HR knowledge graph
Your workforce as a graph, joined by what relates it.
The common data model is loaded into a knowledge graph, where an employee is linked to their manager, organisation, job, pay, reviews and events. Questions follow those links, so “regretted leavers this year, by manager and pay against their band” is one query, not a project.
Regretted leavers this year, by manager
- 1HAS_EVENT → exit, voluntary
- 2HAS_REVIEW → rated 3 or above: regretted
- 3REPORTS_TO → grouped by their manager
HR knowledge graph, part by part.
- 01
Relationships first
Reports to, member of, has job, has event: the graph keeps what connects people and roles, and a question walks the links.
- 02
Organisation design, built in
Reporting lines and memberships are in the graph, so spans, layers and team shapes are a question away.
- 03
Dated records
Pay versions, HR events and reviews carry their dates, so questions about them can be asked as of a past date.
- 04
One store for every suite
Answers, KPIs, dashboards, forecasts and root causes all read the same graph, so they agree with each other.
Load, Link, Query, Answer.
- Step 1
Load
The common data model's records arrive.
- Step 2
Link
Each is joined to the people and roles it concerns.
- Step 3
Query
A question becomes a walk along the links.
- Step 4
Answer
The records it found, with the query kept.
The links a question follows
- REPORTS_TO
- An employee to their manager: spans, layers and team views.
- MEMBER_OF
- An employee to their organisation.
- HAS_JOB
- An employee to the job they hold.
- HAS_COMPENSATION
- An employee to each version of their pay, valid from and to.
- HAS_EVENT
- An employee to their HR events, each of a catalogued type.
- HAS_REVIEW
- An employee to their reviews, ratings and cycles.
- HAS_APPLICATION
- A candidate to their applications, requisitions and offers.
Questions it answers, in your words.
- “Which managers have more than ten direct reports?”
- “Which teams lost the most regretted leavers this year, by manager?”
- “How has pay in Engineering changed since last April?”
On the same model and graph
- Common data modelOne common language for every HRIS and HRMS — Workday, Keka, SuccessFactors — with the analytical layer prebuilt.
- AI analystAsk in plain English. It plans the work, runs it on the graph and shows its working.
- Security & governanceEvery query bound to your organisation; every figure leads back to its source.
Your next people decision, with the why attached.
Request a demo. We'll email you a two-minute walkthrough today, then set up a time that suits you.