Bid separately on the homeowner with drawings and the homeowner with a folder of photos
Project stage predicts close rate better than project type does, and the two buyers need different ads.
One person has plans, a designer and a start month in mind. They search for a builder who can price a defined scope. The other has a saved folder of pictures and a vague number, and they search to find out what things cost.
Both are worth having. They are not worth the same bid, the same page or the same promise. Selling a fixed scope to the first is a straightforward pitch. The second needs a process, a range and a reason to talk to you before they have decided anything.
Split them at the campaign level so budget cannot drift from the near buyer to the browser. Ad copy for the ready group should name the scope and the calendar. Copy for the researching group should offer the pricing conversation, because that is the thing they are actually shopping for.
Track them separately from the first click to the signed agreement. If the researching campaign never produces a signature, cut it. If it produces slower but larger jobs, fund it deliberately rather than by accident.
Target the streets a trailer and a dumpster can actually get down
A mileage radius is the wrong shape for a market with hills, a river and three counties.
Distance on a map is a poor proxy for cost. A job in Pflugerville and a job in Bee Cave can be the same drive on paper and completely different in practice once you factor in the crossing, the hour of day and a driveway that pitches down to a garage below street level.
Build the targeting from where your crews already work profitably. Then apply bid adjustments in the places where you take the job but hate the logistics, and exclude the places where the answer has been no twice already.
Watch what MoPac does to a schedule. If a two truck day west of the city means your lead carpenter is unavailable to anything else, that geography deserves a higher bar, not the same bid as an easy run to Round Rock.
Revisit the map quarterly. Crews change, capacity changes, and a geography you could not serve in March may be exactly where you want volume in September.
Electrification and rebate searches look like remodel intent and are not
A large slice of expensive clicks come from people shopping a utility program, not a renovation.
Induction ranges, panel upgrades and efficiency incentives generate a lot of searching. Some of it is a homeowner planning a kitchen. Much of it is somebody trying to find out what their utility offers, and they will bounce off a remodeling page in seconds.
Austin makes the confusion worse in a useful way. Austin Energy is city owned, while much of the surrounding area buys power through a competing retail market, so program eligibility can depend on which side of a service boundary an address sits. Terms change, so never quote a program in an ad. Point people to their utility and keep those clicks out of your account.
Mine your own search terms report every week for the first months, not every quarter. The waste in a remodeling account hides in phrases nobody predicted, and by the time a monthly report surfaces it you have paid for it twice.
Build the exclusions in layers: retail intent, repair intent, rental and tenant intent, employment intent, and the DIY questions that will never become a contract.
Pace the budget against the design queue, not the install calendar
The bottleneck in a remodeling business is usually the front of the process, and spend should follow it.
Most advice tells you to slow down when the crews are full. For a design build remodeler the constraint often sits earlier, with whoever produces scopes and drawings and shepherds a project through any city review before work can begin.
When that queue is backed up, buying more inquiries produces neglect. Leads sit for a week, get a slow answer, and go elsewhere. Nothing in the account looks broken, and the money is gone anyway.
Set the spend by how many priced scopes your team can genuinely produce a month. If a project touches the exterior and picks up a review step, the front end holds that file longer, so the effective capacity is lower than the raw headcount suggests.
Keep a reserve for the weeks after a queue clears. Coming back to full spend quickly matters more than shaving cost per click by a few cents.
Count the site visits that survive the driveway
Cost per lead is a vanity number in a trade where a share of leads die at the first look at the property.
A form fill is not an outcome. Neither is a phone call. The first number worth optimizing is the appointment that turned into a measured scope you were willing to price.
Add one more filter that matters here: the visits that survive the site itself. An estimator who arrives at a hillside lot with no staging room, a protected tree over the only access, or a footprint change that opens a longer review conversation may leave without a viable job. Those visits cost you an afternoon and should be attributed back to the campaign that bought them.
Feed the outcome data back into the platform, but feed it clean. Import the appointment that produced a real scope, and later the signed agreement, so the bidding is being trained on the thing you sell rather than the thing your form counts.
Kitchens and baths carry very different values, so weight them. A single full kitchen can absorb a lot of small bath clicks, and an account judged on blended cost per lead will quietly starve the campaign making the money.
Put your project minimum in the ad text and buy fewer, better clicks
Price posture in the ad copy is the cheapest filter available to a remodeling account.
Every unqualified click you prevent is money you did not spend and an intake call your office did not have to sit through. Stating a starting point, a typical scope or a minimum project size does that work before anyone lands on the page.
Volume drops. Some owners find that alarming for two weeks, then find their consultation calendar has better people on it. Be honest about the tradeoff before you make the change so nobody panics at the first dip.
Keep the same posture on the landing page, in the same words. A price band in the ad and silence on the page reads as a bait and switch, and the visitor who feels handled does not fill in the form.
This is also where a market full of newcomers pays you back. Buyers who moved to Central Texas recently have no reference point for what a remodel costs here. The company that tells them plainly gets the call, and often gets it first.
Questions we actually get
- What should we budget to start?
- Enough to produce a readable number of appointments in a quarter, and never more scope than your team can price properly. We would rather start narrow on the geography and the closest to buying searches, prove a cost per booked site visit, then widen. Any specific figure quoted before we have seen your close rate and your average job value is a guess.
- Are Local Services Ads worth it for remodeling?
- They can be, and they behave differently from search ads. The intent skews toward faster decisions and smaller jobs, which suits bath work more than a full kitchen. Run it as its own line with its own expectations rather than folding the numbers into the search account.
- How do we stop paying for clicks from people who want a countertop?
- With a negative keyword list you actually maintain, weekly at first. Add exclusions for retail, repair, refacing, rental, DIY and employment intent, and put a project minimum in the ad text. Expect to prune for the first couple of months before the account settles.
- Should we advertise in Kyle, Buda and San Marcos?
- Only if you would genuinely take the job at the price those areas support and the drive does not wreck a crew day. The account should reflect a real decision about where you want to work, not a circle drawn around your office.
- Can you guarantee a number of leads per month?
- No, and we will not pretend otherwise. Auction prices move, competitors change their spend, and your close rate is yours. What we will do is report cost per booked site visit and cost per signed agreement, and tell you plainly when a campaign is not paying for itself.
