusiness schema, ideally the specific subtype that fits you: Plumber, HVACBusiness, and so on. It carries your core facts, known in the trade as NAP: Name, Address, Phone.

areaServed: be honest about your radius

On a city page, tell schema the real area you serve from that city, using a GeoCircle with a geoRadius. If your Phoenix page realistically covers a 25-mile radius, say 25 miles. Do not claim the whole state on a city page. An honest, tight radius per city page reads as genuine local coverage, which is exactly what you want.

sameAs: consolidate who you are

Google builds a profile of your business as an "entity." You help it connect the dots with sameAs, a list of links to your other verified profiles: your Google Business Profile, Facebook, Yelp, and any state licensing page. This tells Google "all of these are the same business," which strengthens your identity and trust.

Service, FAQPage, and Review, carefully

• Service schema describes the specific service the page is about. Fine to add on each service page.
• FAQPage schema is only allowed if the FAQs are visibly on the page (the ones you built in Chapter 23). Same questions, same answers, word for word.
• Review schema: only mark up real reviews shown on the page, and never mark up reviews you wrote about yourself. Google restricts self-serving review markup and can penalize fakes.

The homepage block: LocalBusiness

This goes in the head of your homepage. It is the anchor for your whole site: every service page will point back to its @id. Replace every value with your own. Each line is labeled with why it is there. Chapter 9's Prompt 7 generates this exact block from your fill-in blanks.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Plumber",
  "@id": "https://acmeplumbing.com/#business",
  "name": "Acme Plumbing",
  "url": "https://acmeplumbing.com/",
  "telephone": "+1-602-555-0123",
  "priceRange": "$$",
  "image": [
    "https://acmeplumbing.com/img/acme-plumbing-truck-phoenix.jpg",
    "https://acmeplumbing.com/img/acme-plumbing-team.jpg"
  ],
  "address": { "@type": "PostalAddress", "streetAddress": "123 N Central Ave", "addressLocality": "Phoenix", "addressRegion": "AZ", "postalCode": "85004", "addressCountry": "US" },
  "geo": { "@type": "GeoCoordinates", "latitude": 33.4484, "longitude": -112.074 },
  "areaServed": {
    "@type": "GeoCircle",
    "geoMidpoint": { "@type": "GeoCoordinates", "latitude": 33.4484, "longitude": -112.074 },
    "geoRadius": "40000"
  },
  "openingHoursSpecification": [
    { "@type": "OpeningHoursSpecification", "dayOfWeek": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday"], "opens": "07:00", "closes": "18:00" },
    { "@type": "OpeningHoursSpecification", "dayOfWeek": ["Saturday"], "opens": "08:00", "closes": "16:00" }
  ],
  "sameAs": [
    "https://www.yelp.com/biz/acme-plumbing-phoenix",
    "https://www.facebook.com/acmeplumbingphx",
    "https://www.angi.com/companylist/us/az/phoenix/acme-plumbing.htm",
    "https://www.bbb.org/us/az/phoenix/profile/plumber/acme-plumbing",
    "https://azroc.gov/licensee/acme-plumbing"
  ],
  "aggregateRating": { "@type": "AggregateRating", "ratingValue": "4.8", "reviewCount": "237" },
  "hasOfferCatalog": {
    "@type": "OfferCatalog",
    "name": "Plumbing Services",
    "itemListElement": [
      {
        "@type": "Offer",
        "itemOffered": { "@type": "Service", "name": "Water Heater Repair" }
      },
      {
        "@type": "Offer",
        "itemOffered": { "@type": "Service", "name": "Drain Cleaning" }
      },
      {
        "@type": "Offer",
        "itemOffered": { "@type": "Service", "name": "Emergency Plumbing" }
      }
    ]
  }
}
</script>

• @type uses your real trade sub-type (Plumber, HVACBusiness, Electrician), not the generic LocalBusiness, so Google knows exactly what you do.
• @id is a stable identifier for your business that other blocks link to. Keep it identical everywhere.
• address, geo are your real NAP and map coordinates. image points to real photos, named descriptively, not stock art.
• areaServed as a GeoCircle declares your honest radius. The geoRadius of 40000 is 40,000 meters, about 25 miles. Set it to your true reach, not a number that swallows the state. If you would rather list cities than a radius, replace it with the city-list form shown below.
• openingHoursSpecification lists real hours. For genuine 24/7 emergency service, add a block with "opens": "00:00" and "closes": "23:59" across all seven days. Only claim it if you truly answer around the clock.
• sameAs links your other real profiles so Google connects them to one entity: Yelp, Facebook, Angi, the BBB, and your state license page (from your Chapter 25 citations). Note: your Google Business Profile is not a sameAs target, Google already owns that link.
• aggregateRating is optional and dangerous. Include it ONLY if those exact review numbers are real AND visibly displayed on that page. Google restricts self-serving review markup; a rating in your schema that a visitor cannot see on the page can trigger a manual penalty.
• hasOfferCatalog lists the services you offer, tying your service pages to the business.

Each service page: Service plus BreadcrumbList

Every service page gets two more blocks. The first is a Service block that describes just that service and links back to the business by @id, with a real price range from your Chapter 4 matrix.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Service",
  "@id": "https://acmeplumbing.com/water-heater-repair-phoenix/#service",
  "serviceType": "Water Heater Repair",
  "provider": { "@id": "https://acmeplumbing.com/#business" },
  "areaServed": { "@type": "City", "name": "Phoenix, AZ" },
  "offers": {
    "@type": "Offer",
    "priceCurrency": "USD",
    "priceSpecification": { "@type": "PriceSpecification", "minPrice": "180", "maxPrice": "450", "priceCurrency": "USD" }
  }
}
</script>

The provider uses the same @id as your homepage business, so Google knows this service belongs to that entity. The offers price range comes straight from your price matrix. The second block is the breadcrumb trail, which pairs with the visible breadcrumbs from your Chapter 16 checklist.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "BreadcrumbList",
  "itemListElement": [
    { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://acmeplumbing.com/" },
    { "@type": "ListItem", "position": 2, "name": "Services", "item": "https://acmeplumbing.com/services/" },
    { "@type": "ListItem", "position": 3, "name": "Water Heater Repair in Phoenix", "item": "https://acmeplumbing.com/water-heater-repair-phoenix/" }
  ]
}
</script>

FAQPage, only where the questions are visible

On any page that genuinely shows a set of questions and answers (the ones from Chapter 23), you may add FAQPage schema, word for word with what is on the page.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [{
    "@type": "Question",
    "name": "How much does water heater repair cost in Phoenix?",
    "acceptedAnswer": { "@type": "Answer", "text": "Most repairs run 180 to 450 dollars depending on the part." }
  }]
}
</script>

Validate everything, and keep expectations honest

Always validate your schema before you trust it. Use two free tools:

• validator.schema.org checks that your JSON-LD is syntactically valid and well-formed.
• Google's Rich Results Test at search.google.com/test/rich-results shows which rich-result types Google can detect on the page.

Most home-service blogs are a graveyard of "5 Tips for Spring" posts that rank for nothing and sell nothing. The problem is not blogging; it is blogging without a job. Give every post one job: support a money page. That is the whole strategy.

The tier model

Think of your content in two tiers that work together:

• Tier 1: money pages. Your service-area pages, the ones built to book jobs: "AC Repair in Naperville, IL." These are the pages you want to rank and convert.
• Tier 2: supporting blog posts. Assets that capture question searches and pass internal links up to the tier-1 page with descriptive anchor text. They feed the money page authority and relevance.

A citation is any place online that lists your business name, address, and phone number, together known as your NAP. Google reads these mentions across the web as votes of confidence that you are a real, consistent, local business. When they all match, they reinforce your Google Business Profile from Chapter 17. When they conflict (an old address here, a disconnected number there), they raise doubt at exactly the moment Google is deciding whether to trust you in the map pack.

Consistency is the whole game

The single rule of citations: your NAP must be byte-for-byte identical everywhere. Pick one canonical format and never deviate.

The directories that actually matter

You do not need to be on hundreds of directories. A focused set of high-quality, relevant listings beats a spray of junk. Prioritize in this order.

Data aggregators feed everyone else

Behind the scenes, a handful of data aggregators supply business information to countless smaller directories, apps, and voice assistants. If your NAP is right at the aggregator level, correct data flows downstream automatically; if it is wrong there, the error propagates everywhere. The major players feeding the US ecosystem include Data Axle, Foursquare, and Neustar/Localeze. Getting your listing correct with these sources cleans up dozens of sites you would otherwise never find by hand.

Audit what already exists

If your business has been around at all, listings you never created already exist, often with stale data from old phone books and scraped sources. Find and fix them.

• Search your business name, your phone number, and your old phone numbers in Google. Every result that lists you is a citation to check.
• Claim the listings you can, correct the NAP to match your canonical block, and request removal of true duplicates (two listings for the same location confuse Google).
• Free scanners from services like Moz Local or BrightLocal will spot inconsistencies quickly. You can fix them by hand for free, or pay a service to push corrections; either works.

A backlink is another website linking to yours. For years, people gamed this by buying links in bulk, and for years it worked. It does not anymore. Google contextualizes links now: it weighs who is linking, whether the link makes sense, and whether real people actually click it. For a local home-service business, chasing volume is a waste. Chasing local relevance is the whole game.

The local links that actually count

Each of these earns a real link, and most also send real people. For each: how to get it, roughly what it costs, and what it is worth.

Chamber of commerce

How: join your local chamber; members get a profile and a link from a trusted local domain. Cost: often a few hundred dollars a year. Worth: a genuinely local, trusted link plus real referrals and networking. One of the highest-value, lowest-effort links available.

Better Business Bureau

How: create and, if you choose, accredit a BBB profile. Cost: free to list; accreditation costs a yearly fee. Worth: a link plus a trust signal customers actively check, and a second review stream that supports your reputation.

Suppliers and vendor pages

How: ask the manufacturers and suppliers you already buy from to add you to their "find a certified installer" or "where to buy" pages. Cost: free; you already spend money with them. Worth: a highly relevant industry link, and buyers who start on the brand's site get routed to you.

Sponsorships

How: sponsor a youth sports team, a school event, or a local government or community-facility program. Most publish a sponsors page that links to you. Cost: a few hundred to a couple thousand dollars depending on the event. Worth: links from schools, leagues, and community sites (often strong local domains), plus real goodwill and your name in front of local families who own homes.

A small annual scholarship

How: set up a modest scholarship for local students and let area schools and colleges know. Cost: a few hundred dollars for the award. Worth: links from .edu and local-news domains that are hard to earn any other way, and it is genuinely good for your community. Keep it real, actually award it, and do not turn it into a link scheme.

Churches and community organizations

How: support or partner with local churches, nonprofits, and neighborhood associations. Cost: varies, often small. Worth: trusted community links and word-of-mouth inside tight local networks that convert well.

Local news coverage: be the source

How: pitch local reporters seasonal, useful angles: "how to stop pipes freezing before the first cold snap," "what the heat wave does to your AC." Reporters need local experts on deadline; be the easy, quotable one. Cost: free, just your time. Worth: links from high-authority local-news domains and a credibility halo you cannot buy.

Host or sponsor community events

How: run a free water-heater-flush day, a charity drive, or a neighborhood cleanup, and get it listed on community and event calendars. Cost: your time and some supplies. Worth: event-page links, local press, and a story customers remember.

Fill out the Local Backlink Target Sheet worksheet at the end of this chapter before you move on. Chase relationships that could send you a customer, not links for their own sake.

You just spent an entire part of this book earning organic visibility: fast pages, a strong profile, schema, information gain, citations, links. That work is real and it compounds. But you have to understand what the search results page actually looks like today, because it is not the ten blue links it was ten years ago.

Organic SEO is one lever, and it is a lever that quietly loses ground every year. Google keeps adding paid and machine-generated blocks above the organic results. Each new block pushes your hard-won organic ranking further down the screen. You can be the number-one organic result in your city and still sit below the fold on a phone, where most local searches happen.

What a money search actually shows

Search something with clear commercial intent on a phone, for example "emergency plumber phoenix," and this is the stack you scroll through, top to bottom. The percentages are estimated click share for a high-intent mobile local search. Read down the right column and watch where the clicks go: by the time you reach the organic block, add up everything above the fold and organic is fighting over single digits.

There are two distinct paid products from Google, and owners constantly confuse them. Local Services Ads (LSA) sit at the absolute top and charge you per lead. Google Ads (the text ads) sit just below and charge you per click. Start with LSA, because pay-per-lead is friendlier to a small budget, then layer Google Ads for the searches LSA does not cover.

Local Services Ads: pay per lead, not per click

LSA is the block of pros at the very top with a green badge, a star rating, and a "Request a quote" or call button. You do not pay when someone sees or clicks your ad. You pay only when you get an actual lead: a call or message from a real prospect. That alone makes it the safest paid surface for a home-service business to start with.

Those one-tap call and quote buttons matter more than they used to. As covered in Chapter 27, Google has been stripping the call and website buttons off organic Business Profile listings and keeping them on paid placements, so LSA and sponsored map pins now hold a real friction advantage on emergency searches: the panic buyer taps once instead of digging for your number.

The Google Screened and Google Guaranteed badges

To run LSA you pass a verification: license and insurance checks and background checks. In return you earn a badge, Google Guaranteed for most trades or Google Screened for some professional services. The badge is a trust mark customers recognize, and Google Guaranteed even backs the job up to a cap if the customer is unhappy. Getting verified is step one and it takes real paperwork; start early.

Before you go live, connect the free tools that get your site indexed and let you watch it perform. This is a one-time setup that pays off forever.

Google Search Console

Search Console is Google's free dashboard for how your site performs in search. It is not optional. Set it up:

1. Go to search.google.com/search-console and add your site as a property.
2. Verify ownership. The easiest method is usually a DNS record or an HTML tag the AI can add to your site.
3. Submit your sitemap (Chapter 14) under Sitemaps.
4. Use the URL Inspection tool to check any single page: it tells you whether Google has indexed it and why or why not.

Bing Webmaster Tools

Do not skip Bing. It powers Bing search and feeds some AI assistants, and setup takes five minutes because you can import your site directly from Search Console. Go to bing.com/webmasters, choose import from Google Search Console, and your verification and sitemap carry over.

Google Analytics 4 (GA4)

GA4 is Google's free analytics that shows how many people visit, where they come from, and what they do. Create a property at analytics.google.com, and have the AI add the GA4 tag to every page. You do not need to master it. At minimum, watch: how many visitors you get, which pages they land on, and which sources send them (Google search, your GBP link, direct).

Track the actions that make you money

Traffic is vanity. Calls and form submissions are revenue. Measure those.

• Click-to-call links. Every phone number on the site should be a real <a href="tel:+16025550123"> link so a tap on mobile dials instantly. This is the single most-used action on a home service site.
• Call tracking (optional). To measure which calls came from the website versus GBP versus ads, use a call-tracking service that assigns tracking numbers. Keep your displayed NAP consistent (Chapter 16) if you do this, and follow the provider's guidance to avoid confusing Google.
• Form success and error states as conversions. Your lead form should show a clear success message on submit and a clear error message on failure. Fire a GA4 conversion event on the success state so you can count real leads, not just page views. A form that silently fails loses leads and you never know.

Most owners know their total revenue and their total marketing spend, and nothing in between. That is like knowing the score of a game but not who scored. Attribution is the discipline of tying revenue back to the specific channel that produced it: this job came from Google organic, that one from the Business Profile, this one from a referral. Without it, you cut the wrong things and feed the wrong things.

Give every channel its own fingerprint

The trick is to make each channel identifiable before the lead ever reaches you.

Unique phone numbers per channel

A call-tracking service assigns a different phone number to each channel: one number on your Google Business Profile, another on your website, another on a yard sign or mailer. Every number forwards to your real line, but the service records which one rang. Now a phone call, the highest-intent action in home services, is finally attributable. Keep your displayed NAP consistent per the rules in Chapter 25 and follow the provider's guidance so the tracking numbers do not confuse Google.

UTM discipline on every link

For clicks, tag every link you control with UTM parameters so analytics (Chapter 30) records the source. Your Business Profile link, your email footer, your social posts, each carries a tag like ?utm_source=gbp&utm_medium=profile. Be consistent with your naming; "facebook," "FB," and "fb" will split one channel into three and ruin your report.

A source field that feeds your CRM

Your booking form (Chapter 11) captures the source automatically in a hidden field and passes it into your customer database or CRM alongside the job. Now every online lead lands in your CRM already labeled with where it came from.

Run a monthly revenue report by lead source

Once a month, pull a single report from your CRM: revenue and job count grouped by lead source. Not leads, not clicks, actual booked revenue. This one report tells you where your money truly comes from and, next to your spend per channel, gives you return on ad spend for each.

Every paid surface from Chapters 26 and 27 shows up here as its own row, so you can see the true return on each ad dollar side by side. Notice the pattern this example makes obvious: organic compounds to the best ROAS, LSA earns its keep, and the shared-lead marketplaces run thin. Your real numbers will differ; the point is to measure each paid channel, not to assume.

Are your staff attributing leads correctly?

Attribution breaks at the front desk more often than in the software. The person answering the phone is busy and skips the source field, or guesses. This is the silent hole in most owners' data.

• Call recording. Most call-tracking services record calls. Spot-check a handful each week to confirm which number rang and whether the source was logged correctly.
• Spot audits. Once a month, pick ten jobs and trace each one from first contact to the CRM record. Every mismatch is a hole in your data you can now close.
• Make it easy. Because each channel has its own tracking number, the source is often logged automatically, taking the guesswork away from whoever answers.

When a channel's return drops, diagnose before you cut

A falling ROAS is a symptom, not a diagnosis. Cutting the channel on reflex can kill your best source over a temporary dip. Work through the causes first.

Fill out the Monthly Revenue by Source worksheet at the end of this chapter on the same day every month. The source with the best return is where the next dollar goes.

Here is the trap almost every owner falls into: they obsess over getting more leads while quietly losing half the ones they already have. If ten people call and you book three, doubling your leads doubles your waste alongside your revenue. The single fastest way to grow is often not more traffic; it is a higher booking rate on the calls you already get.

Know your booking rate, then compare it

Booking rate is simply jobs booked divided by qualified leads. Pull it straight from the attribution data in Chapter 31. A healthy home-service booking rate on inbound calls is often well above half; if yours is markedly lower than that, you have found your biggest opportunity, and it costs nothing in ad spend to fix.

Diagnose the leak in order

1. Phone handling

Most lost bookings die on the phone, and most of that is mechanical, not mysterious.

• Answer speed. A panic buyer (Chapter 2) who reaches voicemail is already dialing the next result. Every unanswered ring is a paid-for lead handed to a competitor.
• After-hours coverage. Emergencies do not keep business hours. An answering service or a forwarded line that books the job at 9 p.m. captures revenue your competitors sleep through.
• A real script. A simple, consistent call flow (greet, diagnose the need, give a confident price or arrival window, book the slot, confirm) converts far better than winging it. Use your call recordings from Chapter 31 to hear where calls fall apart.

2. Pricing objections

If callers consistently balk at the price, the problem may be your pricing (Chapter 4), or it may be that whoever answers is quoting a number without first establishing value. A price stated cold sounds expensive; the same price after a confident explanation of the guarantee and what is included sounds fair. Sometimes the fix is the script, not the number.

3. The review gap

A caller who found you organically will often check your reviews before they commit, even mid-call. If a competitor has 400 reviews and you have 30, you lose the wobble even when your price and service are better. That sends you back to the reviews engine in Chapter 17. A thin review count quietly suppresses your booking rate no matter how good your phone handling is.

Measure what happened on the job

Booking the job is not the end of measurement. A short job-completion survey tells you whether the work itself is building or eroding your reputation, and it catches problems before they become one-star reviews.

Acquiring a new customer is the most expensive thing you do. You paid for the ranking, the ad, the lead, and the phone call. Selling the same customer again costs almost nothing, because trust is already built and their number is already in your database. Yet most home-service owners close the job, mark it done, and never contact that person again. That is money left on the floor, every single day.

Home services are on a natural cycle; use it

Your work has a rhythm, and that rhythm is a built-in reason to reach back out. Carpet gets dirty again on a predictable schedule. Water heaters and HVAC systems need maintenance. Drains clog again. You do not have to invent a reason to follow up; the calendar hands you one.

Run these by text and email from your CRM. Text gets read; email carries more detail. A simple automated sequence tied to each job's completion date does the remembering for you.

Small human touches punch far above their cost

The stuff that feels old-fashioned is exactly what a faceless franchise cannot do, and it is pure fuel for the owner-operated positioning from Chapter 3.

• Handwritten thank-you cards after a first job. A ninety-cent card is remembered long after a receipt is thrown away.
• Customer birthday or move-in-anniversary notes, if you capture the date. Low effort, high memorability.
• A real follow-up call a few days after a big job, just to confirm everything is still working. It heads off problems and deepens loyalty.

Time the referral and review ask to the peak moment

The best moment to ask for a referral or a review is the instant a customer is happiest, which is usually right when they tell you they are thrilled. Your job-completion survey from Chapter 32 pinpoints that moment: a 9 or 10 is a green light. Strike then.

Everything up to here gets you a working system: a business defined, a machine built, discoverability earned, and performance measured. What separates the businesses that plateau from the ones that keep growing is what they do next. They treat the whole book as an annual loop, revisiting each part with a year of real data in hand.

The annual planning loop

Once a year, block a day and run the loop. It is short because the work is already done; you are refreshing assumptions with evidence, not starting over.

• Revisit Part I. Did median pricing move in your zips? Pull fresh Census data and re-shop competitors. Did a new competitor appear, or did an old one weaken? Update your pricing matrix (Chapter 4) and your positioning (Chapter 3) if the landscape shifted.
• Review the machine. Which pages rank and convert, which do not? Refresh thin content, add pages for demand you have discovered, and re-check your conversion path (Chapter 11).
• Read the performance data. Take your twelve monthly revenue-by-source reports (Chapter 31) and look at the year. What is the real cost per lead and lifetime value now? Adjust the budget model from Chapter 5 to match reality.

Add one lead source per year, within budget

Do not try to be everywhere at once. Each year, add one new lead source and give it enough time and budget to prove itself before you judge it. Master your website and Business Profile first. Then, in later years, layer in the next channel: paid search, a referral program, direct mail to your best zips, local sponsorships. One at a time means you can actually attribute and evaluate each one instead of muddying your data with five simultaneous experiments.

Kill underperformers with data, not vibes

The flip side of adding a channel is cutting one. But cut with evidence. A channel earns removal only after you have run the diagnostic from Chapter 30 (rankings, profile health, budget, seasonality) and it still underperforms across a meaningful stretch, not one soft month. "It feels like the ads are not working" is a vibe. "This channel returned 1.4x over two quarters after ruling out seasonality" is a decision.

The compounding effect of boring consistency

None of this is flashy. Refresh a page here, add a channel there, ask for every review, follow up on every job, run the report every month. It is boring. That is precisely why it works: most of your competitors will not do it consistently. Reviews compound. Rankings compound. A retained customer base compounds. Five years of unglamorous, consistent execution builds a moat that a competitor cannot buy their way past in a season.

Fill out the Annual Iteration Review worksheet at the end of this chapter once a year. It is where you get honest with the numbers and set the one big bet for next year.

Here is the whole field manual on one runway. Do these in order; each step produces what the next one needs. One rule to remember: through the middle of this sequence your website is in a staging state (Chapter 9). You do not truly go live and tell Google about it until everything validates. Decide, build, polish, then flip the switch.

Part I: Build the Business

• 1. Know your market. Build the zip-code sheet from Census data: income, owner-occupied rate, housing age, seasonality. Map competitors. Decide compete or flank. (Chapter 1)
• 2. Know your consumer. Name your buyers (panic, planner, price) and rank who you want most. (Chapter 2)
• 3. Set positioning. Write one provable positioning statement and your de-positioning angle. (Chapter 3)
• 4. Price it. Build the price matrix (cost-plus or market-entry) and plan your raises by review milestone. (Chapter 4)
• 5. Package the offer and budget. Turn positioning and price into one offer. Pick a budget model and number on purpose. (Chapter 5)

Part II: Build the Machine

• 6. Research keywords. Seed from your zips and buyers. Build the candidate sheet. Weigh intent over volume. (Chapter 6)
• 7. Run SERP overlap. Collapse synonyms into one page each. Output: your definitive page list. (Chapter 7)
• 8. Map your URLs. One flat, lowercase, hyphenated URL per page. Plan canonicals and one site version. Output: your URL map. (Chapter 8)
• 9. Build with AI on Replit, to staging. Feed it the URL map. Demand semantic, static, fast pages. Publish to a preview state; do not announce it yet. (Chapter 9)
• 10. Fix headings. One clear H1 per page, working H2s, structure not font size. (Chapter 10)
• 11. Wire it to convert. Click-to-call everywhere, short forms, trust blocks by every CTA, fast load. (Chapter 11)
• 12. Lay each page out as a funnel, then audit the above-fold echo. Structure every money page as the five-band arc (discovery, education, relevance, proof, commitment). Confirm each hero echoes the exact price, promise, or USP its meta title dangles. (Chapter 12)