The first media plan I ever reviewed had 14 million impressions in it. Fourteen million. I remember thinking that sounded enormous, like the brand would be inescapable. Then the campaign ran, and nobody seemed to notice. That was my first real education in what an impression actually is, what it isn't, and why the gap between those two things has become one of the most consequential problems in modern advertising.
An impression is the single most fundamental unit of measurement in advertising. It is also, depending on who you ask, one of the most misleading.
An impression is a single instance of an advertisement being displayed to a user. One ad load equals one impression. If a banner ad appears on a webpage and the page is loaded, that's one impression, regardless of whether the user actually looked at it, scrolled past it, or even had the browser tab in the background.
In traditional media (TV, radio, print, out-of-home), an impression is an estimate of exposure. In digital advertising, it's a logged event: the ad server records that the creative was served. The distinction matters because digital impressions feel precise (they come with timestamps and user IDs) but they're measuring delivery, not attention.
The term comes from print advertising, where it originally referred to each physical impression of ink on paper. A newspaper with 100,000 subscribers generated approximately 100,000 impressions per ad placement. The concept carried over into broadcast media and eventually into digital, where it became the foundational metric for Cost Per Thousand (CPM) pricing models.
I find that most confusion around impressions comes from conflating them with advertising reach and advertising frequency. These three metrics are deeply related but measure fundamentally different things.
Impressions count total ad deliveries. If one person sees your ad five times, that's five impressions.
Reach counts unique individuals exposed to your ad. That same person seeing your ad five times counts as a reach of one.
Frequency is the average number of times each reached individual sees the ad. It's calculated as impressions divided by reach.
The relationship is simple: Impressions = Reach x Frequency.
This is why 14 million impressions didn't feel like 14 million people noticing the brand. With a frequency of 7, it was actually 2 million people seeing the same ads repeatedly. Understanding this math is essential for any marketer planning media buys or evaluating campaign performance.
| Metric | What It Measures | Formula | Example |
|---|---|---|---|
| Impressions | Total ad deliveries | Reach x Frequency | 14,000,000 |
| Reach | Unique individuals exposed | Impressions / Frequency | 2,000,000 |
| Frequency | Avg. exposures per person | Impressions / Reach | 7.0 |
| GRP (Gross Rating Point) | Reach % x Frequency | (Reach / Universe) x 100 x Freq | 140 GRPs |
The way impressions are counted varies significantly by medium, and the inconsistencies have been a source of industry debate for decades.
Digital display and programmatic. An impression is counted when an ad is served by the ad server. The IAB (Interactive Advertising Bureau) standard distinguishes between "served impressions" (the ad was delivered to the page) and "viewable impressions" (the ad was actually visible on screen). Since 2015, the industry has been moving toward viewable impressions as the standard, but adoption is still uneven.
Social media. Platforms like Meta, TikTok, and LinkedIn each define impressions slightly differently. Meta counts an impression each time an ad enters the screen. TikTok counts video impressions when the video starts playing. LinkedIn counts an impression when an ad is at least 50% visible for 300 milliseconds.
Television. TV impressions are estimated through panel-based measurement (historically Nielsen) combined with set-top box data. An impression represents one household (or individual, in newer measurement systems) exposed to a commercial. Connected TV (CTV) is changing this landscape, offering digital-style impression counting on the TV screen.
Out-of-home (OOH). Billboard and transit impressions are estimates based on traffic counts and visibility studies. The Geopath organization provides standardized OOH impression measurement in the U.S.