When drilling down into the KPIs below the timeline in the Overview page, you may experience that the reported media-generated effect differs slightly between what is reported in the media decomposition and the business insights decomposition, for the same selected period:
Media Insights
Business Insights
The reason for this discrepancy is due to the fact, that the media-generated effect is reported in two different ways, for media decomposition and total sales decomposition, respectively:
The media insights decomposition uses the effect reporting method we call "aggregated effect", which means, that all generated effects from the day of the investment, including the decay effects (the effect from that investment occurring on the following days of the investment), are reported on that day. Let's take an example: you are campaigning for the sales of a dairy product. On February 21st you are investing 30.000 in media - among other days of different investment levels. Some of the actual sales generated from that investment will occur on February 21st. But some of the sales will occur on the following days (not all consumers viewing the campaign exposures, will run out and buy your product on that same day. Some will do it on the 22nd, and some on the 23rd, and so forth). All of those sales occurring on the day of the investment as well as the following days, are all aggregated and reported as a total, on the day of the investment. In this case, the 21st of February. As such, the effect reporting in the media decomposition, is an analytical way of reporting effect. It's designed like this, so that you can define a specific investment period, and then understand the total effects coming from that investment, such as a specific campaign.
The business insights decomposition uses a different reporting method; we call this "observed effect". The observed effect is reported as it happens in reality. Meaning the opposite of the reporting used in the media decomposition. Here, the effect of all dimensions, including paid media, does not aggregate the effect of the following days and report the total effect on a specific day. Instead, only the actual or real effect is reported on each day, meaning that some of the effects of an investment on a specific day are reported on the following days of that investment.
This means, that for the media effect reported in the total sales decomposition, some effect from the investments made in the period before the selected one, will "flow" into the chosen period. And that some of the effects from the investments of the selected period, will "flow" into the period following the selected one.
The way to read these ways of reporting media effects is then; the media decomposition reports the total effect from the media investments of the selected period, even though they will not occur precisely as reported. While the total sales decomposition will report the effects of the media investments of the selected period, as they actually happen during this period.