Closed Bug 726027 Opened 13 years ago Closed 13 years ago

Add legend to KPI dashboard

Categories

(support.mozilla.org :: General, defect, P3)

defect

Tracking

(Not tracked)

RESOLVED FIXED
2012.12

People

(Reporter: atopal, Assigned: rehandalal+mozilla)

References

Details

(Whiteboard: u=contributor c=kpidash p=1)

Each graph should have a legend going with it. 1. Number of active contributors to the English KB Definition: The number of people who have contributed to the KB in the last 30 days. Definition of contribution: edited 1+ article or has reviewed 1+ article 2. Number of active forum contributors Definition: The number of people who have contributed to the forums in the last 30 days. Definition of contribution: wrote 10+ posts 3. Number of active contributors to the non-English KB Definition: The number of people who have contributed to non English KB versions in the last 30 days. Definition of contribution: edited 1+ non English article or has reviewed 1+ non English article. 4. Clickthrough rate: Definition: Percentage of people searching who click on an item on the search results page. 5. Helpfulness of KB articles and Forum Definition: Ratio of “yes”/total votes for KB articles, forum threads 6. Questions responded to within 72 hours Definition: Percentage of questions that get an action within 72 hours 7. Questions solved Definition: Percentage of threads that have answer marked as a solution
Priority: -- → P3
Whiteboard: u=contributor c=kpidash s=2012.4 p= → u=contributor c=kpidash s=2012.4 p=1
Assignee: nobody → rrosario
Whiteboard: u=contributor c=kpidash s=2012.4 p=1 → u=contributor c=kpidash s=2012.5 p=1
Regarding 3: Do the edits need to be approved? If the edit is deferred is still there? What if it's unreviewed?
(In reply to Ibai from comment #1) > Regarding 3: Do the edits need to be approved? If the edit is deferred is > still there? What if it's unreviewed? It is pretty simple right now: users that created a revision in that month.
Ricky, is it simple to remove "deferred" from the list? In other words, this KPI loses a lot of its meaning if we don't do that because a reasonable percentage of "editors" are spammers or trolls. Those edits are usually deferred or deleted. Thanks in advance.
I'd rather not do that. Edits by trolls and spammers should be deleted. We often defer edits not because they are spam, but because we want to improve them before they are approved, those contributions are still useful, and they would disappear completely from the metrics, if we pooled them together with spammers and trolls.
I agree with Kadir. We actively delete spam edits. My request around active KB contributors (English and non English) is to not count someone twice, i.e. if I edit and review I just count as one contributor.
(In reply to Verdi from comment #5) > My request around active KB contributors (English and non English) is to not count someone twice We do only count them once.
Bumping to 2012.6.
Whiteboard: u=contributor c=kpidash s=2012.5 p=1 → u=contributor c=kpidash s=2012.6 p=1
Sprint based milestone :-)
Target Milestone: --- → 2012.6
Whiteboard: u=contributor c=kpidash s=2012.6 p=1 → u=contributor c=kpidash p=1
Assignee: rrosario → nobody
Target Milestone: 2012.6 → 2012.7
Target Milestone: 2012.7 → 2012.8
Target Milestone: 2012.8 → 2012.9
Target Milestone: 2012.9 → 2012.10
Target Milestone: 2012.10 → 2012.11
Assignee: nobody → rdalal
I know there is progress here but there isn't a pull request yet, so I am going to go ahead and push it to the next sprint.
Target Milestone: 2012.11 → 2012.12
For L10n, from https://github.com/mozilla/kitsune/blob/4f21e8410773baed49e53122f9469190de19dd16/apps/kpi/cron.py L10n Definition: L10n coverage is a measure of the amount of translations of the English top 50 articles that are up to date, weighted by the number of visits for each locale.
Landed: https://github.com/mozilla/kitsune/commit/242568ba8b2638c240334ea76d71938466fa9ab0 Pushed to production by Rehan at 2012-06-25 18:54:00 EDT
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.