Nielsen Claps Back At VAB, Corrects Amazon 'Errors And Misrepresentations'
- by Wayne Friedman , August 30, 2023
Nielsen has pushed back on criticism from the VAB concerning Nielsen's effort to integrate Amazon first-party data with its panel data for measurement of “Thursday Night Football” this coming season.
Sean Cunningham, president and CEO of the VAB TV advertising group, complained about the arrangement on Tuesday in a release: “This first/third party data mash-up has the effect of unfairly tipping the scales in Amazon's favor versus the other NFL programmers who have not been afforded the opportunity to supply Nielsen with their own first-party data.”
Nielsen says it has been in active discussions with other clients in doing the same for other clients.
“We look forward to bringing more such integrations into our measurement in the near future,” Nielsen Audience Measurement CEO Karthik Rao said in a letter sent to VAB's Cunningham and released to the press.
advertisement
advertisement
Cunningham said the VAB calls “upon Nielsen to not move forward with its plans to unfairly force first-party data from Amazon’s TNF property into the NFL third-party data sets.”
A week ago, responding to inquiries from Television News Daily, a Nielsen representative said, “This is a capability not just for Amazon, but something we are offering for any client with a similar need with live streaming programming."
For its first season a year ago, Amazon's Prime Video’s “Thursday Night Football” averaged 11.3 million viewers, according to Amazon’s analysis of its first-party metrics combined with Nielsen viewership measurement.
By way of comparison, traditional Nielsen measurement was 15% below this level -- averaging 9.58 million viewers per game.
If Amazon monitored all of the devices---including TV sets----that were tuned in per commercial minute---or whatever---- to its games that's one thing---providing the findings are independently audited. In theory, this gives you a more accurate measurement of how many devices were "reached' by an average commercial. But how did Amazon determine how many viewers were reached and what kinds---by sex, age, etc.? It may know who owns the device and for a cell phone that may be good enough. But what about a TV set where it's likely that more than one household member is "watching" in some---or many---homes? And that's not counting visitors to a household who may also be "watching".
If we are heading towards a new system that accepts a seller's independently audited data about the number and types of devices which were "reached" and all that Nielsen is doing is applying viewer- per -device factors---derived from the people meters---against such info to get a more "accurate" reading, that may be acceptable even though the resulting "impression" tallies vastly overstate the numbers of people who actually watched the commercials. But shouldn't all of the major sellers in a competitive set---like news, sports, etc. be done in the same manner -----if they are willing? Or are we going to create a situation where buyers can't compare the "impressions"---or CPMs---- from one seller to another---even though they compete directly for ad dollars?
It's a worrysome sitution but one that should be approached calmly and with the MRC's help regarding independent auditing. Perhaps it may develop into a positive step for certain kinds of "TV", though not, necessarily, for all TV buys.