The fastest way to waste a restoration budget is to start with a keyword list nobody checked. Keyword Planner will happily hand you thousands of ideas. Some are the searches that book drying jobs. Many are the same search typed six different ways. Plenty are competitors' names, software, rentals and people looking for work. Bid on all of it and the account spends the first month learning what you could have told it on day one.
Here's the order we work in on restoration accounts. First, see who already shows up on your searches. Then let their websites seed Keyword Planner. Strip out the repeats and the junk. Whatever survives fills the small groups from Chapter 4.
Google's Ad Preview Tool Shows Who Bids on Your Searches Without Costing Anyone a Click.
Before you look up a single keyword, look at the auction you're about to join. Search "water damage restoration near me" and see who shows up: franchise locations, independent restorers, plumbers, and the lead sellers and call networks covered in Chapter 16.
Do that inside Google Ads, not on google.com. In your Google Ads account, the Tools menu holds the Ad preview and diagnosis tool, which renders the results page for any search you type into it. Google's help page recommends the tool over searching on google.com yourself, because the tool doesn't rack up impressions on your own ads. That matters once your own ads are running: Google's page says the tool keeps those checks from adding impressions to your ads' statistics.
Resist the urge to tap a rival's ad just to see its landing page. It charges them for a click that wasn't a customer, and it tells you nothing the ad itself doesn't. Write down the names and the websites instead. You'll need the websites next.
We Fed Seven Restoration Websites to Keyword Planner and Got 7,174 Ideas Back. 3,624 Were Actually Different.
Most people start Keyword Planner by typing words. It works better here to start it with a website. Paste in a rival restorer's site, pick your metro as the location, and Google answers with the searches it ties to that site, each with a monthly count and a bid range.
In one example market, Sacramento, we did this with eight restoration companies' websites, franchise locations and independents both. One site returned nothing. The other seven returned 7,174 ideas between them. Once we removed the ideas that showed up for more than one site, 3,624 different phrasings were left. Your own market will produce its own numbers, but expect the same pattern: heavy overlap, because restorers in one town sell the same few services.
Two things stood out. First, mold searches led most of the lists. Mold is a real service with real searches, and Chapter 15 gives it its own treatment, but it means a list built from competitors' sites leans toward mold even for a company that mostly dries out water damage. Second, the bid estimates came attached to every idea. Read them as a ballpark: they're built from past auctions, and the prices you actually pay will settle once your own ads run.
Keyword Planner Shows Six Ways to Type "Water Damage Restoration Near Me." Google Counts Them as One.
Scroll the ideas from those sites and you'll see something that looks like good news. In the example market:
- water damage restoration near me: 1,000 a month
- water damage restoration services near me: 1,000
- water damage restoration companies near me: 1,000
- water damage restoration company near me: 1,000
- water damage restoration service near me: 1,000
- water restoration near me: 1,000
Six searches, 6,000 homeowners a month. Except it isn't. Google's Ads API documentation describes how its keyword data treats close variants: it "deduplicates close variants (for example, merging "car" and "cars" into a single result ...)." Six separate rows that all show the same 1,000 fit that pattern: in practice, they're one pool of searchers counted six times.
That changes two things. When you estimate demand, count the pool once. And when you build the list, you don't need all six. Exact match isn't word-for-word anymore: per Google's guide to match types, it also catches searches with the same meaning. One phrasing per idea will do at launch. The rest are clutter that makes the account harder to read.
Competitors' Brand Names Show Up All Over Keyword Planner's Ideas. Every One Belongs on Your Negative List.
Give Keyword Planner a franchise location's website and it brings back the franchise's name in every combination: the brand with water damage, the brand with mold, the brand near me. Nationally the numbers behind those names are big. "servpro" shows about 90,500 a month in Keyword Planner, as many as the core category search itself. "paul davis restoration" draws about 22,200 and "puroclean" about 9,900.
Every one of those searchers is looking for a specific company. In our method, competitor and franchise names go on the negative keyword list for every campaign, so your ads stay off searches for someone else's business. Chasing searches for a rival's brand is a whole separate campaign with separate risks. A restoration company building its first account should leave it alone.
Your own company name is the opposite case. It goes in a brand campaign of its own from month one (Chapter 2), so a homeowner who already chose you finds you first.
"Xactimate" Draws 22,200 Searches a Month. Not One of Them Is a Homeowner With a Wet Floor.
The rest of the noise comes from the restoration trade itself and from homeowners trying to avoid hiring anyone. Nationally:
- Software and training: "xactimate," a name for estimating software, draws about 22,200 searches a month. "iicrc certification" draws about 5,400. These come mostly from people in the trade, not homeowners with a leak.
- Do-it-yourself: "dehumidifier rental" draws about 5,400 and "how to dry carpet" about 1,000. Some of these people will end up calling a pro, but not from that search.
- Jobs: "restoration technician jobs" and similar searches come from people who want to work for you, not hire you.
- Cleaning: one example-market site's list came back with 57 ideas containing "cleaning service," including "cleaning services" at 720 a month. That site's cleaning pages pulled in searches from people looking for a house cleaner.
All of these go on the negative list before launch. The list doesn't stay finished, though. These negatives are a starting set. When broad match switches on (Chapter 6), it digs up fresh variations of each one, which is why the search terms report gets read daily through that first month.
Mold Gets More of the Local Searches. Water Damage Pays More of the Bills.
In the example market, a mold search sat at the top of the ideas for six of the seven restoration websites we tested; only one came back led by water damage. That isn't Keyword Planner judging the sites. The two biggest searches in the local lists, "mold removal service" and "mold clean up services," each showed 1,300 a month, more than any water damage phrasing, so they float to the top of almost any restoration site's ideas. Volume isn't value: in Cleanfax's 2026 survey, 52% of restorers named water damage their most profitable service, against 19% for mold. Last, point Keyword Planner at your own address and look past the top rows. If your water damage searches show up readily, your pages are doing their job. If they barely appear at all, your water damage pages may be too thin for Google to connect them with those searches, and Chapter 4 is where that gets fixed.
What survives all the cutting goes into the small service groups from Chapter 4: water damage searches together, mold together, fire and smoke together, about five or six close searches per group. Here is what a first list looked like in the example market, using one phrasing per idea and the monthly counts Keyword Planner showed for the Sacramento area:
- Water damage: water damage restoration near me (1,000), water damage restoration (390), water damage repair (90), water mitigation (70), water extraction (30).
- Mold: mold remediation (720), mold removal (590), mold remediation near me (170), black mold removal (140).
- Fire and smoke: fire damage restoration (210), fire restoration company (30).
That's eleven keywords, and every one names a service a homeowner is ready to buy. Notice what isn't there: no "water heater leaking," even though it drew 210 searches in the same area. It's a real problem search, but at launch it would pull plumbing calls along with water damage ones. It waits for broad match and a page of its own (Chapter 4).
Notice also how small the local numbers are next to the national ones. In the example market, mold remediation came in at 720 a month and water damage restoration at 390, a sliver of the nationwide figures in Chapter 2. A metro area's water damage list may add up to a couple of thousand searches a month. That's normal, and it's why each search on the list has to earn its place.
Every keyword on that first list starts on exact match, with bids typed in by hand (Chapter 7). The first month's money stays on searches you chose on purpose, and the account starts recording which of them turn into signed jobs.
A starting restoration list doesn't need hundreds of keywords. A few dozen good ones, grouped tightly and backed by the right negatives, will teach Google more in the first month than a thousand loose ones.
Next: The Leak Searches Exact Match Never Sees.
A launch list built this way stays inside the lines you drew. That's a feature, not a limit. Wider reach comes later, on purpose, once the account has booked jobs to learn from and someone is watching the search terms. Chapter 6 covers the searches you didn't pick: how homeowners describe the leak in words no list predicts, and how broad match finds those searches once the account has enough booked jobs to steer it.
Want the whole plan first? The guide home lists all sixteen chapters. Chapter 4 covers the ad groups and pages this list feeds.
On restoration accounts we run for Google Ads management, the negatives go in first and the keywords second.
