Generate reports for Sales' data needs

[mkdir-washington-edu]

Following on from the Rev Ops project:

• Generate report of deal to org mappings for revops to review prior to ORG ID / date range work #1536
• PRD

The goals here largely mirror hypothesis/lms#6159, some of which is already complete:

1. Fetch contract dates from HubSpot
2. generate usage reports for each ORG as a background task
3. store the results in the DB
4. Write the user count back to HubSpot

Billable user definition: https://www.notion.so/Billable-User-14c5d0cdd2c24950b696d6801c586050
Also: https://github.com/hypothesis/report/blob/main/docs/decisions/billable_users/user_metrics.md

Sales' data needs
What data is used now, even if the counts are not accurate ("not accurate" because: some counts might be active instead of billable users; the time periods counted may not be accurate, especially if they depend on contract dates). Counts based on:

• this semester, last semester, this academic year, last academic year, 12 months, 24 months, this billable year, last billable year
• • Sales' comment on the above list: "these are helpful regardless of contract dates".
• • I.e.: there's value in knowing usage both by the relative contract dates, and also the non-relative measures like last x months.
• • We'll likely need a replacement for the semester metrics, as that's unknown and changes from institution to institution (including institutions that use quarters, not semesters).
• • There's still some confusion (on my part) on how to treat partial- and multi-year contracts w/r/t "last billable year".

Why do we need Billable users by these time periods (so we can brainstorm useful solutions):

• About 2/3rds of our customers use contract dates of July 1 -> June 30th
• • we want to know if they're using as much as they've contracted
• • the primary use case is driving up usage earlier in the year to get to our contracted billable usage numbers
• • "for some of these measures, lining usage up with our contract dates would be confusing for us" (because we're looking at fixed time periods)
• Other use case: we want to be able to see growth year over year to track

Other case: "Billable teachers"

• The HubSpot "Billable teachers" metric, when correct, can be useful comparatively, e.g. "are we gaining or losing teachers".

[mkdir-washington-edu]

PRD: https://docs.google.com/document/d/10dSpp6QRN3PipDpyxvwfyAxKmF3zoR9hEdxyLZdmLBA/edit?usp=sharing