Split the account at the county line, because a Franklin click is not a Donelson click
Metro Nashville and Davidson County run as one consolidated government while the surrounding counties do not, and the buyer changes at that line as reliably as the paperwork does.
A ring drawn around your shop is the wrong shape for a remodeling territory. Drive time in this metro is not symmetrical, and neither is the work. Build targeting from counties and from the routes your crews actually take, then let each one have its own campaign and its own budget.
Williamson County, and Franklin and Brentwood in particular, tends to bring larger scopes, a designer in the conversation and more competitors bidding. Paying more for a click there can be correct. Paying that same price for every click in the account is not, and a single campaign forces exactly that trade.
Rutherford, Sumner and Wilson counties are different again, and Murfreesboro, Hendersonville and Mount Juliet are worth separating from each other once you have enough volume to justify it. Start with two or three geographic splits, not ten.
One setting is worth checking on day one. Location targeting should be limited to people in or regularly in your areas, not people merely showing interest in them, or you will buy clicks from every household planning a move who has typed the city name into a search bar.
A share of your clicks want a restaurant on Broadway, not a remodel
In this city the word kitchen pulls hospitality, tourism and job hunting, and the word bath pulls spas, so the exclusion list has to be built for Nashville specifically.
Pull the search terms report weekly for the first two months, then monthly. Read every line. The waste is never where you assumed.
Expect to exclude restaurant and bar queries, hiring and jobs language, apartment and rental listings, spa and bath house traffic, appliance repair, cabinet refacing if you do not offer it, and the endless supply of do-it-yourself questions. Add supply retailers and big box store names.
Stay on phrase and exact match until you have enough conversion history to justify anything broader. Broad match with a thin account is a way to donate money to the auction.
Keep a separate campaign for your own company name if anyone is bidding on it, and never let brand traffic sit in the same campaign as prospecting. Mixed together, the brand conversions make a failing prospecting campaign look healthy.
One bathroom and a whole house gut cannot share a budget or a landing page
Scope determines price, sales cycle and which crew runs it, so scope should determine campaign structure too.
Run separate campaigns for the job types you actually sell. A single bath. A primary suite. A kitchen where the layout holds. A kitchen where walls and plumbing move. Each one gets its own ads and its own landing page.
Ad copy should disqualify as hard as it qualifies. If you do not take jobs under a certain scope, say so in the ad. A click you never receive costs nothing, and every unqualified call costs somebody an afternoon.
Send the click to a page that matches the search. Somebody who typed a bathroom phrase and lands on a general remodeling homepage will bounce, and you paid for the privilege.
On bidding, small accounts rarely have enough conversion data to feed an automated strategy well. Start tighter and more manual, get the conversion tracking honest, and move to automated bidding once the account is regularly producing real outcomes rather than form fills.
January sells here, because the work happens indoors
Real winters, ice included, stop the exterior trades, and a kitchen or bath is the one project that does not care what the weather is doing.
Plenty of contractors cut their spend after the holidays. The auction gets cheaper at exactly the moment a household that just hosted twelve people in a bad kitchen is most annoyed about it.
An ice week is worth planning for rather than reacting to. Crews may not move, but inquiries still arrive, and the company that answers the phone during a week when nobody else does starts spring with a booked calendar.
Pace against the production calendar, not the month. If you are sold out into April, spend less and spend it on the scopes you are short on. If a job cancels, that is the week to push, not the week to cut.
Watch the platform's own pacing behavior too. Daily budgets can be overspent on individual days, and a campaign that quietly front-loads a month leaves you dark when a good week arrives.
Feed signed contract values back to Google, or the algorithm optimizes for junk
A form fill is not a job, and a bidding system that only knows about form fills will cheerfully buy you more forms from people you cannot serve.
Define stages and track all of them: inquiry received, qualified, in-home appointment held, scope priced, contract signed. Most remodelers stop at the first one, which is why their reporting looks better than their bank account.
Import the later stages back into the ad platform with values attached. Once the system can see which clicks became signed work, it starts buying more of them and less of everything else. Without that feedback it is optimizing toward whichever ad group produces the most cheap forms.
Use call tracking with dynamic numbers so phone inquiries are attributed properly, since a large share of remodeling contact still starts as a call. If you record calls, the rules on consent are worth confirming with your own counsel.
The number that matters at the end of the quarter is cost per booked job by campaign and by county. Cost per lead is a vanity figure and it will tell you the wrong campaign is winning.
The Gulch rental operator wants a scope and a date, so write the ad that way
Managers buying on turnaround time need different copy, a different form and a different measure of success from a homeowner planning a once-a-decade renovation.
A meaningful share of housing downtown and in the neighborhoods around it is run as a business. The person deciding may hold several properties, approve remotely, and care mostly about how fast you can be in and out.
Give them a separate campaign and a landing page that opens with turnaround, availability windows and invoicing rather than with design. The form should ask for unit count, access details and the dates the property is empty.
Search volume from this group is thin, so do not expect the campaign to carry itself on clicks alone. Pair it with remarketing to people who reached the operator page, and treat it as support for direct relationships with management companies rather than as a standalone channel.
Say nothing about short-term rental rules in the ads or on the page. Those rules are contested and change, and specifics are worth confirming with Metro rather than asserting in copy you cannot update quickly.
Questions we actually get
- What should we budget to start?
- Enough to gather real data in one geography rather than a token amount spread across five. Pick the county that produces your best margin, fund it properly for a full quarter, and expand from what you learn. A budget stretched thin across Davidson, Williamson and Rutherford at the same time usually produces three campaigns that never gather enough information to be optimized.
- Should we run Google and Meta at the same time?
- Not on day one. Search captures people already looking, which is the shorter path to a booked job. Meta earns its place later using finished project photography to create demand and to stay in front of people who visited but did not call. Adding it before search is stable splits attention and budget.
- How do we handle leads from outside our service area?
- Tighten targeting first, then handle the rest at intake. Ask for the county or the town in the first field of the form so an out-of-range inquiry never gets a callback promise. Keep a short referral list for the ones you cannot serve, since a clean handoff is worth more in a referral-poor market than a polite dead end.
- Are lead-generation platforms worth using alongside our own campaigns?
- They sell the same inquiry to several companies, so you compete on speed and price rather than on the work. Some remodelers use them to fill gaps in a soft month. Just track them as their own channel with the same booked-job math you apply to search, because they usually look cheap per lead and expensive per contract.
- How quickly should we expect the account to stabilize?
- Nobody can promise a timeline, and results depend on your close rate as much as on the ads. What is predictable is the sequence: the first weeks are spent cutting waste from the search terms report, the next stretch is spent getting conversion tracking to reflect booked work, and only after that does bidding optimization mean anything.
