In a restaurant town, the word kitchen buys you commercial build-out clicks
Your most expensive waste is traffic from a completely different industry that happens to share your vocabulary.
The city runs on restaurants, and Google does not know you only work in houses. Searches for hood systems, walk in coolers, food truck fit outs, equipment repair and commercial build outs will find a broad kitchen campaign and spend residential money on them.
Build the exclusion list before launch rather than after. Commercial, restaurant, hood, cooler, equipment, supply, wholesale, repair, jobs, salary, apprentice, DIY, refacing, cabinet doors only, and every appliance brand you do not install.
Then read the search terms report weekly for the first two months. Nothing else you do in the account will return as much per hour. After that, monthly is usually enough as long as somebody actually does it.
Match type does more of this work than the negative list ever will. Broad match paired with an automated bid strategy is very good at finding the commercial traffic on its own, and it will do so with your budget.
Orleans, Jefferson and St. Tammany deserve three budgets and three ad texts
One campaign covering the metro hides the fact that you are subsidizing the work you least want.
Three parishes, three markets. An Uptown raised house, a Metairie slab ranch and a Covington subdivision are different scopes at different prices sold to different people, and they should not be sharing a budget or competing for the same impressions.
Split them at the campaign level so each carries its own money and its own report. Otherwise a cheap Kenner click quietly funds itself by starving the Orleans Parish campaign that produces your better jobs.
Set location targeting to presence rather than presence or interest. The default setting shows your ads to people researching New Orleans from other states, which is not the same as people standing in a house here.
Decide about the Northshore deliberately. Crossing the lake for a measure and again for the job is real cost, and if you want that work it deserves its own budget, its own ad text and its own landing page instead of leaking in through a loose radius.
Naming historic district review in the ad filters clicks before you pay for them
Ad copy is a qualifier you are already paying for, so use it to repel the wrong buyer.
An ad that mentions raised houses, older homes and work in protected neighborhoods speaks directly to the owner in Bywater or the Garden District. It also gently discourages the person with a 2004 slab who wants the cheapest possible counter swap.
Fewer clicks at a higher cost per click can be a much cheaper account. The number that matters is cost per booked job, and a click you did not buy costs nothing.
Do not promise an approval in ad copy or on the landing page. Say you have worked in these districts, and say that review requirements are worth confirming with the city or the relevant commission. Honest hedging in the ad survives contact with the sales call.
If you also want straightforward suburban slab work, run it as its own ad group with its own plain language and its own page. Two audiences, two conversations, one account.
Bid on the out of level floor and the failing shower pan, not on the word remodel
People search the symptom before they search the service, and the symptom is cheaper.
Sloping floors in an older house. A soft spot in front of the tub. A shower pan that has started leaking into the room below. Cabinet doors that stopped closing a year ago. Owners type those before they ever type the word renovation.
Those searches convert more slowly and cost less. Run them in their own campaign with their own budget so a longer path to sale does not distort the numbers on your main campaigns.
Send them somewhere honest. A page explaining how you diagnose the problem, what a paid assessment includes, and when the right answer is a licensed structural engineer rather than a remodeler. Free estimate language attracts the wrong response to that query.
Track them separately for at least two quarters before judging. Symptom traffic often looks poor on a thirty day view and looks fine on a six month one.
Spend should follow the selections backlog, because one stalled job blocks the next start
Your constraint is calendar capacity, not budget, so pace against the calendar.
A kitchen occupies a crew for weeks and occupies your designer well before that. Once selections are backed up, another signed contract does not help you. It hurts the jobs you already sold.
Agree on a comfortable number of active jobs and a comfortable design backlog. When you get within one job of the ceiling, reduce spend by a set percentage rather than switching campaigns off.
Switching off and back on costs stability and costs whatever the bidding system had learned. Reducing budget keeps the account alive at a lower volume, and ramping back is a single change instead of a restart.
The reverse is equally true. A soft month is exactly when spend should go up, and most remodelers do it backward because the soft month is also the month cash feels tight.
A kitchen closes over weeks, so the default attribution window undercounts you
Short reporting windows make good accounts look bad and produce confident, wrong decisions.
Somebody clicks an ad in the first week of research, talks to a spouse, sits on it, comes back through a branded search two months later and signs. Default settings will credit whatever touched last.
Extend the conversion window to the longest sales cycle you can actually measure. Import the signed contract as the conversion, not the form fill, so the system optimizes toward money instead of toward volume.
Accept that recent spend has not finished paying out. The last thirty days of an account always looks worse than it is, which is why so many remodelers cancel campaigns a month before they would have worked.
Judge the account by quarter and by cost per signed contract. Cost per lead is a useful diagnostic and a terrible way to decide anything.
Questions we actually get
- What should we actually count as a conversion?
- The signed contract, or at minimum a booked in-home measure that a person confirmed. Form fills and phone calls are useful for diagnosing which part of the account is working, but if you optimize toward them the platform will find you the cheapest possible form fills, and cheap form fills in this trade are usually people pricing a countertop on a Sunday.
- Should paid search run alongside SEO or after it?
- Alongside, if the budget allows. Paid search answers this month's calendar and tells you within weeks which language and which parishes produce booked work, which is the best possible input into what to write organically. Organic pages take longer and cost less once they land. They are complements, not alternatives.
- Do we need separate landing pages for each parish?
- You need separate pages when the content is genuinely different, which for this trade it usually is. An Orleans Parish page can speak to raised houses, crawl space plumbing and exterior review. A Jefferson Parish page speaks mostly to slab houses and a different price posture. Copying one page and changing the parish name helps nobody and reads as filler.
- How much of the budget goes to bathrooms versus kitchens?
- Split them into separate campaigns first, then decide, because they have different economics and different sales cycles. Bath searches are usually higher volume and lower value, kitchen searches the reverse. Sharing a budget means whichever gets clicks faster eats the money, which is rarely the split you would choose on purpose.
- Is Meta worth running for a remodeler here?
- It works differently. Search buys people who already decided to look, and social buys attention with photography. Finished work in recognizable neighborhoods can perform, but instant lead forms tend to produce inquiries that go nowhere for a considered purchase. Treat it as demand creation with a longer payback and judge it separately from search.
