Ad Network for Publishers: Four Numbers That Decide Whether You Get In

What Each Ad Network Requires Before It Will Take Your Site

Bar length shows how high the entry requirement sits, not how good the network is. The units differ, so lengths are indicative rather than proportional. A dashed track means there is no number to clear: the site is read instead of counted.

Rows below the first are sorted by how hard the network is to get into, lowest bar first. The first row is the publisher of this page, disclosed above, and it sits at the top for that reason rather than on merit. Nobody is ranked by quality here, because an ad network that won't accept your site isn't a worse one, it's an unavailable one.

The first row works differently from the other six, and the difference is worth stating precisely. Every other network here gates mechanically: at 24,000 pageviews Raptive says no, and no amount of context about your audience changes that answer. AdsCompass sets no number and reads the site instead, which makes it the only route on this page where a small site can be accepted on what it is rather than on how much of it there is.

That cuts both ways, so here is the cost. You cannot self-qualify before you apply, and there is no published rule to hold the network to afterwards. Secondary sources have quoted three different figures in the past eighteen months — 30,000 monthly visits, 100,000 unique visitors, 300,000 pageviews — and not one traces back to AdsCompass itself. Treat all three as noise. The only way to get the real answer is to submit the site.

What Changed Between October 2025 and February 2026

Raptive went first, on October 16, 2025, cutting the minimum from 100,000 monthly pageviews to 25,000 and retiring Rise, the separate tier that had run at 50,000. Its own explanation is the interesting part: the company said traffic had become a less reliable measure, with AI inflating pageviews on thin sites and algorithms moving a site's numbers overnight. That is an argument against the pageview test made by one of the networks still using it. Two geo rules replaced the single number: sites between 25,000 and 99,999 pageviews need half their audience from the US, UK, Canada, New Zealand or Australia, and above 100,000 that drops to 40%.

Mediavine's answer came in January, and it wasn't a lower number so much as a different kind of number. From January 15, 2026 the main network asks for $5,000 of annual ad revenue, which you can only demonstrate by already running ads somewhere. Journey breaks that circle: it opens at 1,000 sessions, runs on the Grow plugin, and upgrades a site once trailing twelve-month revenue crosses $5,000. Publishers who have mapped the tiers report 70% on Journey against 75% after conversion.

Ezoic moved the opposite way on February 19, 2026, setting 250,000 monthly active users as the floor for new publishers. Sites integrated before the cutoff keep access regardless of size, and an Incubator program admits twenty applicants a month from below it. No network on this page has moved its bar since.

How to Pick the Best Ad Network for a Website You Own

Three things decide the answer, and only one of them is the size of your audience. The unit a network counts in, where your readers live, and what your pages are about will each disqualify a site on their own, in that order of how often they catch people out. Work through them before you fill in an application, because two of the three cannot be fixed by growing.

Why 50,000 Pageviews Is Not 50,000 Sessions Is Not 50,000 Users

Take one site: 20,000 monthly sessions, 2.5 pages per session, 1.4 sessions per user. The last two are assumptions, so substitute your own; the point survives either way.

The site has 50,000 pageviews, double Raptive's bar. It has 20,000 sessions, twenty times what Journey asks. It has roughly 14,286 monthly active users, or 5.7% of what Ezoic requires. Same site, three units, and the reading swings from twenty times over one bar to seventeen times under another.

Ezoic takes that figure from Google Analytics, where active users and total users are different metrics. Confirm which of the two your report is showing before you measure yourself against the 250,000 line.

Traffic Geography: Why a 26,000-Pageview US Site Beats a 60,000-Pageview One

Raptive publishes its geo requirement. Monumetric's appears only in publisher accounts, which consistently describe at least half of traffic from the United States, United Kingdom, Canada or Australia at the entry tier.

Compare the two lists. New Zealand counts toward Raptive's 50% and is absent from Monumetric's, so a site built on New Zealand traffic reads differently to each. A 60,000-pageview site with 70% of its audience in South Asia fails Raptive outright while a 26,000-pageview US site walks in.

Self-serve networks don't check where your readers live before accepting you. They pay less for some of them instead.

The Categories a Managed Ad Network Blocks by Default

Every managed ad network keeps a default block list, and Monumetric publishes its own in its FAQ. Blocked by default: gambling and betting, dating, references to sex and sexuality, drugs and supplements, get-rich-quick offers, weight loss, cosmetic procedures, religion, and anything that blinks or shakes.

It is a demand-side block, and the distinction matters. The list doesn't only constrain what you write about; it means advertisers in those categories cannot buy your inventory even when your readers are exactly who they want. For a sports-betting site or a dating blog the managed tier isn't a slow approval, it's a structural no, and the money for that audience sits with networks that accept the vertical.

What "Best Ad Network" Means on the Payout Line

Payout terms hide three separate questions behind one number. When does the money actually arrive, how much of the advertiser's bid survives the chain before it reaches you, and can you see the rate before you commit a page to it. A network can win the first and lose the other two, which is why the lowest threshold is rarely the best deal.

Payout Floors Are Set Per Method, Not Per Account

AdSense is the simple case, and the only ad network here with one flat number: a $100 threshold, paid between the 21st and the 26th each month, with verification starting at $10 and a PIN mailed by standard post that can take two to three weeks to arrive.

Everywhere else the floor depends on how you take the money. Adsterra publishes $5 through Paxum, $25 through PayPal and $1,000 by bank wire, paid biweekly on Net 15. PropellerAds advertises weekly payouts from $5 with a four-day hold. AdsCompass releases direct-publisher payouts on request from $50, Net 30 by default on RTB.

If the only question is which network pays out soonest on small earnings, the $5 networks win it and it isn't close. But a threshold only matters once you are earning, and on a site the managed tier has turned down you are earning nothing at all. Weighed against acceptance, the verticals a network will actually serve, and whether you can read the rate before committing a page, the difference between $5 and $50 is a rounding error on the first month. Whichever you pick, check the number for the method you'll actually use: a $5 threshold that becomes $1,000 on wire is not a low threshold for anyone who banks by wire.

Revenue Share or Auction Bid

Google states its terms plainly. Publishers keep 80% of AdSense for Content revenue after the buy-side platform takes its fee, working out to roughly 68% when the advertiser bought through Google Ads. The same help page adds one more line: if your site is linked to an AdSense platform partner, a second revenue share is taken after Google's.

Most managed ad networks are exactly that kind of partner. An advertised 75% is therefore 75% of what reaches the network, not of what the advertiser paid, and the figures aren't comparable across tiers. Journey's reported 70% sits downstream of Google's split the same way. Self-serve networks answer differently by publishing the bid instead of the percentage. The AdsCompass Traffic Calculator returns CPC and CPM by GEO and format before you place a tag and before you deposit anything, so you can hold a real number against the RPM you are earning today. None of the managed platforms will tell you what your inventory is worth until it is already running on your pages.

Formats a Publisher Can Run Below That Bar

Six formats are available on the publisher side of a self-serve ad network: push, in-page push, native, popunder, banner and video pre-roll. Telegram rewarded posts and rewarded video exist on the same platform but are buy-side only, so they aren't options for monetizing a site.

Classic push collects a browser subscription and keeps paying after the visitor leaves, the only format here with a tail. In-page push needs no permission prompt and works where classic push doesn't, including iOS Safari. Native and banner sit closest to what a managed network would have placed on the page anyway.

The Number You Apply With and the Number You Get Paid For

Every threshold in the table above is verified through Google Analytics. Every payout an ad network sends you is settled on rendered impressions. Those two counts used to be close enough to treat as one number, and they are separating.

An agent reading your page programmatically appears in analytics and fires no ad. A crawler respecting robots.txt may not appear at all. The gap runs both ways, no network publishes a coefficient for it, and the consequence is that you can clear a bar on paper and be judged unprofitable thirty days after approval. Segment your analytics by browser and device, then look at what survives once everything that cannot render an ad is stripped out. That residue is what an optimization team sees, and it decides whether you keep the account rather than whether you get it.

Strengths and Limits of a Network That Doesn't Publish Its Criteria

What to Verify Before You Paste the Tag

Two things are worth settling before a single tag goes live, because both are far harder to unwind afterwards.

Ezoic's grandfathering deserves a closer look, because the two pages describing it don't agree. The Incubator page protects sites active before February 18, 2026; the support knowledge base says February 19. For anyone who integrated that week, one day decides the status. The same knowledge base voids protection if the integration is removed for more than seven days, while publishers who received the February notice report ninety. Seven days versus ninety is the difference between a theme migration and a business decision, so get the answer in writing before you touch anything.

Then find the exclusivity clause. Managed networks want your display inventory whole, self-serve networks generally don't care, and which one you signed decides whether you can test a second source next quarter. If a threshold on this page has moved since we checked it, send the network's own page through contact and we will correct it.

Frequently Asked Questions