Expand my Community achievements bar.

Join us for the next Community Q&A Coffee Break on Tuesday April 23, 2024 with Eric Matisoff, Principal Evangelist, Analytics & Data Science, who will join us to discuss all the big news and announcements from Summit 2024!

Report Statistics


Level 2


When viewing Page Views, Visits, Visitors and other core reports there is no granular information available about the report aside from the core data e.g. Numbers over time.  Additional embedded statistical data would be of interest to many who consume the reports.  As an example: hourly, daily, weekly, … average, range, standard deviation, median.   Presenting this information within the report would provide additional insight to the consumer of the report.

1 Comment


Level 1


I absolutely agree with this.  I pull this data down into Excel so that I can calculate whether or not a data point falls outside of 95% confidence interval ranges.  It's nice to see, for example, that traffic was up on Monday, but is that spike statistically significant for a Monday when compared to all Mondays over the trailing last six weeks?  It's the difference between data and actionable information!


I also run linear regressions on my trended keyword data to determine if the slope for certain strategic keywords exceeds a certain angle.  If, for example, the per period rise of the linear regression line for a particular keyword exceeds 5% of the average for the trailing six week period, then I know there's something going on.  This is an invaluable tool when you're tasked with keeping a "close eye" on nearly a thousand strategic keywords at one time.


Combining the confidence interval per period with slope helps make trends JUMP OUT at you!  I'm able to zero in on creepers and spikers alike within seconds and I can even identify product buzz ahead of a launch.