ComScore, known in the industry as the expert in digital measurement, which is taking an aggressive cross-platform position with its acquisition of Rentrak, also recently announced a measurement
deal with Viacom and partnership with Adobe. I sat down with Andrew Lipsman, comScore’s VP of marketing insights, to get his perspective on all of these recent advancements in conjunction
with comScore’s efforts at perfecting the calculation of true digital ad performance.
ComScore is embarking on work in reach and frequency measurement as a way to link devices and
platforms under common metrics. This effort is a result of the tech debate to measure viewability, fraud and ad blocking. The question comScore posed: What is the value of a digital ad? —
will, once answered, help identify clean impressions in terms of true, accurate reach and frequency.
“We conducted three different pieces of research over the past year, that ultimately
provided insight into what validated reach and frequency looks like when you qualify impressions based on the opportunity to see," said Lipsman. "We synthesized that research into our latest
Value of a Digital Ad report, which provided new insight, [showing] how a better understanding of those metrics changes what we thought we knew about digital ad effectiveness, such as brand and sales
lift.”
Reach measurement is challenged in digital, according to Lipsman. Gross reach and frequency in digital is overstated because it includes non-viewable and fraudulent
impressions.
There is also overstatement of reach due to server-side reportage versus unique audience visitor counts, which can often vary by a factor of two or three. This is because
server-side measurement counts the incidence of unique cookies, but there could be multiple browser uses or deleted cookies that are then reloaded. Also, when you overstate reach with server-side
analytics, you are also understating frequency by a corresponding amount.
Once you can account for gross reach and frequency correctly, you then need to examine the traffic of bots or ads
below the fold to determine the actual validated reach and frequency. “Roughly half of all digital ads are not viewable,” noted Lipsman. “When the reach and frequency are properly
validated, it is a much better indicator of how to effectively build reach and how much frequency can be tolerated. Validated reach and frequency, while lower than gross, offers more accurate insights
and less waste.”
In cross-platform measurement, it’s important how to understand digital in relation to television. Part of what makes digital comparable to TV is the opportunity
to view, but digital must be accountable for those instances where there is an inability to view. This is quite a challenge. According to comScore, only 48% of digital impressions are viewable. 45%
are non-viewable human (often those ads that are below the fold) and an additional 7% are non-viewable invalid traffic (bots).
This effort takes on a bigger context beyond the accurate
reporting of cross-platform performance. When you can optimize media allocations, advertisers can spend more effectively.