This past summer, much of my TV viewing was dedicated to watching the series “Downton Abbey” and “Breaking Bad” in their entirety. “Downton Abbey” continues this January, but “Breaking Bad” concluded its five-season run before I started watching it. I was concerned that I would accidentally learn Walter’s and Jesse’s fates before seeing the final episodes. My friends who had seen the series were quite accommodating. But it’s tough keeping secrets in the digital age, and unfortunately, I did learn what happened in advance of watching. Jesse’s outcome was revealed by Seth Meyers during this year’s Emmy broadcast. And Walter? Well, I guess I’m to blame, since I was stupid enough to read a New York Times article in which the first paragraph states “Warning: Contains spoilers about the new age of television.”
The article, by Emily Steel, discusses the social ramifications of revealing dramatic plot twists. She cites a study by Grant McCracken, which Netflix plans to use as the basis for a digital promotion which creates a flow chart to classify people by their propensity for spoiling. At the root of all this is an attempt to understand how people view television content in the age of time-shifting and streaming, which has critical impacts on TV’s business model.
As viewing patterns change, so does water cooler conversation. You can’t simply blurt out, “How crazy was it when Danny took out Joey last night?” You need to first establish that the episode was watched by the people in the room. But the burden seems to fall more so on the one who isn’t caught up; I fell a few episodes behind my friends watching “Sons of Anarchy” this fall – so I had to make sure that they weren’t talking about it when I was around.
But with all of this conversational jockeying going on, I needed to ask a pretty basic question: how much time must elapse before what happened in a popular TV show becomes “fair game” – no longer subject to Spoiler Alerts?
To find out, we surveyed TV viewers from our online consumer panel. We know from conducting new product development market research studies (using conjoint and Bracket) that the way a question is framed influences how respondents answer. We wanted to look at the issue from both sides, so we randomly split our sample into two groups, and posed essentially the same question to both: how many days need to pass before people can communicate freely about a show and not face criticism for spoiling it for someone who hasn’t watched it yet? We asked each group to assume a different role: one group was told to assume that they had just watched the episode, and the other group was told to assume the episode had aired, but they hadn’t watched it yet. We upped the stakes by describing the show as the final episode of a series that they liked a lot.
We speculated that those who had hypothetically watched the episode would propose a shorter “spoiler alert” timeline than those who hadn’t watched it yet, since they would be eager to talk about it. Were we right?
Well, sort of.
We started out by looking at the mean number of days the respondents proposed (the mathematical average). Those who saw the program have an average spoiler alert timeline of 13 days. Those who didn’t have a timeline nearly thrice as long – 36 days. This seems consistent with our theory that the people who see a show and want to talk about it are less sensitive about doing so than those who haven’t seen it yet.
But we dug a little deeper, and two other data points seem to contradict our initial findings. The median number of days is, in fact, one day higher for those who have seen the show than those who haven’t. And the percentage of people who say it’s okay to talk about the show the same day it airs (a 0-day spoiler alert timeline) is actually lower for that group than for those who haven’t seen it.
So taking in all this information, it seems that there’s a small group within the group of people who haven’t watched the show who would like a lengthy timeline for them to get their viewing in, but for the most part the two groups are similar.
We then dug even further. What seems to really make a difference is the actual viewing behavior of the people themselves. People who time-shift programs propose a longer spoiler alert timeline than those who only watch TV programs in real-time. This is consistent across all 3 data points. Moreover, it’s consistent whether they had hypothetically viewed the program or not.
So what did we learn? Those who time-shift are more sensitive to spoilers than those who watch in real-time. This makes sense; some people may simply forget that not everyone watches programming in the traditional manner. So the real moral of the story is, if you don’t want your program spoiled… watch it as close to the air date as possible. And choose your conversation partners -- and New York Times articles -- wisely!
VP / Research Management
Michele likes to hijack TRC's online consumer panel to get relevant answers to her burning research questions. She loves asking questions relating to her favorite hobbies - TV and movies, golf, casino gambling and travel - and more often than not the answers can be generalized across industries.