All posts
Article1 min read

Banner Ads vs. CTV Ads: Raindrops and a Reservoir

A banner impression is a raindrop. A connected-TV ad is a reservoir. The market trades them as the same unit, and that is the whole problem.

By Shailin Dhar

A banner impression is a raindrop. It is scattered, fleeting, and worth almost nothing on its own. It lands somewhere below the fold, on a screen no one is looking at, and evaporates. You only get value from raindrops by collecting billions of them and hoping the arithmetic works out.

A CTV ad is a reservoir. It is a held pool of concentrated human attention: someone on a couch, leaned back, unable to skip, watching a 15-second spot on a 60-inch screen. That attention did not scatter. Someone dammed a river to collect it.

Same unit, different substance

The trouble is that both trade as an "impression." One word, one line item, one CPM comparison. But a raindrop and a reservoir are not the same commodity just because both are made of water. The buyer who compares a banner CPM to a CTV CPM and calls the banner "cheaper" has confused the price of the drop with the price of the pool.

  • Raindrops: abundant, cheap, near-impossible to verify one by one, easy to fake by the billion.
  • Reservoirs: scarce, expensive, watched by an actual person, far harder to counterfeit at scale.
  • The measurement stack treats both as interchangeable liters. It should not.

You cannot fill a reservoir with raindrops and call it a lake. And you cannot buy attention by the drop and expect it to pool.

Shailin Dhar

Fraud follows the unit that is easy to fake. Raindrops are trivial to manufacture, so the banner supply chain drowns in them. Reservoirs are harder to counterfeit, but the fraud simply moves upstream (spoofing the device, the app, the household) because the payout per unit is so much larger. Price the substance, not the label, or you will keep buying weather and calling it water.

AdTechMediaMeasurement

Keep Reading

AI-assisted recommendations of related work elsewhere on the internet.