A made-up Pinterest, built from public figures.
Every person is simulated. Country and state figures are allocations of Pinterest's three reported regional totals.
- Stored batch
- Loading
What if Pinterest changed one thing?
Follow one what-if through a made-up Pinterest built from public figures. Every result is a simulated range, not a forecast.
Only three totals are reported.
Pinterest reports monthly users for three regions. The dots regroup into those totals. Anything finer is the model's own allocation.
2
Pick a what-if.
Choose a change to follow. Features show decisions and outside events show visits. Each group shares one scale.
Reading the stored world. Nothing is running.
What it is, and what it rests on.
Loading the scenario.
- Loading
- The evidence, the assumed change and the behaviour it touches will appear here.
Real things like this:
Read the assumptions
The result, and how to read it.
A world is one full set of assumptions. What follows is a range under stated assumptions, not a forecast.
Loading the stored result.
5
Watch it happen.
One simulated world runs twice with the same people and the same random seeds, once with the change. The chart shows visits over the trailing seven days.
Nothing is running.
Two versions. The same starting people.Press Play one world to watch visits with and without your chosen change.
6
Where it lands, and who it happens to.
Look closer at a place, or follow one simulated person. These two pictures show where to go. They carry no result data.
7
How much to believe it.
Matching public figures constrains a baseline. It does not validate simulated people or feature effects.
8
What to do with it.
Use the result to choose what to measure first.