SAMPLE: Children's Hope Alliance × BI Consulting Services

1Workforce at a glance

Period-end values; deltas compare with the same-length prior period. Turnover is annualized.

2Headcount and annualized turnover by month

Bars: period-end headcount. Line: trailing-12-month separations ÷ average headcount.

HeadcountAnnualized turnoverTurnover target 30%

4Where the workforce is — North Carolina footprint

Circles are sized by headcount in the selected scope; colour is annualized turnover. Dots mark community-based staff outside the three offices.

Turnover under 30%30–40%Over 40%Legacy campus (Barium Springs 1891 · Grandfather Home 1914)

5Turnover by program and office

Annualized, ranked highest to lowest. Bars in clay exceed the 30% target.

6The Hope Tree — tenure and retention

Each ring is a tenure band; the canopy shows how much of today's team has stayed past each milestone.

7Open positions and time-to-fill

Requisitions open at period end; median days from posting to accepted offer for hires in the period.

8Overtime hours per FTE

Monthly OT hours per full-time equivalent, by program.

9Training and compliance completion

Share of staff current on each required course at period end.

10Ask the data

Natural-language questions over the same model, answered with the numbers on this board.

11Sensitive fields excluded at the source

The Paylocity → Azure SQL load pulls only the columns the board needs. Struck-through fields never leave payroll.

Social Security number

Date of birth / age

Race, ethnicity, gender

Home address, bank, dependents

Medical, leave reason, disability

Individual pay rate (aggregated only)

Employee ID (hashed), program, cost center

Hire / term dates, status, FTE, hours, training

12How the numbers get here — nightly, with no manual export

The architecture Kourtney and Kevin sketched in early 2025, built as a low-code, least-maintenance pipeline.

Source

Paylocity API

Employees, positions, time & labor, learning. Column allow-list enforced here.

Load

Power Automate

Scheduled flow, 02:00 ET. Incremental by modified date; retries and failure alerts to HR.

Store

Azure SQL

Star schema: employee, position, program, location, calendar. 36 months of history.

Serve

Power BI

This board, row-level security by office, board-ready export each quarter.

Private AI on your own server

The "ask the data" panel above is built to run on a GPU server inside your own network: an open-weight model over your own documents, with every answer citing the file it came from. No per-seat licence, works offline, and nothing leaves the building.

How it sits in your environment

One appliance-class server on your rack or in a secured closet. Your files are indexed where they already live; the model runs locally; answers come back with the source page attached.

What that means in practice

1
Your data only, on your hardwareAn open model over your own documents and exports. No third-party AI processor, no data-transfer agreement to review.
2
Every answer cites its sourceReplies point to the file and page they were drawn from, so people check the answer instead of trusting it blind.
3
No per-seat licenceThe whole team can ask questions. Cost is the server and our build, not a monthly number that grows with headcount.
4
Works offlineIf the internet connection goes down, the answers still come. Nothing depends on a vendor's uptime or pricing changes.