What happens to 85,000 people the day after the election
Sweden voted on 13 September. Two of our clients ran quizzes into that election, and both did well.
One part ran a television-marketed personality type quiz campaign from June to polling day. 95,000 plays by 85,000 unique players, alongside several smaller quizzes asking voters about specific concerns.
Dagen, Sweden’s Christian daily, ran a valkompass in the months before the vote. 51,000 plays by 47,000 unique players.
Those numbers are not the interesting part. The interesting part is what those people were doing when they arrived.
The most willing audience you will ever meet
Engagement design is usually described as the work of getting attention. That is the wrong half of the problem. Attention is abundant and cheap. What is rare is willingness, the state in which somebody hands over information about themselves because they want something from you right now and the exchange feels fair.
A voter working out how to vote is in that state almost uniquely. They are actively looking for insight, they have not yet decided, and they will answer twenty questions about their views on tax and defence and schooling without being coaxed, because the answer they get back is worth the effort. Nobody has to be persuaded to take a valkompass. They arrive already wanting it.
You cannot manufacture that condition. You can only turn up when it exists.
Which is why the day after the vote is the number that should worry a publisher, not the eighty five thousand.
The audience is fleeting by design, and that is fine
The obvious objection to an election campaign is that the audience evaporates. The decision gets made, the need disappears, and the traffic that took three months to build is gone in a week.
That is true and it is not a flaw. The campaign did exactly what it was built for. A valkompass has a terminal answer, and a person who returns to it repeatedly has not been served well, they have been left uncertain.
The flaw is what almost everybody does next, which is nothing. The campaign is scoped to the event, the event ends, the report gets written, and eighty five thousand people who were, for one brief period, willing to tell a stranger what they actually think, are allowed to go home.
The quiz was never the product. It was the lead magnet. The product is whatever you offer them on the fourteenth of September.
What a transition actually requires
Three things, and all of them have to exist before the event ends, because after it ends you have lost the reason to write to anyone.
The next thing has to be ready. Not conceived, ready. If the follow-up is still a good intention on 13 September, it will still be a good intention in November. Build it while the campaign is live and you are still interesting.
It has to be about them, not about you. A newsletter sign-up is about you. “Here is how your answers compared with everyone else in your kommun” is about them, and it is made entirely from data you already hold. The most reusable asset in an election quiz is the aggregate, because everybody who took it wants to know where they landed relative to other people.
It has to ask for less than the first thing did. They gave you twenty answers about their politics. The follow-up should cost them thirty seconds, not another twenty questions. Willingness decays, and the ask has to decay with it.
The parties’ smaller quizzes on specific concerns are, in effect, this mechanism running during the campaign rather than after it. Somebody who has already completed the main quiz is warm enough to answer a short one about a single issue, and each of those answers is more useful than the first set, because by then you know who is asking.
Dagen’s two elections, and why the comparison matters
The clearest lesson this autumn came from one publisher compared with itself.
Dagen’s quiz for the church election last year was played 173,000 times by 139,000 people. Their general election Valkompas quiz this year, 51,000 plays by 47,000 people. Same publisher, same format, roughly a third of the participation.
Dagen is a Christian publication with a Christian readership, and a church election is the subject that belongs to that readership more specifically than almost anything else. A general election belonged to everyone in Sweden, which is another way of saying it belonged to nobody in particular.
But the number is not the point. The point is that those two campaigns had different conversion problems and should never have been given the same follow-up.
The church election quiz was largely serving people who were already Dagen readers, or close to it. The transition there is shallow, a matter of deepening an existing relationship.
The general election quiz reached a much wider and much stranger audience, most of whom had no prior relationship with a Christian newspaper at all. That is a harder conversion and a more valuable one, because those are genuinely new people. It needs a different offer, and the offer cannot be “read our paper”, because they did not come for the paper.
Same format, same platform, two entirely different jobs. If I could give a publisher one diagnostic it would be that one: before you design the follow-up, work out whether this campaign brought you your own people or other people’s.
What I would measure instead
Not plays. Plays tell you the promotion worked.
How many left something behind. An email, a preference, a kommun, an opinion. That is the willingness converted into an asset.
How many came back for something that was not the quiz. This is the only number that tells you whether a transition happened, and it is the one nobody reports because it requires you to have built a second thing.
How long the tail lasted. If traffic is at zero seven days after the event, the campaign was a moment. If it is still trickling in October, something was constructed.
I build these things for a living and the platform these campaigns ran on is ours, so treat my bias as given. But the argument would hold if somebody else had built them. A fleeting audience at peak willingness is the most valuable thing a publisher gets all year, and the standard practice is to celebrate the reach number and let them go.
