Hand Google's Keyword Planner a boutique hotel's web address and a thousand or two ideas come back. Scroll through them and a strange thing happens: most of what you're looking at isn't yours to buy. It's the names of the big hotel down the street, a resort at Disney World, a hotel in another city, a bar, a job posting. The list looks like a menu. It works more like a sieve, and most of what goes in shouldn't come out the other side.
That's not a flaw in the tool. It reflects how travelers search for hotels. This chapter walks through the research the way we run it, start to finish, and spends most of its time on the part that matters most for a hotel: deciding what to block.
Google's Preview Tool Shows Who Bids on Your Searches, and Nobody Gets Billed
Before you research keywords, look at who's already buying them. The obvious way is to google "boutique hotel french quarter" yourself and scroll. Don't, and above all don't click the ads to see where they go. Every click costs that advertiser money, and searching your own market over and over muddies your own numbers once your ads are running.
Google built a tool for this. It's called Ad preview and diagnosis, it lives in the Tools menu, and it lets you see the real results page for "pet friendly hotel garden district" as a traveler in Chicago on a phone would, while Google counts no impression against anyone. Run your ten most important searches through it with the location set to your city, and write down every advertiser you see: other hotels, chains, booking sites. Run them again with the device set to a phone, since Google weighs the device when it ranks ads and the lineup can change. Those names are your first competitor list, and the next step uses it.
A Competitor's Website Is the Best Seed Keyword Planner Can Get
Keyword Planner has an option to "start with a website": give it a web address and it suggests keywords based on what's on that site. Work down your competitor list one hotel at a time, pasting each one's web address in. Then feed it your own. Your own site shows you what Google thinks you're about, which isn't always what you think you're about.
We did this for 12 hotels in our example market, New Orleans, and each one returned somewhere between 328 and 2,146 keyword ideas. Every idea comes with a monthly search volume and a top-of-page bid range. Treat the bid range as a rough guide, plus or minus 30%; it's an estimate based on what advertisers have paid, and the price your own ads end up paying will depend on how well they match the search.
One setting matters more for hotels than for almost any other business: the location. Keyword Planner counts searches from the area you choose, and hotel guests mostly search from home. In our pulls, "hotels in new orleans" showed 40,500 searches a month across the US and just 3,600 from inside the New Orleans area. Set the planner to the whole country, or to the places your guests actually come from, and you see the real demand. Set it to your own city and you mostly see locals and people already in town. Chapter 9 makes the same point about the campaign's own location setting.
Two of Every Three Lodging Searches in Our Example Market Named a Hotel
When we pooled the 12 lists, removed duplicates and kept only lodging searches tied to New Orleans, we were left with about 2.1 million searches a month. About two-thirds of that volume named a specific hotel. The single biggest was "new orleans hilton riverside" at 60,500 searches a month, followed by Hyatt Regency (49,500), the Ritz-Carlton (40,500), the Roosevelt (40,500), the Royal Sonesta (33,100) and the Marriott (27,100). Behind them came more than 700 other hotel-name groups.
That two-thirds figure is a market-level number, and it's worth being precise about it. Any single hotel's list is mostly its own name plus generic city searches; it's only when you pool many competitors' lists that the full spread of hotel names shows up. The sorting was done by rule, so treat the share as directional. A quick check in a second city, using two Charleston hotels' websites, also turned up hotel names as a large share of each list. Your market will have its own numbers. The pattern is that travelers search for hotels they've already heard of, a lot.
For a hotel buying its first keywords, that changes the job. Someone searching "hyatt regency new orleans" has already chosen the Hyatt. Your ad on that search isn't competing for an undecided traveler; it's trying to talk someone out of a decision, and paying for the attempt. Other hotels' names go on the negative list. Your own name goes to its own brand campaign, as Chapter 2 set up. What's left after both is a much smaller pool of searches from people still choosing, which is exactly why every one of them is worth getting right.
A Disney Resort, a Hotel Bar and a Job Search Land in One Keyword List. None of Them Books Your Room.
Once the hotel names are out, the rest of the list still needs cleaning. Here's some of what our 12 example lists turned up, with monthly US searches:
- Other cities. "hotel port orleans" (90,500) is a resort at Walt Disney World that happens to share a name with the city. Searches naming other cities, from New York to Nashville, added up to more than 645,000 a month across the lists, though the rule we sorted by can't separate every look-alike name, so treat that total loosely. None of those travelers are coming to your hotel.
- The hotel's own bar. "monteleone carousel bar" (14,800) is a famous bar inside one of the example hotels. People searching it may want a drink, not a room. For that hotel, those searches belong in their own place, pointed at the bar's page, not mixed into the room-booking campaign.
- Job seekers. "hotel that hiring" (8,100), plus careers searches for individual hotels.
- Booking sites. Searches with Expedia, Booking.com, Tripadvisor or Priceline in them were tiny, about 4,600 a month combined, but they're people who've chosen where to book. Block them.
- Apartments. "apartment hotel" and similar searches for longer, kitchen-equipped stays, which only belong if that's what you sell.
Keep all of this in one shared negative keyword list rather than typing it into each campaign. A shared list can be attached to the brand campaign, Isolation and, later, Exploration in one step, and a name added once is blocked everywhere. Block other hotels' names as phrases, so "roosevelt hotel" also covers "roosevelt hotel new orleans parking." But read the list against your own name before you save it. If your hotel shares a word with a competitor, a careless phrase negative can block your own brand searches.
One technical point catches people out. Here's the catch with blocking: a negative is literal in a way a normal keyword isn't. Google's help page says negatives now handle misspellings on their own, but singular and plural forms and synonyms still have to be added by hand. Block "careers" and you still need "career" and "jobs."
A Hotel Shouldn't Block "Free." Travelers Use It to Find Parking and Breakfast.
Standard keyword research advice says to weed out searches with "cheap" and "free" in them. For most businesses that's sound: "free consultation" or "cheap quote" signals someone shopping on price alone. Hotels are different.
"hotels free parking" came back at 6,600 searches a month, and the city-specific version, "hotels in new orleans free parking," at 1,000. In a downtown where overnight parking runs $45 to $56 a night at most of the example hotels, that's a traveler telling you exactly what would win their booking. Free breakfast, free cancellation and free Wi-Fi are things travelers filter on, and if your hotel offers them, those are some of the most qualified searches you'll see. Blocking "free" outright would throw them away. Block the specific phrases that don't fit, such as "free hotel stay" or "free night," and keep the ones that describe what you offer.
"Cheap" is a positioning decision, not a rule. A budget motel near the airport should probably bid on "cheap hotels near the airport." A luxury property shouldn't. Decide what kind of guest you're built for, and let that settle it.
One Theme per Ad Group, on Exact Match. A Pile of Hotel Keywords Isn't a Campaign.
What's left after blocking gets sorted into the themes from Chapter 4: one ad group per neighborhood, landmark or amenity your hotel genuinely fits, five or six keywords each. Each one launches on exact match, so it only shows for searches that mean that keyword. Together those groups make up Isolation, which is what we call the first non-brand campaign: exact match, hand-set bids, searches you can name in advance.
At launch the negative list is a starting point, not a finished product. It gets its real workout later, when Exploration begins. That's our name for the broad-match campaign we build once Isolation is producing about 30 quality bookings a month, which for most hotels is around months two or three and sooner for bigger ones. Broad match brings in searches nobody predicted, including plenty of new hotel names, other cities and odd phrasings. Exploration's first month is a daily job: open the search terms report each morning and block whatever doesn't belong. Sort it by cost, not by clicks, and look first at the searches that spent money without a booking: a new hotel name, a city you don't serve, a question about a show or a game rather than a stay. After that, keep reading it weekly. The list never stops growing.
Next: Isolation Buys the Hotel Searches You Can Name. Exploration Finds the Rest.
Chapter 6 looks at the two campaigns side by side: why the predictable searches go first, and why the long tail of phrasings travelers actually type is where a hotel's account grows.
The earlier chapters set up booking tracking, the order of searches, what a click is worth and ad groups that match. The rest of the series sits on the hotel guide's home page.
Our Google Ads work for hotels writes the do-not-disturb list first and the keyword list second. Running it yourself? Put your three nearest competitors' websites into Keyword Planner this week, and count how many of the top 50 ideas are somebody else's name.
