Popunder

What it is

A full-page advertisement that opens in a new browser tab beneath the active window. It loads on a user-initiated click, which is why anti-AdBlock catches less of it than display, and why the format has outlasted every prediction of its demise since I first covered it at AdExchanger in 2014.

The comparative framing

The interesting question is not “is popunder good.” It is “popunder versus what.” Three head-to-heads worth running before you commit.

• Popunder vs interstitial. Both go full-screen. Popunder gives the user the active tab they wanted plus a backgrounded creative they can ignore until they close the foreground; interstitial blocks the foreground for a frequency-capped window. For iGaming player acquisition in Tier-1 between Q2 2024 and Q1 2025, parallel-buy data across PropellerAds, Adsterra and Adsy showed popunder eCPMs roughly 2.1x to 2.4x interstitial at matched targeting. Interstitial wins on creative impact for app-install offers where a 5-second sequence is needed.
• Popunder vs in-page push. Popunder is the volume play; in-page push is the consideration play. The in-page push unit gives you 30 characters of headline and a 192x192 image, which forces a tighter creative discipline than popunder’s “the whole landing page is your creative” affordance. For VPN trial signups in DACH, in-page push converted at 1.6x the popunder rate per click in the same period, on roughly one-tenth the click volume.
• Popunder vs native. Different products entirely. Native is editorial real estate at scale; popunder is interstitial real estate at higher friction. Mixed-buy testing is the right answer here, not picking one.

Where popunder is the wrong call

• Offers that require more than five seconds of consideration before the click. The format does not reward dwell-time creative.
• Brand-safety-sensitive advertisers whose legal teams will not sign off on adjacency. The publisher network for popunder skews to long-tail inventory; that is the structural reason the eCPM is what it is.
• Tier-1-only buys at sub-$500 monthly spend. The auction dynamics mean small advertisers in the deepest GEOs are bidding against the largest, and the AM allocation reflects that.

Specs

• Pricing models: CPM and CPA available across the major networks; CPM dominates Tier-1, CPA dominates Tier-2 and Tier-3.
• Devices: Mobile and desktop. Mobile is the larger pool by volume; desktop carries higher per-click intent on most verticals I have seen the math on.
• Geo coverage: 240+ countries claimed by the major networks. Real depth, in my testing, is Tier-1 plus a roster of Tier-2 GEOs that varies by network. Trust the publisher concentration disclosed in the network’s investor or partner deck before you trust the country count.
• Anti-fraud: Multi-layer filtering with ML scoring on every impression is the table-stakes claim. The real test is whether the network credits back fraud caught by your tracker after a campaign ships. Two networks in my Q3 2024 testing did; the rest negotiated.

Methodology in the appendix. The verdict above. If the verdict surprises you, the methodology will explain why.