Both the Planticipate app and its workshop equivalnt seek to predict, and explain different categories of people’s plan preferences.
Earlier prototypes correctly predicted people’s top-ranked plan at least 70% of the time, and the current versions should improve upon this.
They constitute an extremely rare, if not entirely unprecedented application of scientific rigor to the intensely humanistic process of people-oriented policy making.
Every app user is asked to consider some “situation”, comprising a client, a goal and two or more alternative plans for achieving that goal. The situation can be either one of the examples that come with the app, or a new situation entered by the app user.
App users then score the situation's plans, both for twelve, key plan-evaluation criteria and for overall desirability, and their judgments are automatically sent to the cloud where they are closely protected and NEVER revealed to any third party (they are of dubious commercial/political value anyway).
This enables our cloud-based dataset to slowly accumulate knowledge about relationships between plans’ criterion scores and their overall desirability and, since every app user has also supplied their demographic details, such relationships are updated for every demographic category to which the user belongs.
It follows that in ANY subsequent situation, provided that you first score alternative plans on the twelve criteria, the apps will forecast each plan’s overall desirability according to up to 93 different demographic groups. And the more people that use our apps the more accurate they should become - they are self-improving.
Moreover, our apps always graphically locate their plan-desirability forecasts within margins of error. This makes it obvious which forecasts are statistically significantly different from each other – if two plans' error margins do not overlap they are not forecast to be significantly different from one another in terms of overall desirability.
The apps also calculate Bayesian probabilities that their plan predictions are correct – they are self-monitoring.
In addition, our apps always suggest why their forecasts turned out the way that they did. For this they use our innovative 'face charts' method which is demonstrably superior for clarifying multi-criteria data. A face chart showing the priority placed upon each criterion by the relevant category of people is set alongside face charts for every plan indicating how they each scored on the criteria. Our apps are, again graphically, self-explanatory.
To see whether or not Planticipate's error margins are shrinking and/or its accuracy is increasing, you should intermittently run it
More generally, if you think we could improve the empathy of your forward planning you should let us know (e-mail: firstname.lastname@example.org)