How Lupe Used Lead Scoring to Automate Game of Thrones Predictions

How Lupe Used Lead Scoring to Automate Game of Thrones Predictions

If you followed Game of Thrones during its run, every episode came with a key question.

“Who is going to die?”

Turns out – a lot of people. But everyone had a different opinion about who (and how).

That’s how the Game of Thrones Death Predictions game started – as a way to keep track of death predictions and declare a winner at the end of each season.

There’s a lot that goes into running a death predictions game. You need to…

  • Manage a group of unwieldy Game of Thrones Fans
  • Keep track of multiple predictions for the show’s many characters
  • Send a ton of updates on the status of everyone’s guesses

And from this, the ActiveCampaign Success team’s Lupe Puga saw an opportunity – to use our very own platform’s lead scoring and automated emails.

Email is coming.

How to keep track of everything (without using paper)

“For the season premiere, people online were posting pictures of a death predictions game they created to guess what characters would live and die. They had their friends fill out and hand in this piece of paper.

I loved the idea so much, but thought that I would save some trees and optimize the experience by using ActiveCampaign.”

What do you do to keep everything organized and episode update emails sending on schedule?

Use automation!

Here’s how Lupe automated the manual work of running the game. She:

  1. Created a form to collect predictions
  2. Used 4 campaigns to send out updates
  3. Automated campaign sends with lead scoring, predictive sending, and tags

The ActiveCampaign form she created let people submit their live-or-die predictions. Submitting the form triggered the first welcome email.

5ofkcuh4g screenshot2019 06 17at9.18.56amThe form to fill out live-or-die predictions.

Once people filled out the form, they were added to a Game of Thrones contact list and the automation began.

Lupe’s automation had four triggered email campaigns:

  1. A welcome email. Everyone on the list got this email before the first episode premiere. This email had the instructions and a link to every contact’s predictions from the form.
  2. A reminder email. A segmented campaign for those who did not take the required steps for participating in the predictions. She used Tags to mark people who were already submitted to avoid sending this campaign to them.
  3. A progress email. A progress email sent after the second-to-last episode, with updates on everyone’s prediction progress.
  4. A final update email. Sent after the final episode, with the last updates on who lived and died.

3kbbr8xd3 image2019 06 25at8.27.02amThe welcome email that all contacts on the Game of Thrones list received.

The show aired every Sunday night. Should Lupe send the emails on Monday morning? Or when they were most likely to be opened?

After all, the truth about Cersei’s fate is not an email anyone wanted to miss.

ActiveCampaign’s predictive sending lets you check a box to send your emails the right time for each recipient (when they’re most likely to engage).

For the last two email campaigns, Lupe used predictive sending. Based on predictive sending’s analysis, emails arrived in people’s inboxes on Monday evening.

The result?

Lupe saw her highest open rates in the last two emails (sent using the predictive sending).

When you play the Game of Thrones [prediction game], you win or you die lose…based on lead scores

“Before the season premiere, I created a form that was due an hour before showtime. Then I used Contact Scoring to match up answers and award points.”

She had the form, she had the campaign ideas, she had the contacts. But there was one more necessary component.


Each update email would include prediction progress information – in other words, who was winning or losing with their predictions.

“If you do it by paper, how do you compile all those answers? Who’s going to count everything? I don’t have the time for that. I still had to put some effort into the lead scoring, but that was more fun and efficient than counting by hand.”

After each episode, Lupe used lead scoring to create points for major events (and if you correctly guessed whether characters lived or died).

5qtztcu screenshot2019 06 17at9.16.31amLead scoring based on predicted character deaths.

Every character was assigned two leads scores – 1 point given if the guess was correct, and 1 point taken away if incorrect.

For example, if someone guessed that Arya died and became a White Walker, that was negative 2 points. (Come on, they never would have done that. Silly prediction).

“I’ve learned enough from this first experience that I think I could make it even more seamless next time”

Lupe’s first experience using ActiveCampaign for this project was inspired by listening to many customers describing their own good experiences. Lead scoring turned out to work better than she anticipated, and the automations were a great time saver (especially after watching on a Sunday night).

But she also learned some things that will make the next automated show prediction game even better than her first successful run.

Her top takeaways are:

  • Choosing double vs. single opt-in. Because this list wouldn’t keep getting emails after the show’s end, Lupe wished she had made the form single opt-in instead of double. A double opt-in wasn’t necessary because this wasn’t a recurring subscription.
  • Submitting multiple forms. Everyone had the chance to submit a max of two prediction forms. Since every contact was registered to one email address, a second email address was needed to submit the extra form. That meant the list had a few double contacts who got more than one update email.

Game of Thrones probably won’t be the last television prediction game that Lupe automates. After all, World Cup Soccer and March Madness are still on the table…What show should she automate next? Suggestions are welcome.

Never miss an update