FY19Q3 PAWS to Panorama Development by ICPI

[jb3436]

FY19Q3 Schedule for Development and POCs (ICPI-Specific)

Assignments

• First 90 (Jacob) Single OU https://github.com/pepfar-datim/Analytics/issues/2191 Global https://github.com/pepfar-datim/Analytics/issues/2192

• Second 90 (Nicole, Michelle, Aliyah, Imran) Single OU https://github.com/pepfar-datim/Analytics/issues/2193 Global https://github.com/pepfar-datim/Analytics/issues/2194

• Third 90 (Anu, Yaa, Mackenzie, Imran) Single OU https://github.com/pepfar-datim/Analytics/issues/2195 Global https://github.com/pepfar-datim/Analytics/issues/2196

• HHS (Yaa, Travis, Mackenzie) Single OU https://github.com/pepfar-datim/Analytics/issues/2197 Global https://github.com/pepfar-datim/Analytics/issues/2198

• DREAMS (Kimi, Aliyah) Single OU https://github.com/pepfar-datim/Analytics/issues/2199 Global https://github.com/pepfar-datim/Analytics/issues/2200

• Age and Sex Disaggregates (Jacob) Single OU https://github.com/pepfar-datim/Analytics/issues/2205 Global https://github.com/pepfar-datim/Analytics/issues/2206

Timeline

FY19 Q3 Iterative timeline (dates may shift slightly): June 3-14: 2 weeks of requirements gathering & discussions about getting dossier set up June 14: Requirements Submitted June 21: Requirements approved June 17 – July 30: 7 weeks of dossier development specific dossier development and workplan management July 1 (Monday): first systems check in (after first 2 weeks of initial development) July 8 (Monday): second systems check in July 11 (Thursday): third systems check in July 18 (Thursday): fourth systems check in July 25 (Thursday): final check in before handoff for engineering August 1 (Thursday): Pre-Check in before handoff

Instructions

• review existing dossiers for ideas (reveiw them in PAW and see how they are built
• submit requirements document (your stakeholders should be involved in this process)
• get approval and start building your dossier in PAWS or in DEV environment. If you build in PAWS, it will need to be copied over to DEV. (see below)
• Share questions on slack in #analytic workspaces channel - these conversations will be added to an FAQ or "How To" SOP
• check in with systems team regularly via in person meetings that are scheduled or using slack channel and tagging them
• update the GitHUb ticket created in the analytics repo for development of the dossiers - also log any issues and status updates here.

Development Environment Development of your dossiers will be in the PAW DEV environment: https://dev-workspaces.pepfar-panorama.org/MicroStrategy/asp/Main.aspx:

Requirements Documents Saved Here

ICPI>DIV>Projects>Panorama Redesign

Panorama Guidance Documents for Analytics Here

ICPI>DIV>Projects> Panorama Redesign> Instructions

[jb3436]

FYI @jaliasd @achafetz @cnichols3h @HoustonJamie @jla7 @wvp3

[MicSherlock]

The 2nd 90 team is thinking that 1,2, 3 90 dossiers might only be needed at OU and site. We aren’t sure we need a Global, it’s so high level that we think CC dossier might cover it cover needs. Putting in the parking lot…

[jb3436]

Check in With Dossier POCS 6/25

Discussed:

• expectations around timeline and will send out calendar invites for check ins
• expectation to check in regularly with stakeholders throughout the process
• use of GH and slack channel for open communication
• best practice of building in PAW prod and not worrying about dev until later in July

Action Items

• Jasmine to send follow up summary to systems team
• Katie and Jasmine to schedule "showcasing" dossier progress during inbriefs
• Teams to schedule their stakeholder engagement demo sessions now

First Systems Check In 7/1

Roma, Josh, Kimi, Renee, Aliyah, Jeff, Janice Rachel, Jasmine, Travis,

Overview

• Dataset created - a custom dataset that the systems team helps to develop and create
• Dynamic - it's meant to be a start for the types of objects
• Want you to be able to use a customized dataset that you will use for your dossier
• Met with Travis and HSS - have the requirements
• Rachel met with Anu and VL for the requirements
• DREAMS - do you need any new objects - you should have everything
• 2nd 90 has everything it needs at the moment

Datasets

• could we have someone like Imran create datasets for ICPI?
• Don't use SQL - they use the schema
• Will require more training - probably not anytime soon
• Net - new, where did this come from?

Collaboration

• Multiple people working on dossiers?
• Multiple people cannot update a dossier at the same time
• Suggest - make a copy with the dataset, then divide up that copy amongst multiple people and then once you are done with the seconds, the systems team will merge them for you later.
• When we create datasets for you, it lives outside. When you make changes, it saves it as a local copy - and this causes problems. If you want to make modifications, go to the external datasets. Don't modify it inside the dossier. It makes it local and breaks the link, then essentially you are linking to two different datasets.
• Does this include custom/derived metrics or attributes, then it is fine.
• Any derived metric only lives in the dossier, unless you import a dossier into another one
• You cannot copy/paste metrics between

Performance considerations keep the numbers of derived metrics as low as possible

• If it can be created in the system, then please ask for it

Presentations of Dossiers:

DREAMS

• Most of it has been built out, but making improvements to the visuals and tables
• There was a recommendation to build a single and global dossier - is it possible to have global pages and then have a dropdown and a selector to choose the OU
• Preference is for a global dossier
• Because DREAMS only shows DREAMS Ous and DREAMS SNUs, it should be fine to only have a global one and not create a single OU
• Filter already has an OU
• Action: Remove OU selector prompt from DREAMS
• Janice will test performance
• Kimi can try to select all of the Ous to see what the performance looks like
• How to change the HTML containers in DREAMS?
• Changed the visuals in the AGYW PREV chapter
• Presented DREAMS SNUs and DREAMS SNUs
• HTML will go away once it is converted to global only

Question:

• Is it possible to make the tables "condensed" at default
• If you want to make the tables condensed at default, you can use outline mode - like the accordion feature
• Still uses the same amount of space - but hides the data. You would need to adjust the container

HSS

• Still only have HRH - others are working on building out the other HSS indicators
• Plan is for it to be a single OU with site level and the Global one will have the
• Site may only have HRH and lab
• Created an infographic with KPIs -
• Country level - top line numbers,
• Inserted graphics
• HRH needs to be presented in context of other indicators and change over time
• Using the visual to filter the table
• Adjust the color palettes so they are more categorical
• Percent achievement - need to come up with a achievement color palette

2nd 90

• Haven't added anything new since last check in
• Had a few discussions about the requirements
• Jamie, Nicole, Jacob Dee and Aliyah discussed next version
• Jamie helped update the requirements document submitted (ACTION)
• Custom charts may not work when they are upgraded to the new version of microstrategy
• D3 - visualization gallery - people can grab their D3 objects
• There may be no support for the visual
• Not all objects were copied over to dev
• If you bring in cool D3 visuals, then keep in mind that it may not work when MS is updated
• Who will document the visual in the ticket - Dossier - IMRAN can put the visuals in the ticket
• Insert the info about this type of visual

Level selector

• could have it so that if OU is selected then the second selector would then have OU and if you select agency, then in the right the agency would be selected
• Make sure we include which chapter filters to show and not to show
• Add to the ticket what should be included
• Need to do a full QC of the net new
• Will change the bars at the bottom to be KPI so that the yield chart isn't misleading in the way that it looks like it is at 100% when it is on a scale of 2%
• What's the update on the net new map and table? ACTION: Aliyah and Imran

Hyperlink considerations

• There will be code that will interfere with the HTML code
• Bringing the dossiers into an iframe
• There will be JavaScript that interacts with it
• Send the script to Janice and Renee ACTION: IMRAN
• Renee will review the code and make updates
• Email code and update GH ticket

3rd 90

• To avoid some of the issues the 3rd 90 group had, it would be ideal to go over the requirements and review what is in the system
• Then it would be good to give them some ideas and start building something
• In some cases, instead of starting with dynamic dataset - it is better to start with a custom dataset
• Permissions issues prevented them from being able to do what they need to do
• Not sure where the line of what I need to create and what you need to create is
• Initial discussion
• Timeline to get them up and running took longer than anticipated

[jb3436]

2nd Systems Check In with Dossier POCs 7/8

Janice, Rachel, Yaa, Travis, Jamie, Kimi, Joseph, Jasmine, Aliyah, Imran, Shaylee, Anu...missed a few ppl.

• Workplan review

  • Janice walked the group through the excel workplan
  • ACTIONS:
  • Janice to send a copy of the workplan
  • Jasmine to maintain updates

• 1st 90 Dossier Custom dataset created

  • Joseph following the requirements document
  • May import some of the visuals from existing dossiers
  • Janice will help with the importing
  • Joseph is working on this based on Jacob's requirements doc
  • Shaylee to make sure that Joseph has updated version of requirements document from HTS stakeholders. (ACTION: Shaylee)

• Question : - how do you get visuals that already exist in another dossier into your dossier?

  • You copy the whole dossier into your dossier and then delete what you don't want
  • Janice demoed the process
  • If you want to import from PAW prod to PAW dev - let the team know and then they could copy it to DEV and then you could import it into your dossier (ACTION: all POCs)
  • Process
  • File - import dossier, select dossier, use the same Ous as your current OU
  • Run dossier
  • If it is using 2 different datasets it brings both datasets into the dossier
  • You can then move around the pages and delete what you don't want to use
  • This is also how you merge a dossier being built by multiple people into one

• HSS Dossier

  • Status update
  • Review analysis from Ademola presented during the brownbag and use that use case as an example to explain how it will be used in field
  • HSS Dataset is site level - avoid running all Ous
  • Dossier prompts by OU - then the rest of the items are preselected based on the requirements
  • It already has FY filters, indicators, and already has the disaggregates requested as well
  • Currently using the generic metrics -
  • No derived metrics were modified since we are using the same generic metrics
  • There are many custom metrics - the dossier was running slow because the dossier had to calculate the metrics each time,
  • the plan was to convert all of the FXs into built in metrics in the dataset
  • Metrics List
  • Systems team needs the list of which ones are valid/useful
  • Support type/cadre - the PDH team is going to parse it on their end and create a separate element
    • Once that exists it will be a separate attribute - this will take time - waiting on an LOE
    • Creating a new column that parses the disagg option
  • Need to verify which metrics actually need to be created
  • Sent in the requirements document that has the list of calculations. Travis and Yaa to confirm final list of elements to be created and those to be deleted(ACTION: Travis and Yaa)
  • Janice will create metrics and will delete the ones that are not being used (ACTION: Janice)
  • Will follow up in a separate meeting (ACTION: Travis, Yaa, Janice)
  • Ping analytics repo POC about adding Travis (ACTION: Jasmine)

• 2nd 90 Dossier

  • Added additional visuals from requirements document
  • Question:
  • There are a lot of metrics in the global second 90 dossier in dev
  • Should the metrics be recreated in the single OU? Could the dataset be copied? Could it be at site by IM level
  • Response and Suggestion:
    • Derived metrics are not transferrable from one dossier to another
    • Copy the global dossier - and if you need a separate dataset - it could be created for the team
    • Prompt it for OU - then put site in it
  • Add this to the GitHub issue (ACTION: Aliyah)
  • Janice is working on creating a facility level dataset
  • Is the net-new analysis - is it going in for Q3?
  • Maybe - the team is doing what can be done now
  • Should we work on site counts? - it is dependent on the net new analysis
  • HTML follow up script? Is it useable in the HTML box?
  • Renee is running the test - she will follow up
  • Need to create a facility dataset and Aliyah will confirm if everything else will stay the same and she will update the GH ticket (ACTION: Aliyah)
  • Then Janice will create the facility dataset (ACTION: Janice)

• DREAMS Dossier

  • HTML code what should I do with it? Currently unable to edit it
  • Janice will modify the HTML (ACTION: Janice)
  • Will review permissions to edit HTML code (ACTION: Janice)
  • Images - have the dreams ICON as part of the HTML code (ACTION: Janice)
  • Action - add the narratives dataset and the AGYW to DREAMS dossier (ACTION: Janice)
  • Narratives cannot be used in the same dataset as results and targets
  • AGYW narratives dataset
  • Narratives into DATIM dataset

• 3rd 90

  • Got the metrics needed
  • Created GH comments on all of the aditional oustanding elements
  • Need to check permissions to allow for more ability to make modifications without needing to ask (ACTION: Janice)

[jb3436]

3rd Systems Check In 7/11

Room : Katie, Anu, Shaylee, Aliyah, Nicole, Jasmine , Joseph, Travis Phone : Janice, Kimi, Renee, Mackenzie, Erin, Jeff, Jamie, Yaa, Jacob

• General Updates

  • Levels Dataset - should we consolidate
  • Agency and SNU3 added
  • Creating the case statement
  • Schema overview and Maps PPT
  • Need to play around the layers in the map
  • They are all added at the same time, they need to actually be layered
  • One way is to Zoom
  • Other way is to filter , but users might forget about the filters

• Jacob sent email to reiterate the levels selector and levels dataset

• Derived vs. Systems Metrics: Once systems metrics are created, you must swap your derived metrics for the systems metrics. This should not take too long and it is important to build first and make sure the visuals function instead of waiting for the metrics. (ACTION: All)

• POCs should be checking in with their stakeholders next week to give updates and get feedback (ACTION: All)

• DREAMS

  • Got the narratives dataset added
  • AGYW and DREAMS implementation narratives
  • Cleaning up the pages
  • Mechanism and partner - it concatenates the Partner and IM
  • Great option to save space in terms of filters
  • Kimi will add request to GH
  • Need to create the time in dreams metric (ACTION: JANICE)
  • Logo needs to be added into the server - then it will be referenced to the HTML
  • If you want an image added, send it to them

• 2nd 95 Dossier

  • Levels selector allows for comparative analyses
  • Single OU dossier - looking at Cameroon, coarse age retention proxy
  • Levels selector lets you move between elements
  • Added TX_ML,
  • May want to look at the current dossiers
  • Add 3 additional metrics
  • Number of deaths and the number of tracked people LTFU
  • Total volume - getting rid of transfers and net/loss gain
  • If there is a need for production dossiers to dev, let the systems team know (ACTION: Aliyah)
  • Need to update the geo org attribute to add the new SNUs added to the levels selector
  • Working on net new and tables adapted form TSD
  • Jamie - move forward with the OU one being site level - since there are a limited number of indicators, it should be good
  • Single OU and site level
  • Will test it out - Already added facility to the dataset
  • Do a dta review and think about which indicators only need total numerator or age/sex which will also limit the number of data amount
  • Clinical cascade was filtered by specific disaggregates for specific indicators, but site
  • FUTURE REQUEST: Once everyone has created visuals, then we should go through the exercise of cleaning up the dataset so that it performs better - 5th check in (ACTION: All)
  • Don't exclude site level analytics

• 3rd 95

  • Still working on metrics
  • PMTCT_ART sum of last 4 quarters - the metric doesn't show the right values at the moment
  • VL coverage and suppression by age/fine
  • When you add in age/fine as a selector, it is not filtering unless agefine is added in as a metric in the table, then it can be filtered
  • VLC age/sex
  • The derived metric needs to have a + at the end of the formula instead of a -
  • Several systems metrics were off
  • You cannot replace a derived metric with a systems metric
  • Derived metrics that are simple, are fine locally
  • But a derived metric off of a derived metric will slow down performance
  • Need to sit down to figure out how the elements are derived
  • Will compare the metrics in the 3rd 90 and will create
  • Will create a systems metrics catalog
  • Set up a meeting to discuss process of dataset creation and metrics: Jeff, Janice, Rachel, Jamie, Katie, Jasmine (ACTION: All )
  • Describe the levels selectors
  • It is a shell
  • Need to create case statements to be able to join the text in the level dataset with the MER dataset
  • Several steps to follow
  • Joseph will work with Anu on it (ACTION: Anu and Joseph)

• HSS Dossier

  • HRH PRE - generally we have displayed the indicator and the budget, which is an external dataset
  • Are we able to add external datasets, like excel dataset
  • Use the budget data as a separate dataset from the S3
  • Is there a greenlight to use the data
  • Yaa to follow up with the PET team to ask them which files to use from the S3 bucket in the budget visualizations for Panorama (ACTION: Travis)
  • Janice will compare budget data in dossier to see what the source of the data are (ACTION: Janice)
  • SC_STOCK are ongoing - building them this week
  • HSS dataset is still in progress according to the dataset
  • How to tag and use as a filter sites that are above or below a certain proxy retention
  • Following up with Jacob to see how we can achieve the custom threshold metric of successful sites
  • Will create a previous period
  • Systems metrics do not replace what you do, you need to replace it with the systems metric and then remove the derived metric
  • Still in the process of creating systems metrics
  • SC_STOCK and EMR SITE, they were not in the list originally - they were added
  • Review the
  • LAB_PTCQI and other metrics

• 1st 90 - on track - did not discuss - will start here during next check in (ACTION: Jasmine)

[jb3436]

4th Systems Check In 7/18

General Notes

• Let's organize and label them
• Minimum set of selectors -
• Will be reviewing comments as a group tomorrow
• Use the existing visuals
• Some level of duplication is okay
• 101 analyses could be the same across cascades to each of the

1st 90

• Not doing a stakeholder demo for this round

2nd 90

• Working in tandem
• Need to combine dossiers
• Generic code - example dossier with the case statements for the levels selector
• Plug in issue on the treatment summary page
• Aliyah will add this to the GitHUb issue to follow up on the visuals and why they aren't working in dev but are working in PROD

3rd 90

• Got the PMTCT metrics needed
• Final ones
• Mostly resolved
• Inserting the selectors and filters and other elements
• Experimenting different layouts to see what will work
• Lab piece in 3rd 90 - what is the status? Is it in there

No HSS representation

• Concern around non HRH indicators in the dossier - not as much built out
• There is a need
• No DREAMS representation
• Recommendation for FY19Q4
• For metric creation
• Set up tables that show all metrics - get them settled, then begin working on the visuals
• Will try to reload the visuals

HTML Code

• It is ready to go (except for hyperlinks)
• Thinking of different ways to support the hyperlinking
• Can we have lachandra update the content for all the new dossiers

[jb3436]

5th Check In with Systems Team 7/25

Attendance Phone: Renee, Nicole,

In Person: Kimi, Yaa, Aliya, Jasmine, Rachel, Janice, Pooja, Joseph, Henry , Jamie

Announcements

• Reminder that everyone is presenting their dossiers to ICPI on Monday at the inbrief (ACTION)
• Objective of Inbrief Showcase - sensitize people to the work done so far, get generic reactions, not the time for deep dive feedback, this should have happened during the weekly Thursday development

First 90

• Index testing chapter needs to be double checked (ACTION)
• The cascade labeling is a bit off - double check the existing index testing dossier (ACTION)
• Look at chips tool (ACTION)
• Selector for "highest and lowest" - how to implement this for other dossiers (ACTION)
  • Filtered by 20% but works only for absolute values or
  • the entire cascade at the site level
• Edit qualification
• Operator - by rank, Highest %
• Filtering a metric
• No community testing - will add to the title and notes
• Remove the HTS Index from the name of all of the pages in the chapter - it is implied because it is in the chapter (ACTION)
• Create a community specific visual - could either add a new page or a new viz on the same page as the facility only one
• Add which disaggs were used in the notes (ACTION)
• Define "hotspot" or revise name
• Add a different color in the chart for the difference between recent and pos and add the legend

Legend Question

• Can you change the title of the legend - NO
• Can you wrap text? - NO
• You could change the name of the attribute that you put it in the color by section

2nd 90

• Formatting
  • Reference lines
  • Chapter filter
• Visual stopped working
  • Net new and treatment cascade
• Don't yet have the simple visual showing tx new and net new achievement
• Treatment summary table - update format
• Tables -grand total hovered at the bottom
• Retention
  • Retention metric proxy is not consistent in all of the dossiers -
  • Retention metric is derived created on their own
• Age selector needs to be resolved
• Retention metric proxy
• Confirm that the derived metric in 2nd 95 is the same as the CC metric
• There are 2 retention proxies - there is the 4 quarter one, and the one quarter one
• Add the 4 quarter retention proxy metric to the HSS dossier - Janice
• Labels at the top and the right
• Text - arial
• Can you do both absolute values and percentages
• TheQ3 data are dummy data and don't function properly - deselect the Q3 data and it is better
• Merged the two dossiers into one

DREAMS

• Filtered out a lot of disaggs and the team will be doing a full qc
• Janice will ask Vitara Chart if they can add a line into the options
• Added reference line for 100%
• Sort elements - edit right click on it and sort
• Nice use of analytic questions in the notes
• Not sure why the all OU filter is not showing the SNUs
• Janice to take a look to see why the AGYW PREV chapter isn't working
• Add "in list" AGYW PREV - Kimi to add to the GitHUb issue
• Chapter Filter - let them know if you don't want to show the chapter filters
• If the chapter filters

3rd 90

HSS

• Analysts getting much for savvy and sophisticated
• Add nots with analytic question to the pages
• Sorting items in a selector so they are alphabetically listed - JANICE TO ADVISE
• Cadre system specific attribute
• Next week
• Add the cadre as a filter
• Add Cadre for HRH PRE
• Add a levels selector to the HSS dossier
• Updating visuals in the overview page to the KPI 123

To Do:

• Share Checklist and Style Guide
• Janice to help write the levels selectors for each of the dossiers
• Checklist to go through all dossiers
• Follow up with Pooja and Lachandra

[jb3436]

6th and Final Systems Check in 8/1

Attendance: Phone : Kimi (in TZ), Anu (in India), Jacob (in DRC), Yaa, Renee, Erin, Mackenzie, Katie, Nicole, Aliyah, (others on the line I missed) In Person: Janice, Pilar, Rachel, Jamie, Pooja, Henry, LaChandra, Travis, Alice, Jasmine

Notes and Actions

• 3rd 90 -

  • Removed all the extra items and pages
  • Almost done with all the visuals
  • Add a reference line to charts with percentages on the axis (ACTION)
  • Can we make the tables below the visuals centered? They are set to left with a lot of padding (ACTION)
  • Will move selector next to the tables since it only works on the tables (ACTION)
  • Add the levels selector to the VLC and VLS page - it is possible and is how all the site level pages in the CC work (ACTION)
  • All summary graphs have the level selector
  • Janice will create the levels selectors for the 3rd 90 dossier and will then tell Anu/Yaa when it is available (ACTION)
  • Janice will make a system metric (ACTION)
  • First chapter - PMTCT ART
  • Cover Page - HTML Content (ACTION)

• DREAMS

  • AGYW PREV page isn't working anymore - Janice got it to work by adding a clause (ACTION)
  • Cannot get the chart to show with just one of the disaggs filtered
  • Discussion about prioritization levels
  • For Q4 - think of how to reflect different DREAMS Ous and prioritization

• 2nd 90

  • Content for the overview page to be sent to LaChandra
  • Sex selector still not working - it was populating before, but not working now (ACTION)
  • TX NEW TRENDS by AGE/SEX
  • Copy of Second 90
  • Pooja to assist with layout and color palettes (ACTION)

• HSS

  • Could the ratios be presented as fractions instead of decimals? (ACTION)
  • Decimals are hard to interpret for ratios - requires mental math
  • MSTR solution
  • 100% stacked proportions and the absolute totals were split out
  • How to control the sort of one in the sort of the other - needs investigation (ACTION)
  • VLS denominator was not in the table so it didn't work - (ACTION)
  • Remove filters for the HRH PRE since it is only at the OU level - insert note - this indicator is only at the OU level (ACTION )
  • SC STOCK - make sure the data label is centered on both - inside the end of the chart (ACTION )
  • Janice working on FIX for the lab specimen test type received - visual isn't populating (ACTION)
  • Need time trends in provider to patient ratio - may push for Q3 Clean (ACTION )
  • Axes not visible for horizontal bubble chart - Will make sure that the axes are visible (ACTION )
  • Actions Lab
  • Move lab testing type below the generic filters
  • Org level selector - TRAVIS will add that to the page
  • Actions Overview Pages and Notes
  • Will move the overview to the cover page
  • Move consice analytic questions into the visuals
  • Remove the chapter overviews and create a larger landing page

• QC - UAT Process

  • Analysts have attempted to do a QC to ensure the visuals are correct, want to make sure there will be additional testing and documentation
  • Will do the test once the dossiers are migrated to dev pano from dev paw
  • Will make analysts a part of the initial testing (ACTION)
  • If analysts find things that are off/missing/not accurate - put a note in the ticket (ACTION)
  • CANNOT ADD NEW THINGS to visual
  • Removing editing privileges for analysts once the dossiers are in dev pano
  • Use the pepfar login - (ACTION) circulate this

• Server

  • Having memory issues with the server
  • Needs to be restarted
  • Switching the servers on Monday

• Handover

  • On Monday will circulate the pano dev login to the dossier POCs and let them know on Monday
  • Archive Folder - Create an archive folder in each one for old versions (ACTION)
  • Finalization:
  • Jacob Request - Provide line by line responses to the most recent comments (ACTION)

• Formatting in MSTR

  • Layout - not aligned / symmetrical - will submit a ticket to MSTR (ACTION Rachel)

• Later Iterations

  • Insert hyperlink to MER guidance videos

[jb3436]

Final Fixes Notes Hi all,

We are almost at the finish line. The dossiers need to be migrated today. However there are still some issues to address to make the dossiers ready for Q3 initial release. Here are the final issues to address for your dossiers to go forward. Most of these are very quick fixes.

There is no shame in kicking off a problematic page until a fast follow.

2nd 90

• To make things more clean and gain a little space, suggest deleting the X axis titles when they are obvious ("Period") or related to the Org Selector.
• P5: I may be wrong, but I'm a little confused about the time period for this page. It has a quarterly filter, but then it has "TX_CURR prior APR." So if I select Q4, what will it show? I may be wrong here.
• P6: Suggest using same colors as CC dossiers. Two of these colors are very similar.
• P9: Should the axis scales be allowed to be independent by row? Otherwise \<15 will always be tiny. I think it's ok. Trends is more about seeing progress, whatever the volume level, not comparing volume levels.
• All pages need text notes, where appropriate. All pages need text notes, where appropriate. Especially if you use any kind of calculated metric
• P19: Stacked age should always be descending. Also check age selectors to see if this is done.
• P11: This should be limited to Q1 and after, right?
• P14/15:
  • Make legends larger
  • WOuldn't it be better to swap axes? If you have 15 IP, you have more vertical space than horizontal space to compare them
• P16:
  • Patient died / TX_CURR: a decimel should show. Right now the axis says 0, 0, 1, 1,
  • Is this TX_CURR current Q?
• P16:
  • If you don't have time to stratify by sex / age, please add them as a Filter so the user can dig.
  • Swap axes for table, since if you have 15 PSNU you will have a wiiiiide table
• P16: Ditto, but just for Sex
• P28/29: Just to confirm that you want to use quarterly retention proxy and that the Period selector is just "Choose one." (it seems like it is Multiple Options and All, which doesn't work with with the Q Proxy) The other option would be to remove the Period filter and make it the 4 Q retention proxy (which is a little more stable given more data).
• P29:
  • sort descending by retention prxy
  • Manage retention proxy scale: set min to 0%? And add a text note. It's really hard to have \<0% retention with the new proxy, unless it is a program shift (artifact).
• All table padding should be set to Small.
• P30: Remove facility code from table.

3rd 90:

• All pages need text notes, where appropriate. Especially if you use any kind of calculated metric
• P2: Org level should be on top and period on bottom
• P3: Why isn't this by Q? People will want to see Q by Q results
• P4-6: Remove the word "Constant" from ref lines. Ensure all have 100% ref lines.
• P7/8:
  • Suggest making "summary" graph smaller. The site level graph needs more space.
  • Site level graph is not site level
  • Suggest moving graphs around and making table tall and long instead of wide and short—more painful to scroll through many sites with only a small number can show. Reduce to key variables. Agency and PP not truly needed given the Selector filters and the Summary graph filter (they can also be added to tooltip I think). VLC is there twice.
  • The site graphs are pretty hard to read/manipulate because the VLS is very clustered. I suggest:
  • Manipulating the axis scale, if possible?
  • Adding a bar graph for VLS (and also one for VLC if you have time), which can be pulled directly from the CC Site dossier.
  • I don't see a site-level VL non-suppressed. That can also be quickly borrowed from CC site.
• Still missing VLS by age/sex/Org level and VLS trends by age/sex/Org level—can be easily pulled from CC
• P4-6: Given that you have four different colors going on here and two kinds of marks, I strongly suggest removing the mark labels as it makes it more cluttered. They will still have the tooltip.

HSS:

• P3/4: Suggest swapping x and y axis. If you have a lot of IPs this is going to get crunched.
• P5: Suggest inversing the metric so that it is pts per provider. Much easier to interpret.
• Suggest deleting all non-essential x-axis titles, like "Org for display"
• P10: This is awfully confusing. Suggest making these bubbles. It took me a while to realize that the bars were sites.
• P11: Suggest swapping x and y axis for same reason as p3/4
• P12-15: Need a way to deal with axis scale issues. Impossible to interact with/consume when all the sites are so clustered.
• P16/17 and elsewhere: Why no standard Selectors for Prioritization, SNU1, IP?
• P19: This kind of graph looks beautiful but is not useful. Please remove and we can try to modify the approach for a subsequent version.
• P22: This has an OU filter but I think it should be an Org level selector?
• P25:
  • Suggest swapping y and x axis.
  • I confess I don't understand this graph. Both colors are "1". Perhaps this should be for a fast follow
• P26: As I have expressed many times, raw TX_NET_NEW is not acceptable. Please change this to % net new achievement or remove and we can add it at a later point.
• P25/27: could these be combined with a filter for program area

[jb3436]

Today’s Timeline: 11:00am – 1:00PM

• Pano Server upgrade and changing server instance (DEV DOWN during this time)
• Dossier POCs - prepare responses and plan to address final fix comments from Jacob (ACTION)

1:00PM – 5:00PM

• Dossier POCs make formatting, context, content layout udpates to make dossier visually 100% ready to go (ACTION)

• Catalog items that are outstanding in terms of filtering/selectors and UPDATE GITHUB TICKET (ACTION)

• Before OOB 8/6 Jacob gives final call on dossiers being migrated (ACTION)

Next Steps:

1. Handover to DataStrong COB 8/5
2. Jacob gives final call of visual integrity for dossiers to be migrated by OOB 8/6
3. Migration of Dossiers from dev Workspaces to dev Pano 8/6
4. Testing begins Date TBD
5. Dossier POCs to work with DS to do thorough testing and QC of dossiers once they are using the Pano schema Dates TBD
6. Final QC data correctness review to determine if dossier can go live in Pano – Date TBD