After a couple of years in the Prediction Market business we have seen many projects with clients from various industries. Each industry and each client is unique but more often than not the factors for successful Prediction Market implementations are similar across clients and industries.
We’d like to share our learnings to give you an inside from Prediction Market professionals on the critical success factors of PMs:
- Is there a steady stream of questions, e.g. weekly/monthly sales forecasts for various products/services, or regular innovation weeks etc.? If not, you do not need a regular Prediction Market system and you may just book it on demand.
- Is there a real-world outcome for each question which is known no later than 3 months (better <1 month) after the end of the trading period of that question? If not apply special incentive systems (contact us for more on this).
- Is there a real benefit from improved forecasts and is this benefit at least 10x the effort of deploying a Prediction Market? If not don‘t try it.
- Are forecasts taken seriously or does the CEO override them anyways? If the latter is true, there is no need for a Prediction Market (or any forecasting, for that matter).
- Can a traditional survey solve the problem better? If yes, don’t use a Prediction Market.
- Are there enough participants? This depends on the amount of information you want to collect. Here are some rules of thumb: 50 participants for 30 forecasts, 100 for 60, 200 for 80.
- Last but not least: Is there a clear commitment for budget, because Prediction Markets do not come for free
We hope that these rules of thumb will give you a clearer view on the critical success factors for deploying Prediction Markets to your organizations. Prediction Markets have excellent forecasting accuracy but if the above conditions are not met even the best forecast could be useless.
