The Drum Awards Festival - Extended Deadline

-d -h -min -sec

Media NFL Amazon

Amazon’s dismal NFL audience shows that TV’s naysayers deserve penalties


By Samuel Scott, The Promotion Fix columnist

October 9, 2017 | 8 min read

"Views" is a useless metric. I “view” ads on social media networks as I quickly scroll past them. I “view” banner ads on websites as I consequently ignore them. I "view" every TV show that I see for half a second as I channel hunt with the remote control.

amazon nfl

But tell that to many of the media reporters who covered Amazon Prime’s global live-streaming of the first of eleven Thursday Night Football games during this season of American football. The New York Post, Fortune magazine, and Reuters, for example, all highlighted NFL figures stating that a reported 1.9m people “viewed” the kickoff show.

At first glance, the number looked impressive -- especially for a tech company’s first foray into big-league sports programming, which cost Amazon a reported $50 million for the total package. The result seemed to lend credence to the opinions of people such as creative director JP Manterola and MobiTV chief executive Charlie Nooney who think that sports will soon be watched online along with everything else. (Of course, the latter has an obvious bias.)

But the picture becomes fuzzier as one digs more deeply.

Todd Spangler of Variety and Peter Kafka of Recode wrote two of the best media reviews of the event. Their articles put the right numbers in the correct context and stated that the average audience watching Thursday Night Football for at least 30 seconds on Amazon Prime Video was 372,000, with each person watching an average of 55 minutes.

Spangler and Kafka deserve credit for discussing average viewership in their stories and not highlighting the sensationalist 1.9m number instead.

(Note: I filed this column before the second Thursday Night Football game, which was scheduled to occur a few days before publication.)

Why is this important?

Many marketing pundits – myself included – discuss the positives and negatives of digital mediums compared to traditional ones. But such analyses are always going to be faulty because the metrics that online and offline marketers use are completely different. We have only apples-to-oranges comparisons.

TV and radio have average viewers and listeners. Print has average circulations. These numbers represent human beings, are usually audited by independent third parties, and have standard definitions throughout the industry.

Online ad platforms have “impressions,” which is the number of times that one browser made one request for one advertisement to load. Digital video platforms have “views,” which means little. Furthermore, these numbers are very prone to fraud, not yet independently audited, and are defined differently by every ad network in the context of measures such as “viewability.”

So, it is difficult to contrast online and offline because we cannot compare marketing apples to apples. There should be one metric to rule them all in the online and offline worlds. One example: as Facebook releases more original programming on the Watch platform, I challenge the company to report the exact same metrics as TV networks do.

Comparing online to offline

In the case of Amazon’s Thursday Night Football, we can finally make some concrete comparisons in the context of the single metric of average audience size. According to the Variety and Recode reports and my own calculations, here is what we can surmise:

  • 2.5% (372,000) of the total average audience (15.1 million) watched on Amazon Prime.
  • 97% (14.6 million) of the total average audience watched on traditional television.
  • 3.5% of Amazon Prime members viewed the game (there are an estimated 54 million in the US).

In another example, Business Insider advertising editor Mike Shields recently wrote that “cable TV should be worried” because the BuzzFeed food show Worth It had gained 280 million views. But again, “views” means nothing. Anyone can throw some money at a bot network or click farm and get millions of “views” of anything.

I want Buzzfeed to report the average viewership of Worth It consisting of a decent length of time. Then, we can see whether the company is truly anywhere close to television. Just remember that Yahoo’s streaming of an NFL game in 2015 garnered 15m “views,” but the number plummeted to 2.4m when television’s metrics were applied. The average game gets 10m to 20m on TV.

These are yet more examples of how television is nowhere close to being “dead”.

Quantity versus quality of views

It’s not only about the number of views. It’s also about the quality of views. Even if Amazon were to obtain a large viewership, it might not necessarily mean that the programming would be useful for marketers.

We in the marketing profession are in the business of selling stuff. So, our chief concern should not be which mediums are most popular but which ones are best at helping us to do our job. Whether more people watch American football on television or Amazon is not the primary issue.

Just because a medium is popular does not mean that it is good for marketing. A medium that is used by millions of people can still be a bad marketing channel. A medium that is used by few people can be a great marketing channel. Popular mediums and effective marketing channels are not always one and the same.

Mediums themselves have their positives and negatives. We need to know which ones are the best at selling stuff by their very natures. In this context, television soundly defeats online video such as Thursday Night Football as well.

In this email newsletter, Ad Contrarian Bob Hoffman summarizes a presentation by Karen Nelson-Field, a professor at the University of Adelaide and the founder of Media Intelligence Company, at a recent ReThinkTV event in Australia:

television attention

More people actively view an advertisement on television compared to the same ad on YouTube or Facebook. (All of the research is here.)

tv coverage cut through

Advertisements on television take up the entire screen on television but not on YouTube or Facebook.

tv sales

Hoffman notes:

“The combination of an ad's time, pixels and coverage Dr. Nelson-Field calls 'visibility.' It should be obvious that ads with higher 'visibility' would be more effective. But in an irrational advertising environment in which anything online or 'digital' has been imbued with magical properties, it is significant that someone as made a conclusive effort to check the science.”

People go online to get something done, which makes the cold internet a set of mediums that are best suited for tactics such as direct marketing. People watch television to have experiences such as watching sports on the large boxes in their living rooms – and that makes the warm medium best suited for advertising campaigns.

Marketers who think that television is “dead” need go back to the practice squads.

The Promotion Fix is an exclusive biweekly column for The Drum contributed by Samuel Scott, a global marketing speaker who is a former journalist, newspaper editor, and director of marketing and communications in the high-tech industry. Follow him on Twitter and Facebook. Scott is based out of Tel Aviv, Israel.

Media NFL Amazon

More from Media

View all


Industry insights

View all
Add your own content +