Quick answer
A banner ad is a rectangular graphic advertisement, shown as an image, animation or HTML5 creative on a web page or in an app, that links to the advertiser's landing page when clicked. Standard sizes include 300×250, 728×90 and 300×600.
Formula: CTR = clicks ÷ impressions; CPC = spend ÷ clicks
What is a banner ad?
A banner ad is a rectangular graphic advertisement placed on a web page or in an app, usually as an image, animation or HTML5 creative, that links to the advertiser's landing page when clicked. It is the oldest and most common form of display advertising. What makes an ad a "banner" is its fixed shape and its position around the content, such as the top of the page or a side rail, rather than inside the text.
A banner ad has three parts you can measure separately: the impression (it was loaded or seen), the click (someone chose it) and what happens after the click on your own site. Most arguments about banner performance come from mixing up these three layers.
Where did the banner ad come from?
The first widely cited banner ran in October 1994 on HotWired, the web site of Wired magazine. It was an AT&T ad from the "You Will" campaign, and its copy asked "Have you clicked here? You will." According to the founder of the agency that built it, it initially drew a 44 percent click-through rate, as reported in Digiday's history of the banner.
Treat 44% as a historical curiosity, not a benchmark. It was a recollection of a novelty at a time when few people had seen an ad on the web, and nothing like it applies to the same format today. Other early clickable ads are also claimed to be first, so "first banner ad" is better read as "first widely remembered".
What sizes do banner ads come in?
Display sizes are standardised so one creative can fit many sites. The Interactive Advertising Bureau (IAB) publishes the New Ad Portfolio, first introduced in 2017 and last updated in February 2025. Ad platforms also publish the sizes they accept. Google's list of display sizes marks four as its top-performing sizes and lists others as supported:
| Size (px) | Name | Notes |
|---|---|---|
| 300 × 250 | Medium rectangle | Listed as a top-performing size |
| 336 × 280 | Large rectangle | Listed as a top-performing size |
| 728 × 90 | Leaderboard | Listed as a top-performing size |
| 300 × 600 | Half page | Listed as a top-performing size |
| 320 × 50 | Mobile banner | Mobile only |
| 320 × 100 | Large mobile banner | Mobile only |
| 970 × 90 | Large leaderboard | Supported |
| 970 × 250 | Billboard | Supported |
| 160 × 600 | Wide skyscraper | Supported |
| 468 × 60 | Banner | Classic size; still supported |
"Top performing" here is the platform's statement about its own network, not a guarantee for your page. Pick sizes by where your inventory sits and who buys it. If you are the advertiser, produce the four listed sizes plus one mobile size first, then add others only when a placement asks for them.
How are banner ads priced and counted?
Banner ads are sold mostly on two bases. CPM (cost per thousand impressions) charges for exposure; CPC (cost per click) charges for visits. See CPM for how the exposure price is built. The metric that links the two is click-through rate:
CTR = clicks ÷ impressions
Google's own definition uses the same ratio: 5 clicks on 100 impressions is a 5% CTR. The cost per click follows: CPC = spend ÷ clicks, and for a CPM-priced buy, spend = impressions ÷ 1,000 × CPM.
How do you read banner ad results?
Sample data: a campaign buys 1,200,000 impressions at a $2.50 CPM. The ad platform reports 840 clicks. On your own site, analytics record 690 sessions from the campaign and 21 sign-ups.
| Metric | Calculation | Result |
|---|---|---|
| Spend | 1,200,000 ÷ 1,000 × $2.50 | $3,000 |
| CTR | 840 ÷ 1,200,000 | 0.07% |
| Platform CPC | $3,000 ÷ 840 | $3.57 |
| Click-to-session rate | 690 ÷ 840 | 82.1% |
| Cost per landing session | $3,000 ÷ 690 | $4.35 |
| Session conversion rate | 21 ÷ 690 | 3.04% |
| Cost per sign-up | $3,000 ÷ 21 | $142.86 |
The gap between 840 clicks and 690 sessions is real. A click can fail before the page loads: the visitor closes the tab, the connection drops, or a non-human click is filtered. The cost you actually pay for a visit is therefore $4.35, not $3.57. Use the platform's numbers for the first three rows and your analytics for the rest. See conversion rate and cost per acquisition for the downstream metrics.
Is a high CTR a good sign?
Not by itself. CTR measures how often a banner got a click, not whether the click was worth having. Sample data: two creatives, each served 1,000,000 impressions.
| Creative | CTR | Clicks | Session conversion rate | Conversions |
|---|---|---|---|---|
| A: curiosity headline | 0.07% | 700 | 2.5% | 17.5 |
| B: specific offer | 0.03% | 300 | 6.0% | 18.0 |
Creative A has more than twice the CTR of B, yet B delivers slightly more conversions per million impressions. Ranking by CTR would pick the wrong creative. Rank by conversions per impression, or by cost per conversion, once you have enough conversions to compare.
Two more reasons not to chase CTR. First, there is no universal benchmark: CTR depends on placement, audience, size and what is being offered, so compare a creative with your own history on the same placement. Second, people learn to ignore the places where ads usually sit. Nielsen Norman Group calls this banner blindness, first documented in 1997 and replicated since. On one page it analysed, 132 of 148 eye fixations fell in the content area and only 1 landed in the right-hand rail where ads sat.
Can a banner ad hurt the page itself?
Yes, in two ways: it can make the page jump, and it can be annoying enough to break standards.
Layout shift. If the page does not reserve space for the banner, content moves down when the ad loads. web.dev defines the layout shift score as impact fraction × distance fraction, and rates a CLS of 0.1 or less as good and above 0.25 as poor. Sample data: a phone viewport of 390 × 844 px, with article content filling the top 600 px.
| Case | Content moves | Impact fraction | Distance fraction | Score |
|---|---|---|---|---|
| 300 × 250 ad injected above the text, no reserved space | 250 px | 1.00 | 250 ÷ 844 = 0.296 | 0.296 (poor) |
| 320 × 50 ad injected, no reserved space | 50 px | 0.77 | 50 ÷ 844 = 0.059 | 0.046 (good) |
| Slot reserved with the final size | 0 px | 0 |
The fix is to reserve the slot's height in the page layout (for example with a minimum height) and to show a placeholder rather than collapsing the space when no ad is returned, as the web.dev optimisation guide recommends.
Annoying formats. The Coalition for Better Ads' Better Ads Standards list experiences that fall below its standard. On both desktop and mobile these include pop-up ads and auto-playing video ads with sound, plus prestitial ads (on desktop, those with a countdown). Ad density above 50% on desktop or above 30% on mobile is also below the standard.
What should you check before running a banner campaign?
- Match the creative to the landing page. If the banner promises a specific offer, send the click to the page that shows that offer, not the home page.
- Produce the core sizes. 300 × 250, 728 × 90, 300 × 600 and one mobile size, then add others on request.
- Tag every destination URL with source, medium, campaign and a content value for the creative or size, so results can be split later.
- Decide the success metric first. Use conversions or cost per conversion for a performance buy; use reach and impressions only if you are buying awareness.
- Set a stop rule. For example, pause a creative after 1,000 clicks if its session conversion rate is under half that of the best creative.
- Exclude your own traffic and bots from the results so clicks from testers and crawlers do not inflate the picture.
Measuring banner ad traffic with TapCub
Back to the $3,000 campaign above. TapCub does not serve or buy banner ads and does not count impressions; it measures what happens after the click on your own site.
- Tag the landing URLs with
utm_source,utm_medium=display,utm_campaignandutm_contentfor the size or creative. In Web analytics › Sources, the UTM tab splits source, medium, campaign, term and content, with visitors, sessions, pageviews, bounce and duration for each row. See sources and UTM. - Set a goal (a sign-up event or a thank-you page) in Web analytics › Goals. The channel table in Web analytics › Sources then shows conversions per channel, which gives you the session conversion rate for each creative.
- Import the spend in All-platform analytics › Business › Channel ROI on the Costs tab, by day, channel, source and campaign. TapCub's cost per click is spend divided by sessions, so it will show $4.35 rather than the platform's $3.57, which is the click-to-session gap. Its cost per acquisition counts purchases on the order ledger, so for sign-up goals divide the spend by the goal conversions yourself.
- Keep bots out of the denominator. Recognised robots are counted separately in Web analytics › Bots, so they do not inflate visitors or lower your conversion rate.
- Watch the page after launch. Monitoring › Performance shows real-user CLS for the landing and article pages; it has no previous-period comparison, so compare two date ranges and mark the launch in Configuration › Annotations.
Sources
- Digiday, How the banner ad was born (HotWired, October 1994, AT&T, 44% click-through rate).
- Google AdSense Help, Supported ad sizes (display sizes and top-performing sizes); IAB, New Ad Portfolio.
- Nielsen Norman Group, Banner Blindness: Old and New Findings.
- Coalition for Better Ads, Better Ads Standards.
- web.dev, Cumulative Layout Shift and Optimize CLS.
Was this explanation clear?
We use your answers to fix and expand this page.
Couldn’t send. Please try again in a minute.
Thanks for the feedback. We’ll use it to improve this page.You’ve already sent feedback on this page. Thanks.

Product Adoption Consultant
Explores customer onboarding, live conversations, and self-service, turning common usage questions into practical guidance and improvements.