Last updated: July 31, 2026
Performance Media Buyer
I have been buying push, popunder and native inventory since 2016, running performance campaigns across Tier 1 and Tier 3 GEOs at monthly budgets between four and five figures. Since Google's 2024 core updates I have spent an increasing share of my time on the other side of the transaction, advising site owners whose traffic and monetization stack were built on assumptions that stopped holding.
I write the comparison on this site and I am the person to contact when something in it is wrong.
How I got here
Buying media teaches you to read the sell side sceptically, because you are paying for what it produces. Years of watching which source IDs convert and which are padded, which GEOs price rationally and which are quietly arbitraged, leaves you with a specific and slightly cynical understanding of what an impression is worth and where the money goes between the bid and the payout. That understanding is unusual preparation for writing about publisher monetization, and it is the reason this site pays more attention to fee stacking than to headline revenue-share percentages.
The move toward the publisher side was not planned. After the 2024 core updates a number of site owners I knew went from stable organic traffic to half of it inside a quarter, and the advice available to them was written for a market that no longer existed. Most of it still is. That is the gap this site tries to close.
What I actually know
Fee stacking and revenue share. Google publishes its terms plainly: publishers keep 80% of AdSense for Content revenue after the buy-side platform takes its fee, which works out to roughly 68% when the advertiser bought through Google Ads. The line most people miss sits further down the same help page — if a site is linked to an AdSense platform partner, a second revenue share is taken after Google's. Most managed networks are exactly that kind of partner, which means an advertised 75% share is 75% of what reaches the network, not of what the advertiser paid. Comparing those percentages as though they measure the same thing is the most common analytical error in this subject.
Entry requirements and what they actually gate. Networks measure in four different units and each of them disqualifies different sites. I track what each network currently requires, in the unit it uses, and I check it against the network's own documentation rather than against other comparison articles.
Formats, and what each one costs you. Push collects a browser subscription and keeps earning after the visitor leaves, which makes it the only common format with a tail. In-page push needs no permission prompt and works on iOS Safari where classic push does not. Popunder earns the most per visit and carries the most policy risk. These are trade-offs about the direction of a site, not placement tests, and I write about them that way.
Demand-side category blocks. Managed networks publish lists of advertiser categories they switch off by default — gambling, dating, supplements, weight loss and others. That is not only a constraint on what you write about: it means advertisers in those categories cannot buy your inventory even when your readers are exactly the audience they want. For some sites that single fact settles the entire question of where to monetize.
How I work
Every threshold published under my name is checked against the documentation of the network that set it, and the document is named in the text so a reader can look it up without taking my word for anything. Where a figure exists only in publisher accounts and not on the company's own site, I label it as reported in the sentence where it appears rather than presenting it with the same confidence as a documented rule.
I do not publish invented measurements. If a claim would require an account I do not hold — a withdrawal time, an approval turnaround, an RPM from a live integration — I say the claim cannot be evidenced and leave it there. There is a real cost to that rule, because a fabricated case study is more persuasive than an honest gap, and readers cannot tell the difference. That is precisely why the rule has to hold.
No network named in the research has reviewed the text before publication. The commercial relationship behind the site is disclosed on about us, and the sourcing rules I work to are set out in full in the editorial policy.
Where I am wrong
The honest limitation of this site is that it is documentary rather than experiential. I can tell you precisely what Raptive requires because Raptive publishes it, and I can tell you where its own pages contradict each other. I cannot tell you how long its approval queue was last month, because I have not been in it.
Publishers who have been through those queues know things I do not, and I would rather publish their experience with attribution than pretend to it. If you have applied somewhere recently — approved or rejected — the dates and the outcome are genuinely useful to other readers, and I will credit them.
Contact
Corrections, source links and questions about method are welcome through contact or via LinkedIn. Corrections come first and are usually made the same day. My current research is the comparison of Ad Network; site conditions are in the terms of use.