Different Methodological Approaches

propella.ai is currently in the process of experimenting with different ways to draw insights from our Worker Activity data. While the underlying data source informing the insights is the same, some users may notice differences between what is currently shown in propella.office and newer experimental insights. In order to rapidly prototype and provide these new insights to the commercial property market, we have not yet standardised on one single format to visualise on. Discrepancies between different visualisations are all known and in every case are the result of aggregation and simple moving averages. We have gone to great lengths to ensure the data is consistent and can be identically transformed between insight types.

For example, the propella.office Worker Activity in market insights is aggregated by the week and has a 3-week moving average applied to it. While our Daily Return to Work estimates is simply the rawest estimate we can create of return to work. Consequently the numbers will be different but the underlying data is the same. It is possible to transform the Daily Return to Work estimates into propella.office's Worker Activity exactly, and we have done so.