For each of the past 90 days, how many times did users reset their password?
This is a basic count of user activity: measuring a specific user activity over a specific time range. This sort of query is often the starting point for more in-depth explorations of your data.
Building the query
Although we're measuring activity over time, we want to see the number of users performing a specific action grouped by day. So we'll use a Table view for this query.
Select the time range (90 days ago to now), and Count Events as a measure (because we're looking at events, not actors).
Next, group by the automatically generated
__day__ column. Interana provides columns that let you group by mine, hour, day, and week.
Instead of showing us a count of all events over this period, we want to look specifically at password reset events. In the Wikipedia data set, this is represented in the
log_action column as a
reset event, so we'll filter by that.
Tip: adjusting the order of the results
By default, Interana orders the chart by the measure (in this case, Count Events). The chart will be ordered with the largest number of events at the top of the table.
Use the chart controls to can change this to order by the days:
Interana orders the chart with the most recent date with activity at the top of the chart. Note that if there were no reset password actions during any day, that day will not be included in the chart.