What Google Ads Cost for Guides

Every guide who asks this question wants a number, and every article that supplies one is describing somebody else's market. The reason is not evasion. It is that search advertising has no price list, and the mechanism that replaces a price list re-runs from scratch every single time somebody types something into the box.

What follows is the part that can be known. There is a public description of how the price gets set, published by the company setting it. There is a national price series for advertising services, collected monthly by a federal statistical agency, which tells you the direction of the weather even though it cannot tell you what it is doing over your water. And there is the one figure that actually decides whether any click price is affordable, which is not on any website because it comes out of your own books.

Why nobody can quote your click price

The platform's own help documentation is unusually candid about this. Its page on the ad auction describes it as a process that decides which ads are eligible to appear and in which order, or whether any ads meet the minimum required standard to show at all. Then it says the thing that ends the search for an average: the auction process repeats for every search, each time with potentially different results, depending on the context of the person's search and the competition among advertisers at the precise moment of that search.

Read that literally. Not per day, not per campaign, not per market. Per search. Two people in the same town typing the same words ninety seconds apart are running two separate auctions with two separate outcomes, because in between them a competitor's daily budget may have run out, or a new advertiser may have started, or the second person may be on a phone at a boat ramp rather than a laptop at a desk.

An average cost per click, then, is an average across auctions that no longer exist, in markets that are not yours, for queries you may never bid on. It is not that the published averages are dishonest. It is that the thing they average is not stable enough to average. Where this sits alongside every other route to a booking is laid out across the getting-booked collection, and the honest ranking of channels only starts once you stop looking for a rate card.

Triton Sportfishing, MA

What the auction actually decides

The sequence in the documentation runs in six steps, and the interesting work happens before anyone's money is involved. Somebody searches. The system finds every ad whose keywords match that phrase closely enough, then discards the ones that are not eligible, such as ads targeting a different country or ads that have been disapproved. Then it removes ads that do not clear the minimum quality standards required to show at all, asking whether the ads have a sufficiently high expected clickthrough rate and whether the ads and landing pages will provide a good user experience.

Only the survivors get ordered, and they are ordered on what the documentation calls Ad Rank, described as a combination of bid amount, the quality of your ads and landing page, the Ad Rank thresholds, the competitiveness of the auction, the context of the person's search, and the expected impact of assets and other ad formats.

Six inputs, and your bid is one of them. That ratio is the whole reason a guide with a good page can advertise against a booking platform with a marketing department and not simply be outspent into invisibility.

The clause that lowers your price

The documentation states the consequence outright: even if your competition bids higher than you, you can still win a higher position, at a lower price, with high-quality ads and landing pages.

That sentence is doing more work than most guides realise. It means the lever with the largest effect on what you pay is not a bidding tactic at all. It is whether the page the ad points at genuinely answers the search. A page about your operation in general, reached from an ad about tarpon in June, is a worse user experience than a page about tarpon in June, and the auction prices that difference into every click you buy for as long as the campaign runs.

Which is why the sequence matters. Build the page first, then advertise to it. The order is set out in the introduction to running search ads, and the same logic drives everything covered in pages built around one water and one town. An ad cannot repair a page that does not exist, and it will charge you a premium every day for its absence.

What the public record does say about ad prices

There is exactly one price signal on advertising that is collected systematically, published monthly, and free to read. The Bureau of Labor Statistics runs a Producer Price Index for advertising agencies, an industry series that tracks what the sellers of advertising services charge over time. It is an index rather than a dollar figure, which is the point: it measures movement, not level.

The series is worth reading once, because the shape of it contradicts what most people assume about advertising costs. Read straight off the series, June to June:

Two things jump out. In June 2020 the index sat at 137.8, below where it stood in June 2016, which means advertising services were cheaper in real index terms four years on. Then between the end of 2021 and the middle of 2022 it moved more than it had in the preceding five years put together, and it has climbed steadily since. The figures come from the series itself rather than from anyone's summary of it, and anybody can pull the same numbers.

Reading that index honestly

Now the caveats, because an index used carelessly is worse than no index. This one covers advertising agencies, which is a different thing from an auction for search clicks. It is national. It is monthly, not daily. And it is a producer price series, meaning it tracks what is charged rather than what is bought.

What it is genuinely good for is settling one argument. When somebody tells you that advertising has become impossibly expensive, or conversely that it has never been cheaper, there is a federal series that says the price of advertising services rose roughly twelve percent between its mid-2020 trough and mid-2026. Twelve percent over six years is a real increase and it is not a catastrophe. It tells you the channel has not run away from a small operator, and it tells you that anybody describing a collapse or an explosion is not looking at the data.

What it cannot do is tell you what a click costs on your water. No federal series does, because none of them observe auctions.

Where industry data stops being useful

The same limit applies further up. The Census Bureau runs a Service Annual Survey covering revenue and expense for service industries, with its tables built on the 2017 North American Industry Classification System, and it has been folding into a newer Annual Integrated Economic Survey launched in March 2024. Its 2022 data went out in January 2024. There is a published snapshot of service industries for 2021 giving estimated revenue and expense across categories.

Read the release dates. The most recent detailed figures were published roughly a year and a half after the period they describe, which is normal for a statistical programme of that size and useless for deciding what to bid on Tuesday. Classification is another wall: guiding does not sit in its own category, so whatever bucket contains it also contains businesses with nothing in common with yours.

This is not a criticism of the statistical system, which is doing a different job well. It is a warning about a specific failure mode, which is a guide spending three evenings inside federal data hoping it will produce a bid, when the number that decides everything was in their own booking records the whole time.

The only number that sets your ceiling

What a click is worth to you is fixed by what a booking is worth to you, and by how many clicks it takes to get one. Nobody else can supply either figure. The first is in your pricing, the second is in your booking history, and the second one is where almost every guide guesses when they could count.

Work it in that order and the ceiling falls out. Take what you keep from a trip after the costs of running it. Decide what share of that you are willing to hand to advertising, which is a business decision rather than a formula. Then divide by how many enquiries it takes you to land a booking, and again by how many visitors to the page produce an enquiry.

The discipline this imposes is the real value. A guide who has done it can look at a reported click price and know within a second whether the channel is viable, without any opinion about advertising. A guide who has not done it is negotiating with a number they cannot evaluate. The same backwards arithmetic applies to every paid channel, which is why it is worth doing once properly before comparing search against social or reading what the social side charges.

Shallow Water, FL

The costs that are not clicks

Click spend is the visible cost and rarely the largest one. Somebody has to build the campaign, write the ads, add negative keywords, read the search terms report and turn things off. That is time, and a guide's time in season has a price that is easy to establish, because it is what a trip pays.

An hour spent on a campaign in July costs whatever an hour of guiding costs, which makes July exactly the wrong month to be learning the interface. The building happens in the off-season, when the same hour costs very little, and the season gets maintenance only. Guides who invert that end up paying peak rates for setup work and then resenting a channel that was never given a fair test.

There is also the page itself, and whatever you spend on photography, and the phone that has to be answered. None of that appears on an advertising invoice, and all of it decides the return.

What experienced guides do differently

They start with negative keywords rather than bids. An hour spent listing every search you never want to pay for, meaning the tackle shopping, the free advice, the job hunting, the boat sales, will do more for cost per enquiry in a small market than any bid adjustment, because it stops the spend rather than optimising it.

They advertise narrowly and locally. One water, one species, one season, one radius. A campaign that tries to cover everything spends its budget on the searches you are least likely to convert.

They read the search terms report weekly, which is the only place the campaign tells you the truth about what people actually typed. And they treat the phone as part of the campaign, because a call that goes unanswered has cost the same as one that books.

They also know which enquiries came from where, which is impossible without a phone number or form used nowhere else. Guides who have run the unpaid side of search properly usually have this in place already, and the two channels then stop being confused with each other.

Common mistakes

Sending traffic to a homepage. The ad answered a specific question and the homepage answers a different one, so the visit ends. This single fault probably wastes more guide advertising money than every bidding error combined.

Judging the campaign on clicks. Clicks are what the platform counts and enquiries are what pays for fuel. A week of cheap clicks that produced nothing is not a good week.

Turning it on in the busiest month, discovering it is expensive, and concluding the channel does not work. It is expensive then because everybody else is bidding too. Seasonal timing is its own subject, treated for the social side in the timing piece, and the logic transfers.

Changing three things at once and learning nothing. And spending on brand-shaped advertising before there is any evidence people search for the brand. If nobody yet knows the name, buying the name buys nothing, which is a separate question from whether the name and mark are worth investing in.

What surprises people

The first surprise is how small a working budget can be. There is no minimum you must reach before the auction takes you seriously, because the auction is not evaluating your budget, it is evaluating Ad Rank on a single search. A campaign spending very little per day is competing on the same six inputs as one spending a great deal, and losing only on volume.

The second is that the cheapest clicks are frequently the worst ones. A broad, vague search is cheap because nobody bidding seriously wants it. The search that costs the most is usually the one from somebody ready to book, and paying $12 for that is often better business than paying $0.60 for forty people reading about knots.

The third is how quickly the answer arrives. Guides expect months of tuning. In a small market with a narrow campaign, thirty days of honest tracking usually settles it, because the volumes are small enough that a clear result does not need a large sample.

The fourth is that the platform's cheapest lever is free. Improving the landing page costs nothing per click and reduces the price of every click after it, permanently, for as long as the page stays good.

When not to advertise at all

If the page an ad would point at does not exist, the money is early. If the phone does not get answered inside a working day, the money is early. If the calendar is already full in the months you would advertise, the money is unnecessary, and a waiting list costs nothing.

If you cannot say what a booking is worth to you within a few dollars, the money is uncontrolled, because you have no way to recognise a bad result. And if the plan is to advertise instead of fixing something you already know is broken, advertising will make the broken thing more expensive rather than less visible.

There are also cheaper adjacent routes worth exhausting first. The lead-form product for local trades sits in a different place in the results and is priced differently, covered in the local services piece, and showing ads again to people who already visited is a smaller and usually cheaper problem, handled in the retargeting piece.

Reading the first thirty days

Set the question before you start, in writing, because a question invented afterwards will be answered by whatever the data happens to show. The question is how many traceable enquiries the spend produced and what each one cost.

Then look at the search terms report before anything else, because it tells you whether you bought what you meant to buy. Where most of the spend went to searches that were never going to book, the campaign was not too expensive, it was aimed wrongly, and that is fixable in an afternoon.

Check whether enquiries were answered, and how fast, before blaming the campaign. Then compare cost per enquiry against the ceiling you worked out earlier. If it sits under the ceiling, the channel works and the question becomes how much more of it you can buy. If it sits far above, the channel is telling you about your market rather than your settings.

Keep the spend records properly while you are at it, since advertising is an operating cost of the business and belongs in the same discipline as everything else in the deductions list.

When to stop

Stop when a month of clean tracking puts your cost per enquiry well above the ceiling and the obvious repairs have each had a turn. That is not failure. It is a cheap answer to an expensive question, and it releases the budget to something that works.

Stop also when you cannot tell where enquiries came from, because a campaign you cannot measure will be judged on feel, and feel is generous to spending. Fix the tracking or turn it off.

What is not a reason to stop is a bad fortnight, a fluctuating ad position, or one expensive click. Position varies by design, since the auction reruns constantly. Video is a separate decision with different economics again, taken up in the video ads piece.