Winter is coming are you ready for it?

VALAR MORGHULIS.

‘Dragonstone,’ the Season 7 premiere of Game of Thrones, was the most-watched HBO premiere of the decade, attracting an audience of 16 million viewers. Of these, more than one third streamed the episode through the HBO GO app. As winter approaches for Westeros, fans of the show settle in for a blisteringly-hot final six episodes. With over 6.7 million mentions on Twitter and 65 million trailer views on YouTube already logged, season 8 figures to be one of the most buzzed-about TV finales of all time, and a cultural touchpoint rivaling the finale of the Sopranos (11.9 million viewers).

So what does the end of Game of Thrones mean for marketers? And what does it mean for viewers?

Game of Thrones is not only the most-watched series on HBO, it is also the most-pirated TV series of all time¹. And while the show hasn’t been ad-supported, its cultural impact has been felt across the advertising ecosystem, as seen most notably in Bud Light’s ambitious crossover Super Bowl ad, which saw the ‘Bud Light Knight’ facing a gruesome demise at the hands of ‘The Mountain.’ For advertisers, the game becomes finding ways to use the fervent fandom and engagement with the show to create meaningful engagements with their brands, a process that starts with understanding the Game of Thrones audience.

¹Vanity Fair, https://www.vanityfair.com/hollywood/2015/12/game-of-thrones-piracy-2015?verso=true

As Game of Thrones fans, it’s hard for us to pass up an opportunity to combine our passion for data with our excitement for the new season and so we put some of our analytics prowess to work looking at the GoT audience. Using our combined digital and connected TV data, we’ve put together a look at GoT viewers: who’s binge-watching the series to catch up for the new season, and where will they go after the show is gone?

TV VIEWING BEHAVIOR

Using data from MiQ’s integration with Vizio, we were able to identify over 140 thousand US households who tuned in to Game of Thrones’ seventh season for live viewing. Of these, more than 90% watched more than one episode live, and nearly one-in-five (17%) went on to watch the entire season live as it premiered.

On average, viewers spent about 28 minutes watching each episode premiere, about the same time spent watching the championship game of this year’s March Madness Tournament. One in six households (16%) watched more than an hour of Game of Thrones on premiere days as they caught up on the previous week’s episode before each new season 7 premiere.

As viewers gear up for season 8, many are rewatching prior seasons to catch up and refresh their memories on prior seasons. More than 70% of the viewers who watched season seven (3+ episodes) watched HBO’s series recap marathon this year. Just under one fourth of marathon recappers (23%) are either new viewers or primarily watched Game of Thrones’ seventh season through alternate channels (such as the HBO GO app or downloaded).

Segmenting the Game of Thrones audience

Using their TV viewing behavior patterns and online interactions, we were able to build a clustering analysis to identify three distinct Game of Thrones audience segments:

Lannisters

Lannisters are the most ardent Game of Thrones followers, actively following Game of Thrones content both on TV and social media platforms (Twitter and Reddit). These viewers watch premieres when they are the first broadcast on TV and also tune in for pre-season and pre-episode recaps. Beyond their aggressive consumption of GoT TV content, they are also active on fan pages and discussion forums, reading and actively contributing to online discussions.

For brands looking to do Game of Thrones tie-ins, fanatics are their target audience. Not only will GoT-centric themes capture their attention, but these consumers will be more likely to create word-of-mouth for promotions and tie-ins that feature creative brand executions around their favorite show. These are also the easiest GoT fans to reach, active and engaging with the show across multiple channels, with more available ad opportunities due to the amount of related content consumed. The majority of opportunities to reach these viewers are on their home or work PCs.

TARGETING LANNISTERS:

Lannisters skew younger, female, and affluent. 52% are under the age of 35, but 53% have an annual household income of $100k or more. They’re avid consumers of technology and celebrity content and tend to have an affinity for contemporary high-end fashion labels. They’re also avid consumers of sports content on TV.

Targaryens

In terms of their TV content consumption, Targaryens are similar to Lannisters, except that they are less likely to tune in to the post-episode highlights/making-of segments on HBO and are usually spectators, rather than participants, when fan discussions happen in online forums. Targaryens are regular series viewers and have been eagerly anticipating the final season for more than a year. They’re also may be the most likely to pirate episodes if there is a leak in the middle of a season.

Because they engage less (and more silently) with GoT content online, there are fewer contextual opportunities to reach Targaryens.  On average, there are 13 digital opportunities per day, across multiple devices and platforms, which means you need a more behavioral-focused, cross-device strategy when trying to reach these viewers.

TARGETING TARGARYENS

Targaryens are equally likely to be male or female and tend to have household incomes in the $75k to $100k range. They skew a little older than Lannisters and over-index to technology and sports content online. They’re more likely to shop at discount fashion retailers and watch a lot of syndicated broadcast content (Friends, NCIS) and travel and leisure shows on TV.

Starks

The viewer group who are most likely to fall off during the final season are Starks, who are more casual fans of the series. Starks make up 43% of the total GoT audience but account for only 12% of TV viewing time. They tend to be less inquisitive and invested in the series and are more likely to catch up on the series online the day after a new episode airs, or to watch a repeat viewing after the premiere initially airs. Only 3% of Starks visit or participate in online GoT discussions.

Starks tend to be mobile-first when it comes to internet usage, and average about 17 impression opportunities per day, eight of which are on mobile devices. As the largest individual segment, they still represent a major opportunity for tie-in advertising, but lower levels of engagement will likely present more of a challenge to create experiences that capture their attention.

TARGETING STARKS

Starks skew significantly more male than female (about 1.8 to 1), and young – 58% are under the age of 35. They tend to have lower household incomes and a moderate spending capacity. They over-index to online news content and are more likely to watch news content on TV.

PREDICTING THE FATE OF POPULAR GOT CHARACTERS:

Look, we’re nerds. And as proud nerds, we’re always looking to combine our love of data science with our love of pop culture. So we put our stats skills to work and created a model to identify the survival probability of the series’ major characters during Season 8 based on 25 factors like screen time, page ranks, character traits, house and established allegiances. What to know who the model thinks will win the Game of Thrones? Read on.

  1. Cersei Lannister – Possibly the most-hated surviving character in the series, Cersei is 1.8 times more likely to die by the end of this season than an average character in Game of Thrones. Though the manner in which Cersei goes is fairly uncertain, we think it’s a pretty safe bet to assume that the old witch’s prophecy will come true and Cersei won’t end up on the iron throne at the series’ end.

Death Odds: 15:2

  1. Sansa Stark – It is impressive how Sansa has managed to live on and fight against all odds. Most pundits and Twitter polls rolled out before Season 7 of GoT had predicted Sansa to die during Season 7. But not only did Sansa survive the Battle of the Bastards, she also managed to wipe Littlefinger off the board in Season 7 in a shocking power play. But our survival analysis predicts that Sansa likely won’t make it through Season 7, behind only Cersei as the most likely to bite it.

Death Odds:  9:2

3.Sandor Clegane, “The Hound” – After an impressive run and multiple brushes with death, we think that Sandor’s redemptive arc is likely to end in his death, either to the night King or his brother (during the highly-anticipated Clegane bowl). Surprisingly, our model likes his undead brother, Gregor “The Mountain” Clegane’s um, “survival” odds much more this season – it predicts that The Mountain has a 1.3x higher chance of making it through the season.

Death Odds: 11:3

  1. Bran Stark – Now that he’s delivered the message about Jon’s heritage, there’s not much left for Bran to do in the narrative and so the new Three-Eyed Raven might meet his end before the series does. Bran ranks second amongst the Snow siblings behind Sansa in terms of their probability of death. Bran is twice as likely to die as Sansa and 4 times more likely than Arya.

Death Odds: 7:4

  1. Jaime Lannister – Nikolaj Coster-Waldau’s conflicted antihero has come a long way since pushing young Bran Stark out of a window in the series premiere. But in season 8 we suspect that Jaime may finally find his moment of glory in the battle of the living and the dead. The chance for a heroic last stand pushes the one-handed golden knight into the very top of the model’s top-5 list of characters likely to die, though we really hope that he fulfills the ‘Valonqar’ prophecy before his time is done.

Death Odds: 3:2

Want to know more about Game of Thrones viewers, or how to connect TV and digital data to drive business outcomes for your brand? Let’s talk.