Ad placement advice usually circulates as screenshots of somebody else's earnings, which is roughly as useful as screenshots of somebody else's gym progress. The truth is structural: a placement earns according to two variables, how many real impressions it collects in view, and how much attention it steals from the content the reader came for. Rank placements by revenue per unit of reader cost, and the confusion clears.
This guide does exactly that for blogs. Each major placement gets its earned value, its user cost, and its setup notes, drawn from Google's own format documentation and the recurring patterns of sites that monetize without bleeding readers. We finish with the mistakes that quietly drain accounts, because what you avoid earns as much as what you add.
Quick answer:Two Ideas That Anchor Every Placement Decision
Two ideas anchor every good placement decision. First, viewability: an ad nobody scrolls to has zero value regardless of its theoretical rate, which is why visible-by-default formats and early in-content positions outearn buried units. Second, interruption cost: every ad consumes attention, and attention is what makes pages-per-session and return visits happen. The earning question is always the ratio, never the raw revenue line, because revenue extracted this month at the cost of reader trust is a loan, not income.
Auto ads overviewPlacement beats density, and it is worth killing the density myth early. Adding a fifth or sixth unit rarely adds proportional revenue, because each additional unit competes for the same finite attention and typically lowers the viewability of everything already on the page. Auction rates follow real visible attention, not tag count, so three units in strong positions out-earn six in weak ones while costing you less reader goodwill. When you want more revenue, the sequence is diagnose viewability, improve position, test a different format, and only then consider adding a unit.
Anchor Ads: The Best Mobile Earner, Explained
anchor ads documentationThe setup is nearly free: enable them through Auto ads and Google handles position, size adaptation, and frequency capping. Dynamic anchors can expand, collapse, or stick to the corner, which lifts performance while keeping the dismiss control present. If you enable only one auto format on a blog, this is the one with the best evidence behind it.
Position settings deserve thirty seconds of attention after enabling. Anchors default to the top of the screen where allowed, and Google shifts them to the bottom whenever a top anchor would cover important interface elements. If your header hosts a sticky search box or a persistent call to action, test the bottom-anchored variant for a few weeks, because a top anchor that overlaps your navigation creates exactly the accidental-click surface policy cares about. The settings live inside the Auto ads controls, and the change applies site-wide without touching your template.
The Placement Rankings With Real Numbers Behind Them
In-article units are the manual workhorse. The highest-value slot is after the opening, roughly where the first section settles: the reader has committed to the page, the unit renders in view, and the ad appears between thoughts rather than inside one. A second unit near the two-thirds mark catches the readers who reach deep content, and that is usually plenty for a 1,500-word article. In-article formats render natively between paragraphs, matching your text style instead of interrupting it.
| Placement | Earned value | Reader cost | Setup notes |
|---|---|---|---|
| Mobile anchor | Highest on mobile-heavy blogs | Low, dismissible | Auto ads toggle; frequency capped |
| After-opening in-article | High, strong viewability | Low-medium | Manual unit after first section |
| End-of-content multiplex | Steady, catches finishers | Very low | Auto or manual; grid of related ads |
| Desktop sidebar (long posts) | Moderate on desktop traffic | Low | Sticky-adjacent layouts need policy care |
| Vignette between pages | High on multi-page audiences | Highest, interruptive | Frequency limit, then measure |
The multiplex unit at the end of content is the quiet consistent earner. Readers who finish an article are in a browse mood, and the multiplex grid serves related-content style ads exactly there, with no interruption cost at all because the article has already delivered its value. Place it after your conclusion and before comments or related-post modules, and let it catch the audience that manual in-article units were never going to reach.
Concrete placement maps make this usable, so here are the two templates that cover most blogs. On a standard post page: anchor on mobile, one in-article unit after the opening section, a second in-article unit around the two-thirds mark, and multiplex after the conclusion. On listicle or roundup pages: the same anchors and multiplex, with in-feed units between every third list item, which reads as rhythm rather than interruption. Archive and category pages usually deserve nothing beyond an anchor, because visitors there are navigating, not reading, and low dwell time makes every extra unit pure friction with negligible earnings.
Sidebars, Vignettes, and Measurement That Stays Honest
Sidebar advertising earns reliably on desktop but decays quietly on mobile, where sidebars collapse out of the flow entirely. It works best beside long-form posts where reading keeps the page open for minutes, and it earns little on quick-answer pages where bounce arrives before the second ad impression. Check your analytics split before investing design effort in sidebar units, and if desktop traffic is a minority share, a single anchor plus in-article units will outearn a polished sidebar you spent an afternoon tuning.
Vignette ads are the most powerful and the most dangerous format, showing full-screen between page views. Google allows frequency control precisely because uncontrolled vignettes can shred pages per session. The correct use is measured: cap frequency, enable for a three-week test, and compare revenue per thousand sessions alongside scroll depth and pages per session. Keep them only if the revenue lift survives its own interruption cost, and be honest about the comparison.
Measurement keeps all of this honest, and the reporting path takes two minutes to set up once. Build a custom report in the AdSense dashboard filtered by ad format, so anchors, in-article units, and multiplex stop hiding inside one blended line. Divide daily earnings by sessions from your analytics rather than impressions, because sessions price the reader's experience while impressions price the auction. Then change one placement at a time, wait three weeks, and compare the session-based number plus engagement metrics before deciding. Blogs that measure this way converge on a calm, high-earning layout within a quarter; blogs that chase screenshots rearrange furniture forever.
Mistakes That Drain Accounts Quietly
The fastest way to lose ad revenue is not a bad placement; it is a policy problem or a layout problem that torpedoes everything else. Account health comes before optimization, because a suspended account earns nothing while you appeal. These are the recurring self-inflicted wounds, and the first three have ended real accounts, not just annoyed readers:
- Ads adjacent to navigation, download buttons, or media controls, which risk invalid-click flags and account action.
- Units that shift content downward as they load, inflating layout shift and sending Core Web Vitals the wrong direction.
- Any layout encouraging accidental clicks, including ads styled to mimic content links or placed where double-clicking is habitual.
- Six units on a page nobody scrolls halfway, wasting layout space and lowering every unit's viewability.
- Sticky sidebar ads that overlap content on narrow desktop windows, a complaint magnet and a policy gray zone.
- Judging changes after three days, before auctions relearn and weekly patterns flatten out.
One final habit separates accounts that compound from accounts that plateau: a quarterly placement audit. Templates change, traffic mixes shift toward or away from mobile, and formats you disabled a year ago may have evolved. Spend thirty minutes each quarter walking your three busiest templates on a real phone, confirming units render where intended, nothing overlaps content, and the custom report still splits formats the way you expect. The audit catches drift early, and drift is what quietly turns a tuned layout into an average one while you were busy writing.
monetizing a blog beyond AdSense RPM versus CPC explainer