Broadcast & Media564 words
Average minute audience against reach
A game can be reported as having four million viewers and nine million viewers in the same week without anybody lying. The two numbers answer different questions, and only one of them is a rating.
What is in this piece — 5 sections
Two numbers are published about the same basketball game. One says 4.1 million. The other says 9 million. Both come from the same measurement company, and both are accurate. The difference is not spin; it is a different question being answered, and the answer is in the metric's name.
Average minute audience
The quoted “rating” for a game is normally the average audience per minute across the broadcast window. Take the audience in every minute of a two-and-a-half-hour telecast, add them up, divide by the number of minutes. This is the currency advertising is bought in, because an advertiser is buying a minute, not an evening.
Its property worth understanding: it punishes duration. A long, tight game with a half-empty first quarter and an enormous fourth has a modest average and a huge peak. The average minute audience is the right number for selling advertising and the wrong number for describing how many people cared.
Reach
Reach, or unique audience, counts people rather than minutes: how many distinct viewers watched at least a qualifying amount. For a long event it is always the bigger figure, often by a factor of two or more. It is the right number for the question “how many people saw this game” and the wrong one for “what was an advertisement worth”.
Both are legitimate. The failure mode is comparing one to the other — a reach figure for this year's final against an average-minute figure for last year's, presented as growth. That comparison is wrong by construction and it is extremely common.
What out-of-home measurement adds
Sport is watched in bars, in gyms, in airports and in barracks, and for years that audience was simply missing. Nielsen's national out-of-home service was built to capture it. In the company's own description of the launch, the service leans on portable people meters carried by roughly 77,000 panelists drawn from 44 local markets, projecting out-of-home viewing across about 65% of US television households, and combines that with the in-home national panel to quantify the incremental audience out-of-home viewing delivers.
Those are the numbers Nielsen published for its own method; we quote them as its description of the method rather than as a verified census. The practical point for a reader is directional and not in dispute: sport gains more from out-of-home measurement than almost any other programming, so a year-over-year comparison that straddles a methodology change is measuring the method as much as the audience.
Three questions that settle most audience arguments
- Average minute, or unique? If the release does not say, it is usually average minute — and if it is unusually large for the slot, it is usually reach.
- Does it include out-of-home? And did the comparison year?
- What window? Live, live plus same day, live plus seven days. Each is a different number for the same programme, and all three get called “viewers”.
Who is allowed to carry the game to those viewers in the first place is in blackouts and the word exclusive, and what happens to a local rights holder in bankruptcy is in the regional networks piece. The donor material on this desk looks at the editorial consequences — bet-centric shows and real-time commentary. Desk index: Broadcast & Media; full index: the archive.
What we read
Every figure above comes from one of these documents. If a number is not in them, it is not in this piece.