• by Joe Casale , April 8, 2011

The "reach" metric used to measure digital media is broken. Media planners use comScore monthly ad network reach rankings to make informed buying decisions. Yet few question what these metrics mean -- and where the reach actually comes from. Solely relying on reach metrics as a tool to chart out media buys could lead down a dangerous path at the expense of advertisers.

Networks that purchase data from publishers for behavioral targeting are credited for the reach of the audience. A network can buy enough pixels to target everyone in the U.S. (pixel reach), but not serve a single ad impression (ad reach). Current measurement definitions don't differentiate pixels associated with ads served and pixels associated with data just sitting on a page.

A Tale of 3 Networks

Imagine three networks that all guarantee high reach to an audience target -- e.g., women with young children in the household. While these networks appear identical on the surface based on reach metrics, they are vastly different. Yet most media planners can't easily notice or assess it.

Network A: High data reach, low impression reach

Network A buys data across a wide net of large cooking and home sites to amass audience data on this segment. It is credited for the reach of every exposure its pixel has to this audience. This network has very few publishers -- no established customer base.

Insight: This network doesn't have any buying power, but it appears to reach everyone the advertiser is looking to target.

The truth about reach: Ad campaigns will never have priority on this network because the network has no buying power. This network needs to scale quickly, scrounging around for publishers to take their ads in order to fulfill the advertiser or agency's buy. Thus, the inventory and impressions will be of lower quality.

Network B: Low data reach, low impression reach

Network B is regularly chained between networks. It specializes in highly focused campaigns like remarketing and rarely buys directly, choosing instead to pass most impressions along to the next network. This network is credited with the reach of all the impressions it is exposed to, whether the network passed on them or not.

Insight: This network has no buying power and no data or targeting to an advertiser's audience segment.

The truth about reach: Ad campaigns will run in the worst possible locations, and the ads are likely to be poorly targeted.

Network C: High data reach and impression reach

Network C has the same reported reach as Networks A and B. The network buys a comfortable volume of data in the segment and accesses a large volume of impressions through publishers directly. It rarely passes on impressions and brings both targeting and strong buying power to the table.

Insight: This network can regularly sell impressions in sought-after verticals to a wide range of precisely targeted audiences by incorporating a solid volume of data into the campaign.

The truth about reach: This network seems like an obvious choice, but because of the way reach rankings are measured and reported today, it looks identical to Networks A and B.

The business dynamics and offerings of each advertising network are radically different. However, an unassuming advertiser cannot see the differences hidden behind deceptive reach metrics. How are media buyers supposed to compare real scale between networks? We must fix this problem and separate actual, data and potential reach measurements for ad networks.

Proposed solution

To bring clarity to these ad network numbers, we must disaggregate data from impressions. By separating potential reach (impression exposure) from actual reach (impression fulfillment), ad buyers will have an accurate context behind what type of inventory the network commands to have a better view of what advertisers would get from campaigns.

The industry would have three categories by which to gauge each network -- data reach, potential reach, and actual reach. In fact, the IAB initiated a measurement-working group focused on seeking a change to the current standards that we joined. The sooner we address this issue, the more accountable and transparent the industry will be.