You can clone with
HTTPS or Subversion.
Many Piwik users log every day, every week, or have some kind of "routine". But many Piwik users also login at random intervals, depending on many factors.
It would be nice to reduce the user frustration and highlight clearly all the things that have changed and are significant, since the last time they logged in.
The widget could say for example:
What's changed since Jan 23rd (last login)?
* New significant search keywords: $K1 (1304 new visits), $K2 (23 new visits), $K3 (11 new visits)
* New significant website referers: ....
* New pages with significant traffic: $BLOG_POST1 (155 views), ...
* Some pages lost significant traffic: $P1 (55 visits, -555%), $P2 ....
* Some keywords used to be strong but have sharply decreased: $K1 (4 visits, -800%)
* Ecommerce sales sharply decreased for these products: $P1 (11 sales down from 155: -93%), ....
Insights based on: daily data, weekly data, monthly data
As a first step of this ticket, we could store the last user login datetime.
This would make user experience more enjoyable as it would reduce time spent looking for insights.
When clicking on an insight eg " Nice Homepage (55 visits, -555%)" we could open Row Evolution for this label to let the user know more about why there is an increase/decrease.
It would be quite important to show the sparklines for each insight. We don't want to show the sparkline by default, so it could be shown for all insights when clicking on a checkbox [ ] Display sparklines, or maybe shown on hover on the insight itself.