FireCARES weekly reporting

[meilinger]

Need to create a FireCARES ops/department data weekly report that will include the following:

• Has jurisdiction boundary
• Has staffing information
• "Data Reviewed" status for each available set
• Count of new users (weekly)

Additionally, we should include google analytics reports as well w/ comparisons for the prior week and month.

• Pageviews
• Unique pageviews
• Bounce rates
• Most popular pages
• Popular search keywords

Would be nice to be able to meld this with the helpdesk reports (see #461) as well as individual department/community engagements for cause-effect.

@garnertb @cweinschenk any additional important factors to report on welcome!

[cweinschenk]

Love this!

For your first set could be do it like a running count where the criteria has been met with then a + XX for what grew during the week

[meilinger]

@cweinschenk totally makes sense! Yeah, we'll need a mechanism to persist data so that we can do rolling comparisons, but definitely the way to go!

• User account type (FireCARES native vs IMIS vs Helix -- IAFC)
• Possible facebook/instagram/snapchat/twitter tracking of #firecares keyword mentions

[meilinger]

Additionally, we should be tracking # of verified departments (each "verified" dimension) in this report as well.

[garnertb]

Here are some more:

• Count of stations without a department
• Session counts using advanced tools (ie heatmap, eff, etc)
• Session duration
• AVG Query Position (from GA via Search Console)
• CTR (from GA via Search Console)

[cweinschenk]

I will add:

This should be done by population group:

• [x] departments without risk score
• [x] departments without performance score
• [ ] departments with 2016/2017 data

[meilinger]

Grafana dashboard for some of these metrics (starting w/ data inherent to FireCARES/NFIRS).

• Count of stations w/o a department
• Count of departments w/o risk scores (based on "ALL" hazard level)
• Count of departments w/o performance scores (based on "ALL" hazard level)
• Count of departments w/o jurisdictions boundaries
• Count of departments w/ missing population
• Count of departments w/o FDID
• Count of departments w/o any stations
• Count of departments that do not have department administrator-verified data
• Count of departments w/ no staffing information

Green == seeing 0's across the board, make sense as a first-pass @garnertb @cweinschenk ?

The hope is to see downward trends.

[meilinger]

Data curation dashboard in place, links send via Slack.