Kontagent v2.2 released today introduces a major graphical feature update that many of our customers have been asking for! Offsets and moving average trend lines have just been added that support trending in all time series-based graphical views in the platform.

Historical Time Comparisons

One key aspect when comparing two or more sets of data is to ensure that they are apples-to-apples comparisons. Historical time comparisons allow users to overlay data sets with comparable date ranges. For example, lots of our customers look at week-on-week effects on cohorts with new feature or game mechanic testing. In the case of revenue trending, it allows executives to see how revenue is lining up for a particular day of the week or month and whether momentum is in-line with certain targets set based on a comparable past period.

Overlay multiple offset data lines over the underlying data to provide a good view of current performance versus a historical lookback period. You can see a good example of how offsets are used in the screenshot below, and here in our live demo.

Moving Averages

Anyone who trades stocks is familiar with moving averages and why the view is so critical to trend analysis. If you have a lot of volatility in your data, sometimes the trend is harder to follow. By using a moving average, these spikes and dips in data smooth out. The further back you go with a moving average, the longer it will take for the trendline to react to large movements in your underlying data. This lag effect is a reflection of how sensitive the trendline is to the underlying data. Therefore, the lookback setting for the moving average must be set according to the trend being analyzed. A key benefit in trending analysis is setting benchmarks and targets based on current momentum. It can serve as a proxy into how an aspect of a game will be performing. For more information on moving averages, click here.

Tags: Facebook Game Analytics, Moving Averages, Offsets, Social Analytics, Social Game Analytics, Social Graphs, Trendlines