Leadership might be interested in a regular (hopefully automated) QC process in the future
Do we need to include the actual SR?
Set evaluation parameters
Data sources
311 TPW CSR Power BI dataflow
311 TPW activities report (to replace manually pulling activity data from the Check Service Status public 3-1-1 site)
Timeframe
Last 30 days - that's all that's available via the public site
Stats for summary
Percent completed on time
Average time open
If an accurate update is included to the resident or not
External/supplemental data
Do we need the associated work order?
Do we need flex notes?
Identify SR types "descriptions" to include
Identify all divisions that respond to CSRs
With a manual pull of the Activities, we need a cutoff or this will keep going on and on and on - what is the highest priority / most important to look at now?
Jim originally asked for 10 random CSRs from each division that responds to CSRs
Richard really cares about the ones that are in the COA Top 20 list (TPW is responsible for 3 of these)
But with the Socrata 311 Activities dataset, we can easily tap into them all so why not - definitely provides a better picture of the current state
A filter could bring up 10 random SRs like Jim originally asked for
Executives will likely want a summary, maybe including
Percent accurate
Percent complete
Number of fields completed
There are more work groups than there are divisions and some work groups are quite different from one another, so we'd get a better picture if we break it up by work group
Richard really cares about the ones that are in the COA top 20 list (TPW is responsible for ~5-6 of those)
DTS & OPM should coordinate when communicating with stakeholders, like division CSR/data contacts
In the final deliverable, we should highlight updates that are not available to the public via the 311 site Check Service Status
In our sample calculations, maybe we exclude emergency SR types
to show what the public has access to
to keep our scope smaller
Emergency SR types (like Traffic Sign Emergency for a downed stop sign) are not accessible via the public 311 site
Citizen who made the request for a traffic sign emergency doesn't have access to any updates via this site
Do they get email update?
Who tags it as traffic sign emergency, when, and how?
It's not in the 311 request options online/app
What if you call in?
From Christina: Something is tagged as an emergency through the set of questions 3-1-1 asks while intaking the SR. I don't think citizens can input an emergency SR through the mobile app, but it will automatically get sent as an Emergency Sign SR if they flag a damaged sign that's considered an emergency, which are a few regulatory signs: stop signs, yield signs, one ways and wrong way I think!
SR types with a SLA timeframe of more 30 days are not accessible via the public 311 site
If something is open more than 30 days (which is within the SLA timeframe for some types), the public can't check the status via the public 311 site
Can they check it any other way?
PDF of SLAs (Service Level Agreements) is saved here - TPW on p 32-48
All of our activities get added with "COMPLETED" but there's another mechanism to CLOSE it - Christina will check with John
Consider including a PDF from Knack for each SR in the sample - so we can compare 311 vs Knack
For Road Markings/Striping - most of the sample shows no Detail Update, but we looked at one and it was actually reassigned to another SR type.
To build the dashboard, Susanne is using PBI dataflow + manually pulling activities from the Check Service Status motorola site to a spreadsheet
We should divvy up the workload so Susanne doesn't have to manually pull all those herself - we can each add to the spreadsheet
Christina will send 3-1-1 (Carrie, Hichame, Patrick) a request to create an automated TPW 311 Activities report + ask
Which field is used to populate the public website?
Are there activities that don't get reported?
Maybe we add calculations for
How long it's open
How many days late
% of CSRs that have 3-1-1 feedback activity - certain ones may be inevitable, but maybe others are generally confusing and cause things to fall through the cracks
To limit the scope, OPM identified 11 SR types that showed up in the City Top 20 and Council District Top 20 reports
But these don't include an SR type from each division - we're missing AMD, TED, Utilities, ATSD
Let's look at more/different parameters to get to our defined scope
Include one from each division
Look at count per FY
Maybe top 10 requests by volume
Maybe look at what TPW is responsible for in the top 20 COA list by council district to include more representation across the city
Should we include the top SR by volume for each work group (to capture something from each work group even if they're not in the top 20 COA list)
There seems to be a perception among city leadership & in the public that SRs are immediately closed out
Richard got email from RG about potholes
A larger understanding of what's done / how we respond is needed
Based on the SR type, this is the automatic message sent, it's open for on average this long, and has this many activities on average
We're just looking at SRs, but there's also
Citizen Action Forms (CAFs) - like a CSR via a council member
CUIR
Reminder of request
Jim wants to assess how the public receives communication
What does the public see? What responses are publicly visible?
Timeliness is hugely important to Jim
Visuals
Visuals for 10 SRs from the last 30 days won't really be meaningful
Especially ones that have a longer SLA than 30 days
Since we have the Activities dataset, let's include all the SRs per type - we have the data available, it will give a better picture of what's happening
Add these visuals
Average / median time per SR
Don't need these visuals
% Completed On Time - already on the SR Response dashboard
Method Received - already on the SR Response dashboard
Recommendations/notes might should include
Define terms
CSR or SR
Highlight how the activities are only available online publicly for 30 days
What about SRs with SLAs longer than 30 days
Emergency SRs aren't listed
At any given moment, what's the percentage of open SRs that aren't available to check online?
Define deliverable
Set evaluation parameters
Data sources
Timeframe
Stats for summary
External/supplemental data
Identify SR types "descriptions" to include