Programmatic Display Banners: The Standard Sizes, and Where the Real Waste Hides

Why banner sizes are standardized at all

Before the industry converged on a shared set of dimensions, every publisher's ad slots were sized differently, which meant a creative built for one site often didn't fit another without a redesign — expensive and slow, and the opposite of what programmatic buying is supposed to enable. The IAB's standard ad unit portfolio fixed that: a small set of common sizes — the 300x250 medium rectangle, 728x90 leaderboard, 160x600 wide skyscraper, 300x600 half-page, and 320x50 mobile banner among them — that the large majority of publishers agreed to support. Build one creative at each size and it can run across nearly any publisher's inventory without a custom fit. The 300x250 alone reportedly accounts for over 40% of display impressions on Google's own display network, which is as good a sign as any of how much the standard actually gets used.

Static, animated, and rich media — not the same creative decision

A banner isn't automatically a still image. The same standard sizes accept static JPEGs and PNGs, animated GIFs, and richer HTML5 units that can include interaction, video, or expandable panels. Each carries a different production cost and a different technical constraint — HTML5 units, in particular, typically face a separate weight budget for their initial load versus their total file size, because they keep pulling in assets after the ad first renders, unlike a static image that's fully loaded the moment it appears.

The trade-off is rarely just cost. Rich media and expandable units tend to carry more page weight and a slower load, which some publishers restrict or price differently precisely because it affects the rest of the page's performance, not just the ad slot. A static banner is the safest default for a broad, high-volume programmatic buy across many publishers with different technical requirements; an HTML5 or expandable unit is worth the added production cost mainly on a smaller number of premium, hand-picked placements where the extra interactivity has a real chance of being seen and used rather than scrolled past in a fraction of a second.

A worked example: banner blindness and what it actually predicts

"Banner blindness" — the tendency of experienced web users to subconsciously ignore anything in a recognizable ad-shaped box — is real and well documented, and it's the reason standard banner click-through rates have sat in the low fractions of a percent for years, far below what a first-time internet user in the 1990s produced on the same format. This is worth stating plainly rather than treating as an embarrassing secret: a low banner CTR is not, on its own, evidence a campaign is failing. Banners do a large share of their work through view-through impact — brand recall and consideration — that a click-through metric was never built to capture. Judging a prospecting banner campaign purely on CTR is applying the wrong yardstick to the format, not catching a real problem.

Where the waste actually lives, and it usually isn't the banner shape

The mistake worth naming honestly is spending review time debating banner creative — colors, copy, CTA button placement — while the real waste sits somewhere the creative team never looks: line-item and placement selection. In a 31-day pull across 27 advertisers in our own Amazon DSP book, one in five spending line items produced zero attributed purchases, but between them they burned only 0.7% of the total budget — 259 of 1,325 line items took money and returned nothing, totalling $4,927 of $733,336. That's a strikingly different shape from search-advertising waste, where a losing tail of unchosen search terms can run to a third of spend. On programmatic display, an advertiser chooses every line item, so failures tend to be small, visible, and easy to prune — which means the bigger lever isn't finding the losers, it's making sure review time goes toward the handful of line items carrying most of the spend, not the long tail of small ones that already cost almost nothing when they fail.

What to check before touching the creative

Before redesigning a banner because performance looks weak, check what the banner is actually being asked to do — a prospecting placement judged on last-click conversion will look worse than a retargeting placement running the identical creative, because the audiences are at different points in the funnel. Check spend concentration second, sorting line items by dollars rather than by name, since the handful of large line items usually explain more of the result than the whole long tail combined. Only once both checks are done does the creative itself deserve real scrutiny — and even then, a size or format swap is usually a cheaper first test than a full redesign.

Where Dr. DSP fits

Dr. DSP is Amazon DSP run as a managed product by Full Circle, which has managed more than $500M in revenue across 100+ brands. Standard banner sizes are the easiest part of a programmatic display campaign to get right and, per the data above, rarely the part actually costing a brand money. A reader who never buys anything from us should still leave this page knowing that a low banner click-through rate is not automatically a failing campaign, and that spend-sorted line-item review catches more real waste than a creative refresh usually does.