Most of the games in the library are over in a minute. Predict the Outcome is the exception. It takes a second to play and days to finish, because the answer doesn’t exist yet when people play. It’s built for moments your customers are already watching: a cup final, an awards night, a product reveal, the last episode of a show.
How a player sees it
The page asks one question, with between 2 and 20 answers. The player picks one and locks it in. There’s no score and no instant win screen. They’re told they’re in, and that the draw happens after the deadline. Only the first pick in a session counts, so nobody can hedge by answering twice in the same play. Give the campaign a total limit of one play per person, too, or someone could return and pick every other answer.
That short, quiet moment is the point. The game itself is light. The excitement comes from the event, and from checking back afterwards.
Two clocks, on purpose
A prediction campaign has two times that matter, and they’re separate.
- When entries close. This is the campaign’s live-until time. After it, nobody can play.
- When the draw runs. Either you run it by hand, or you set a “Draw at” time and it runs on its own.
Keeping them apart lets you match the real event. Close entries at kick-off, when the answer is still open. Draw on Monday morning, once the result is settled and someone on your team has confirmed it.
Marking the right answer
After the event, someone on your team marks the correct answer in the dashboard. It’s a separate permission, so you can decide exactly who is allowed to do it.
Mistakes happen, so the answer can be changed, right up until a draw has used it. After that it’s locked. Winners have already been drawn against that answer, and changing it would quietly change who should have won.
The draw
When the draw runs, it only looks at eligible entries that picked the right answer. It draws the number of winners you set, one by default, using the same commit-reveal process as every other Keliplay draw. Your team can re-check it afterwards through the API.
The results tab shows how many eligible entries backed each answer, which is also a small piece of market research in its own right.
One more thing to watch
If a draw runs before anyone has marked the right answer, it draws from everyone who entered, whichever answer they picked. The dashboard warns you when a prediction campaign has no answer set. If you schedule the draw, set yourself a reminder to mark the answer first.
Where it works well
Prediction games suit events with a clear result and a crowd that already cares. Sport is the obvious one. It also works for launches (“which colour sells out first?”), seasonal moments and anything with a public vote. What they have in common is a reason to come back and check, and a reason to come back is most of what a campaign is for.