If you have ever stared at the Google Hotel Ads bidding screen and felt your eyes glaze over at “commission per stay” versus “CPC” versus “smart bidding,” welcome. I do this for independent and boutique hotels for a living, and I still think the interface was designed by someone who wanted you to give up and just hand the keys to an OTA.
So let me save you the headache. This is the exact framework I walk through when a hotelier asks me which bidding model to run. It is not theory. It is three questions about your actual business, and the answers point you to the right model almost every time.
First, what these bidding models actually charge you for
Before any strategy talk, you need to be crystal clear on the mechanics, because the names are slippery.
- Commission per stay — You pay Google a percentage of the booking value, but only after the guest checks out and the stay is completed. If they cancel, you pay nothing. This is the closest thing to the OTA model, except Google is feeding clicks to your booking engine, not capturing the guest on its own checkout.
- Commission (pay per conversion) — A variant where you pay when the booking is made rather than after the stay completes. You still owe the commission on bookings that later cancel. Read your account terms carefully, because the two commission flavors behave very differently for a cancellation-heavy property.
- CPC (cost per click) — You set a maximum bid and pay each time someone clicks your rate in the booking module, full stop. No booking required, no completed stay required. You are buying traffic and betting your booking page converts it.
- Smart bidding / commissions with Google’s automation — You hand Google a target (often a return-on-ad-spend goal or a commission target) and let the algorithm adjust bids per auction using signals you cannot see. It needs conversion data and accurate value tracking to do anything useful.
Here is the mental model I use: commission models push the risk onto Google, CPC keeps the risk on you, and smart bidding only works if you have already proven the unit economics so the machine has something to optimize toward.
The single biggest mistake I see: a small hotel switches to CPC because the per-unit cost looks cheaper than commission, then pays for hundreds of clicks from guests who price-shop, bounce, or book and cancel. Cheap clicks are not cheap if they do not turn into completed stays.
The three questions that decide it
Forget best practices lists. Your answer comes from three numbers you already know (or can pull in ten minutes).
Question 1: What is your cancellation rate?
This is the dominant factor, and it is the one most hoteliers underweight.
If your cancellation rate is high — say you run flexible rates, you are in a leisure market with fickle demand, or you take a lot of “I might come” bookings — then commission per stay is your friend. You only pay on stays that actually happen. Every cancelled booking that would have cost you a CPC click, or a pay-on-booking commission, costs you nothing.
If your cancellation rate is low — corporate-heavy, advance-purchase non-refundable rates, a wedding-and-events property where people are locked in — then CPC becomes much more attractive, because you are not exposed to paying for bookings that evaporate. You can buy clicks confidently knowing most of them stick.
I had a conversation with a boutique innkeeper who was furious about his ad spend. Turned out he was on CPC with a 35 percent cancellation rate on flexible rates. He was literally paying Google for clicks from people who booked, changed their minds, and rebooked somewhere else. We moved him to commission per stay and his effective cost dropped because he stopped paying for ghosts. (Illustrative, but I have seen this pattern more than once.)
Question 2: What is your real margin per booking?
Commission models are priced as a percentage of stay value. So the math only works if your margin can absorb that percentage and still beat your OTA cost.
Here is the reference point that matters: OTAs typically take 15 to 25 percent commission. That is your benchmark. If a guest is going to book through Booking.com or Expedia at 18 percent anyway, then a Google Hotel Ads commission bid of, say, 12 to 15 percent that sends them to your direct booking engine is a win — you are clawing back margin and capturing the guest’s email, their data, and the chance to upsell, instead of renting them from an OTA. I dig into that arithmetic in detail in the book-direct math post.
But if your rooms are thin-margin (heavy fixed costs, low ADR), a high commission bid can quietly eat your profit. In that case CPC with disciplined bid caps gives you a fixed, predictable cost per click that you can model against your conversion rate. You trade away the “only pay on completed stays” safety for cost certainty.
Question 3: How long is your booking window?
This one is subtle but it changes how much you can trust automation.
- Short booking window (guests book days before arrival): conversions happen fast, data feeds back quickly, and smart bidding has a fighting chance to learn. Last-minute urban and leisure properties fit here.
- Long booking window (guests book months out, common for destination resorts and event venues): the feedback loop is slow, cancellations have a long runway, and smart bidding struggles because it cannot see the eventual outcome for a long time. Manual or commission models give you more control while the picture is fuzzy.
My decision table
Here is roughly how those three questions collapse into a recommendation. Treat it as a starting hypothesis, not gospel — you still test.
| Cancellation rate | Margin | Booking window | My starting model |
|---|---|---|---|
| High | Healthy | Any | Commission per stay |
| High | Thin | Any | Commission per stay, low target |
| Low | Healthy | Short | CPC or smart bidding test |
| Low | Healthy | Long | CPC with bid caps |
| Low | Thin | Any | CPC, tight caps, optimize the booking page first |
| Mixed | Healthy | Short, high volume | Smart bidding once data is clean |
Notice that smart bidding only shows up when cancellations are under control, margin can absorb the spend, and you have volume plus a short window feeding fast signal. That is not most independent hotels on day one. It is where you graduate to, not where you start.
Why I almost never start a new account on smart bidding
Smart bidding sounds great — “let Google optimize” — and for high-volume accounts it genuinely earns its keep. But it is a hungry algorithm. It needs a steady diet of conversions with accurate values to learn, and a 20-room boutique property simply does not generate enough completed-booking signal in a month for the model to find its footing. You end up with the algorithm flailing on thin data, spending unpredictably while it “learns,” and you with no idea why.
So my sequence for a fresh account is almost always:
- Start with commission per stay (or CPC if cancellations are genuinely low). You get a clean, understandable cost structure and you start gathering real conversion data.
- Make sure tracking is honest. Your value tracking has to pass real stay values back, or every downstream decision is built on sand. This is also where your booking engine matters — a leaky checkout poisons the data and wastes the spend. I treat book-direct conversion as a prerequisite, not an afterthought.
- Once you have a few months of clean conversion data, carve off a slice of traffic to test smart bidding against your manual baseline. Let it prove itself before you give it the whole budget.
That sequencing protects you from the classic trap of handing a poorly-fed algorithm your entire ad budget and calling the result “Google’s fault.”
A realistic word on what this does and does not do
Let me be straight, because the industry oversells this stuff. Winning the Google Hotel Ads auction does not mean you have escaped the OTAs or that you will outrank them everywhere. It means you have bought yourself a seat in the booking module with your direct rate showing, right next to the OTA listings. That is a genuinely valuable seat — it gives the guest a clean path to book with you — but it is one channel in a healthier mix, not a magic bullet.
The realistic outcome of doing this well is a gradual shift: more direct bookings, a bit more margin clawed back, less of your inventory rented out at 18 to 25 percent. Over a few quarters, not a few days. Anyone promising you a guaranteed jump or a fixed return is selling you something I would not buy. The honest version is: you stack the odds in your favor and you measure.
And paid is only half the game. The other half is your organic and AI-search visibility — showing up when someone asks ChatGPT or Google for “boutique hotel near X” without paying per click at all. That is the compounding asset. Paid Hotel Ads buys you presence today; hotel SEO and AI search visibility build the presence that keeps paying after you stop bidding. If you have not yet, it is worth understanding how OTAs win the search results in the first place, because that is the structural problem all of this is fighting against.
The five-minute version
If you skimmed all of that, here is the whole framework in one breath:
- High cancellations? Commission per stay. Pay only on stays that happen.
- Low cancellations and healthy margin? CPC with disciplined caps, or test smart bidding.
- Thin margin? Watch the commission percentage against your OTA cost, and fix your booking page conversion before you spend a dollar more.
- New account or low volume? Do not start on smart bidding. Earn your way there with clean data.
- Long booking window? Stay manual longer. Automation hates slow feedback.
None of this is set-and-forget. I review these models quarterly per property, because cancellation rates drift, ADR moves with the season, and Google quietly changes how the auction behaves. The right model in January might be the wrong one by peak season.
If you want a second set of eyes on which model fits your property — your actual cancellation rate, your margin, your booking window — that is exactly the kind of thing I do on a free intro call. Bring your numbers and I will tell you honestly where I would start, and whether paid Hotel Ads should even be your first move or whether your money is better spent winning the organic and AI results first.