Google Ads for boat tours and excursions works when you bid on seats, not clicks. A tour operator sells a fixed number of places on fixed dates, so the budget should follow open capacity by departure date, spend hardest 7 to 21 days before departure, pause dates that are sold out, and count a booking only on the confirmation page.
Why does every operator in a town bid on the same keywords?
Because the guest types the same thing. In Zadar it is "Kornati boat tour", in Split "Blue Cave tour from Split", in Dubrovnik "Elaphiti islands day trip". Each town has a handful of these phrases, and every operator with a boat bids on them.
- Zadar and Biograd sell Kornati and Telašćica.
- Šibenik and Primošten sell Krka and Kornati.
- Split, Trogir, Kaštela and Omiš sell the Blue Cave and Hvar five-island trip, plus Cetina rafting and canyoning.
- Makarska sells Brač and Hvar, and Biokovo.
- Dubrovnik sells the Elaphiti islands, Mljet and Montenegro day trips.
The same keywords, the same fixed seats. Once a boat is full, one more click is money in the sea. Yet most accounts we open are built as if clicks were the product. Being the fifth is the same as not existing. Being first for a sold-out date is worse, because you paid for it.
Should a boat tour operator optimise Google Ads for clicks or for seats?
Seats. Two numbers matter: utilisation (how many of your seats you fill) and rate (what you keep per seat after commission). Click-through rate, impression share and cost per click are steps on the way, not the goal.
Our client Malduk Adventures in Omiš sells rafting, canyoning, kayaking and zipline on the Cetina river. We have run its Google Ads for three years. Day-tour bookings went from 2 or 3 a week to 10 to 15 a day, cost per acquisition has fallen 20% year on year, and the company has served over 10,000 customers. That did not come from buying more clicks. It came from treating the account like a boat: a number of seats, per date, and a budget whose only job is to fill them at the best rate.
The two ways of running an account look like this.
| Metric | Keyword-led account optimises for | Capacity-led account optimises for | What goes wrong in the keyword-led account |
|---|---|---|---|
| Budget | One daily budget for the whole season | Budget follows open seats per departure date | You keep paying for a Saturday that sold out on Tuesday |
| Main number | Clicks, click-through rate, cost per click | Seats filled, revenue per seat after commission | Cheap clicks on a full boat look like success in the report |
| Structure | Ad groups by keyword theme | Campaigns by product and departure town | Split and Zadar share one budget and the cheaper town takes it all |
| Bidding | The same bid every day of the year | Bids rise 7 to 21 days before departure, ease off outside that window | You outbid everyone in March for August seats nobody is booking yet |
| Sold-out date | Ads keep running | Date paused, money moves to the next open date | Guest clicks, sees "sold out", books with a competitor and remembers you as full |
| Your own name | Not bid on, "they'll find us anyway" | Defended with a small exact-match brand campaign | A reseller's ad sits above your site and takes the booking with its commission |
| Conversion | Page view, or any form submission | Booking confirmation with the booking value | Google learns to bring people who look, not people who pay |
How should I structure a Google Ads account for excursions?
By product and departure town, not by keyword theme. A keyword theme is how Google thinks. A boat leaving a pier at 9:00 with 12 seats is how you think, and the account should match you.
One campaign per product and departure town
"Blue Cave from Split" and "Blue Cave from Trogir" are two campaigns. Different guest, different price, different boat, different capacity. The same goes for "Krka from Šibenik" and "Krka from Split". If the pier and the seat count differ, the campaign differs.
Ad groups by intent inside each campaign
Inside a campaign, split by what the guest wants: the tour name, "from [town]", "private", "price", "half day". Each ad group gets its own ad text.
One shared negative keyword list
Ferry timetables, "free", jobs, school trips. Every account we open is paying for some of these, and removing them is usually where the first two weeks of savings come from.
How do I bid by days to departure and remaining capacity?
You know something Google does not: how many seats are left on each date. Use it.
Spend hardest 7 to 21 days out
That is when the guest has booked the apartment, knows the dates and is choosing what to do. Dates inside this window get the most money.
Ease off outside the window, but do not stop
Beyond three weeks, keep a low presence so the name is familiar when the guest comes back to decide. Inside seven days, the remaining seats decide the bid, not the calendar.
Pause a date when it is sold out
If your booking system exposes availability, this can be automated with a script or a rule. If not, it is a ten-minute check every morning. Either way it happens daily. Nobody should click an ad for a full boat.
Put the money on the weak dates
A Tuesday with eight empty seats gets the budget. The Saturday that sells itself gets less. A "maximise clicks" setting does the opposite on its own, so the rule has to come from you.
Should I bid on my company name when resellers do?
Yes. Marketplaces such as GetYourGuide and Viator bid on operator names and destinations. A guest who searches for your company by name has already decided. If a reseller's ad sits above your site at that moment, the booking that was yours arrives through the reseller, minus its commission. A direct booking typically saves 15 to 20% in platform commission, and an exact-match campaign on your own name is the cheapest campaign in the account. Say "official site" and "book direct" in the ad and give one reason, such as free photos or a free date change. The same logic runs through our guide on Booking.com commission versus direct bookings. Counted after commission, the brand campaign usually pays for itself first.
Which source markets need their own campaigns?
Croatia recorded 110.1 million overnights in 2025. Germany alone delivered 22.3 million of them, followed by domestic guests with 13.8 million, Slovenia 11.1 million, Austria 8.2 million, Poland 7.3 million, Czechia 4.9 million and the UK 4.1 million (Croatian National Tourist Board, January 2026). Split-Dalmatia County had 20.9 million overnights, Zadar County 15.5 million, Dubrovnik-Neretva 9.3 million and Šibenik-Knin 6.6 million.
Those guests search in their own language before they arrive, so campaigns should be split by language, not lumped into one English campaign with "all languages" ticked:
- German for Germany and Austria, together the largest foreign group on the coast.
- English for the UK, Ireland and every guest who plans in English.
- Polish for Poland, the fourth foreign market.
- Czech for Czechia.
Each gets its own ad text, its own bids and a landing page in that language. A German ad that lands on an English page loses most of what the click cost. If a language cannot be done properly, do fewer languages well. Which markets matter most for your town, you already know from your guest book. The account should mirror it.
What does a landing page for an excursion need to show?
One product. The price. Departure times. What is included and what is not. Live availability for the dates the guest is looking at. This year's photos of this boat. The meeting point on a map. The cancellation rule. One button.
Do not send a "Blue Cave from Split" click to a homepage with 14 tours on it. The guest came for one thing; give them that thing and a way to pay. If your widget cannot show availability, at least show departure days and a price, and make the button lead straight into booking. For a tour operator, these pages are most of what our marketing funnel creation work is.
How do I track a booking instead of a visit?
The conversion is the booking confirmation, with the booking value, and nothing else. Not a homepage visit. Not a click on "Book now". Not a contact form, unless enquiries genuinely become bookings and you count them separately.
If your booking system lives on a third-party domain, cross-domain tracking has to be set up in GA4 and Google Ads, or a share of bookings disappears from the report. Until Google sees real bookings with real values, automated bidding is guessing. This is the first thing we check in any Google Ads management engagement, and it is why our monthly report is one page in euros after commission, not a dashboard of clicks.
What changes when trip planning starts in an AI assistant?
Google AI Overviews now appear on roughly 45% of searches, and a growing share of guests start with a question to ChatGPT, Perplexity or Gemini: "best Blue Cave tour from Split for a family", "is Kornati worth it from Zadar". After our AI visibility work for Malduk Adventures, AI-driven results rose 48% in two months and over 100 bookings in that period were attributed to AI assistants.
That changes what the ad account is for. The assistant recommends you, often on the strength of third-party sources, since brands are about 6.5 times more likely to be cited through other sites than their own. The guest then opens Google and types your name. If a reseller's ad is there and yours is not, the recommendation just paid a commission. The ad account and the AI visibility work are one system. Getting recommended is covered in our guide on being recommended by ChatGPT and in our AEO and AI SEO management service. Ads inside ChatGPT are a separate question, and for a small operator the answer today is mostly "not yet"; our guide on ChatGPT ads in Croatia explains why.
Why this matters for Zadar, Šibenik, Split, Makarska and Dubrovnik
Every destination has its own guest and its own booking rhythm. The windows below come from the account we run for Malduk Adventures and from what operators tell us. Replace them with your own dates once you have a season of tracked data.
| Destination | Departure towns | Typical guest | Peak booking window before departure |
|---|---|---|---|
| Kornati and Telašćica | Zadar, Biograd, Šibenik | Couples and families on a week-long stay choosing one full-day trip | Medium: a full day out is planned a few days ahead, rarely on the morning |
| Krka waterfalls | Šibenik, Primošten, Split, Trogir | Families and day visitors; the trip that goes when the beach does not | Short and weather-driven: decided a day or two out, often on the day |
| Blue Cave and Hvar five islands | Split, Trogir, Kaštela, Omiš | Younger groups and couples; the one big trip of the holiday | Medium to long: peak dates fill first, so planners book before they arrive |
| Cetina rafting and canyoning | Omiš, Split, Makarska | Families with teenagers and active couples | Short: chosen once on the coast, usually for the next day or two |
| Brač, Hvar and Biokovo | Makarska | Families on riviera stays, often returning guests | Short to medium: one sea day and one mountain day per week |
| Elaphiti islands and Mljet | Dubrovnik | Couples and city-break guests on shorter stays | Short: shorter stays mean faster decisions |
| Montenegro day trip | Dubrovnik | Guests on longer stays who want a second country | Longer: a border crossing and a full day get planned before arrival |
In Šibenik and Omiš, where the window is short, most of the budget belongs to people already in town and to the last seven days. In Split and Dubrovnik, where guests plan further out, the home-market campaigns carry more weight. Zadar sits in between. Makarska is the clearest case for separate campaigns per product, since the sea day and the mountain day compete for the same guest's one free morning.
What should be ready the week before the season?
- Capacity per date, per product, in one sheet or connected from the booking system. Without this, capacity-led bidding is a slogan.
- Conversion tracking tested with a real booking: confirmation fires, value is correct, cross-domain works if the widget is external.
- One campaign per product and departure town. Everything else paused, not deleted.
- Brand campaign live: exact-match on your name, common misspellings, your name plus "reviews" and "price".
- German, English, Polish and Czech campaigns, with ads and pages read by a native speaker, not only a translation tool.
- One shared negative keyword list on every campaign.
- Landing pages checked: this year's price, times and photos, and a booking button that works on a phone.
- The daily rule agreed: who checks sold-out dates every morning and moves budget to the weak dates.
- First-month budget set from open seats times the cost per booking you can afford after commission, not from last year's spend.
- A weekly habit: seats filled, cost per booking after commission, brand impression share. Three numbers, ten minutes.
Most of this is winter work, which is why we say the season is won in February. Our winter marketing calendar for Dalmatia lays out the months.
What does it cost and what do you get?
From €300 a month. The first month runs with no contract; after that, 90-day terms. You own the account, the pixels, the creative and the data, and they stay yours if you leave. You get a one-page monthly report in euros, counted after commission, and a reply within one business day. On accounts we take over, wasted spend is usually gone within two weeks and enquiries move within a month.
We will also tell you when not to spend. If you run one boat with 12 seats and your partners fill it by June, Google Ads will not make you money, and we will say so. Spend it on photos, on a page that shows availability, or on nothing. The account only makes sense when there are seats to fill at a rate worth defending.
If you want to know whether that is your situation, start with the free visibility check. We look at how you appear when a guest searches and when they ask an assistant, and tell you plainly what we would and would not do.