No Google Ads, no Meta Ads: what actually brings traffic to a classifieds directory

A new directory launches with a marketing budget set aside for paid advertising, the way any other website would launch. The first campaign is rejected within a day, sometimes with the whole advertising account suspended alongside it, and the money that was meant to bring the first hundred visitors is gone before a single one arrives. This is not a mistake in how the advert was written or who it was aimed at. It is a structural fact about the two networks that carry most of the paid traffic on the internet.

Which means the plan has to be built the other way round: assume those two channels do not exist, and decide from the first day where the money and the hours actually go. What follows is the policy position as it stands, what happens to the people who try anyway, and the channels that remain open, in the order they are worth doing.

Why Google and Meta will not carry these ads

Meta's advertising rules prohibit adverts that promote the sale or use of adult sexual services, and they name adult entertainment businesses directly. There is no exception for a business advertising its own directory rather than the services listed on it, which is the distinction most operators reach for first. The policy is written around what the destination offers, not around how the advert describes itself, so the distinction does not survive contact with a reviewer.

Google Ads keeps a separate policy for compensated sexual acts, which names prostitution, companionship and escort services and bans advertising for them outright. Google does certify dating and matchmaking apps to advertise on its network, and it is tempting to read that as the open door: a classifieds site is a place where people meet, after all. The certification explicitly excludes paid companionship and escort services, which closes the one path that looked available.

There is no appeal that reverses either policy, no disclaimer that satisfies it and no rewording that slips past it, because both companies review the page the advert points to and not only the advert itself. This is worth saying plainly, because a great deal of time gets spent on the assumption that the right wording exists somewhere. It does not. The policies are doing exactly what they were written to do.

What you lose is the account, not the campaign

Some operators try anyway. An advert with no sexual language, a landing page written to look like an ordinary directory of local businesses, a domain that gives nothing away. It sometimes runs for a few days before it is caught, which is long enough to feel like a discovery and to justify spending more on it.

Then it stops, and so does the account behind it. The penalty is normally applied at the advertiser account level rather than to the single advert, so the money already spent is gone and the account cannot be used for anything else either, including whatever other business the same person was running through it. That last part is what turns an experiment into a real loss.

Opening a second account rarely helps for long. The platforms match new accounts against the payment method, the business details, the device and the destination site, and a fresh account pointing at the same domain is the easiest pattern in the world to spot. Each attempt also makes the next one harder, and none of it produces a channel anyone could plan around, because the whole thing is one review away from ending.

The real cost is not the wasted spend, which is usually small. It is the two or three months spent finding all this out, during which nothing was built on the channels that would have kept working. A launch plan with a line in it for paid search and social is a launch plan with a hole in it, and the hole gets discovered at the worst possible moment, which is after launch.

So treat the ban as a feature of the terrain, the same way you treat the payment processors that decline the business, and move on. It is not an injustice to argue with, it is not a fault to report, and there is nobody at either company with the authority to make an exception. Everything below assumes those two doors stay shut.

Organic search, the channel nobody can switch off

Search is the channel this business runs on, and the shape that works is many specific pages rather than one page trying to rank for everything. One page per city, and inside a city one per category, each with its own title, its own text and its own listings. A single page that mentions forty cities ranks for none of them, and the reason is not mysterious: nothing on it answers any of the forty questions in particular.

Write those pages around what people actually type, which is almost always a place plus a category and rarely anything else. That phrase goes in the title, in the heading and in the first sentence, in the words a person would use rather than the words the industry uses internally. The traffic in this market lives in the long tail: hundreds of small precise phrases, each bringing a handful of visitors a month, adding up to more than any single broad term ever would.

The technical side is unglamorous and decides the outcome. Pages that load fast on a phone, a sitemap that is actually submitted, internal links that work, no two pages saying the same thing in different words, and nothing so thin that the search result tells the visitor more than the page does. A directory that already runs the checks that keep fake listings out has an easier time here too, because the pages hold up: real listings, numbers that answer, nothing a visitor abandons after two seconds.

The channels that are actually open

There is a real advertising market for this business, built around it rather than against it. Networks such as ExoClick, TrafficJunky and JuicyAds operate as ordinary ad networks: they buy and sell placement across adult sites, and a classifieds directory is a normal advertiser to them rather than an edge case to reject. Nobody there is going to close the account over what you sell.

Test one of them with a small amount before committing anything serious, and judge it on its own numbers rather than on habits formed with mainstream platforms. Traffic quality varies enormously between placements on the same network, so the unit to evaluate is the individual placement and not the network as a whole. Expect to throw away most of what you test, and expect the survivors to need watching, because a placement that worked last month can be resold to different traffic this month.

The channel almost everyone forgets is the advertisers themselves. Every one of them already has an audience, and every one of them has a reason to send it somewhere that makes them look established. Give each listing a clean shareable link, a decent preview when that link is pasted into a chat, and a simple way to reach it from wherever they already publish. This costs nothing per visitor, and it is the only channel that grows by itself as the directory fills up.

Then there is the traffic you already paid for once. Saved searches with a message when something new matches, a listing page worth coming back to, an alert when an advertiser someone follows arrives in their city. A returning visitor costs nothing and does more than a new one, and most directories have no mechanism at all for asking anybody to come back.

What to measure, and what to spend it on

Measure cost per contact, not cost per click. Clicks are what a network sells you and they are the wrong unit for this business, because the two things a visitor can actually do here, revealing a phone number and opening a chat, are the only events an advertiser will ever pay for. A placement with cheap clicks and no contacts is more expensive than an expensive one that produces them.

That means recording the moment a contact is revealed, per listing and per source, and letting that number decide the budget instead of impressions or time on page. It is a small piece of work and it changes every decision that follows, because it is the first time the marketing spend and the thing you actually sell are measured on the same axis.

As for the split: in the first months most of the effort belongs to search, because it compounds and nobody can take it away, while the spend on networks should stay small enough to lose without consequence. Keep in mind too that money spent has to be money that can also arrive, which is a separate project with its own timetable, and one that starts earlier than most people expect.

None of this is as convenient as typing a budget into an advertising dashboard and watching visitors appear. It is slower at the start, and it does not stop when the spending stops, which is the trade. The directories that survive their first year are the ones that stopped waiting for the two big doors to open and built the entrance somewhere else.