Diversity Sourcing Report Generation
Most diversity sourcing reports get assembled once a month by exporting ATS data into a spreadsheet, cross-referencing self-identified demographic fields against pipeline stage and sourcing channel, and building the same pivot table someone built last month. It's slow enough that the report is usually a lagging indicator rather than something a recruiter checks before deciding where to source next, and small errors in the manual join between systems undermine confidence in numbers that often get shared with leadership or used in compliance reporting.
STARTING PRICE
From €299
Standard tier · Multi-step workflow with AI extraction/decisioning and 2-3 integrations.
Get a quote →Saves roughly 4-6 hrs/month for recruiting ops, plus faster course-correction on underperforming sourcing channels.
How the automation works
We connect self-identified demographic data to pipeline-stage and sourcing-channel data directly from the ATS and generate the funnel-stage breakdown automatically, refreshed on a schedule rather than assembled by hand each month. Reports show where diversity representation drops off through the funnel — a channel that sources diverse candidates well but loses them at phone screen looks very different from one that never sources them at all, and the two problems need different fixes. Aggregate-only reporting protects individual candidates from being identifiable in small requisition pools, and the underlying data stays segregated from hiring decisions per standard EEO compliance practice.
Process flow
- 01
Demographic and pipeline data syncs trigger
Self-identified demographic fields, sourcing channel and pipeline stage data pull from the ATS on a scheduled refresh instead of a manual export.
- 02
Join demographic data to funnel stage ai
Demographic responses are matched to each candidate's current and historical pipeline stage, showing where representation changes through the funnel rather than just at the top of it.
- 03
Aggregate to protect individual privacy ai
Reporting rolls up to aggregate counts only, suppressing any breakdown small enough that an individual candidate's demographic response could be inferred from the requisition pool size.
- 04
Break down by sourcing channel ai
Representation is shown by sourcing channel, distinguishing a channel that sources diverse candidates but loses them mid-funnel from one that never sources them in the first place.
- 05
Generate refreshed report output
A funnel-stage diversity report refreshes on a recurring schedule and is available before the sourcing decisions it should inform, not after.
Inputs
- Self-identified candidate demographic fields
- ATS pipeline stage history per candidate
- Sourcing channel attribution data
- Requisition-level minimum pool size for suppression rules
Outputs
- Funnel-stage diversity representation report
- Sourcing-channel diversity breakdown
- Aggregate-only, privacy-protected summary views
- Scheduled recurring report distribution
Works with
Prefer a fully custom build instead of an off-the-shelf integration? We scope both options during your free consultation — most jobs like this one work fine on standard connectors, but higher-volume or non-standard systems sometimes need bespoke API work, reflected in the complex tier.
Where this goes wrong if you get it wrong
- Reporting demographic breakdowns at too granular a level — a single small requisition pool, for instance — can make an individual candidate's self-identified response inferable from the aggregate numbers, which is both a privacy problem and, in many jurisdictions, a compliance one; suppression rules for small pools are not optional.
- A report that only shows top-of-funnel sourcing diversity without breaking down by stage hides where representation is actually dropping off — a channel that sources diverse candidates well but loses them disproportionately at phone screen has a very different problem than one that never sources them at all, and lumping both into one number obscures which fix is needed.
- Demographic data must stay strictly separated from active hiring-decision workflows and never surface to a hiring manager making a pass/advance call on an individual candidate — the report is a sourcing and process diagnostic, not an input to any specific candidate decision, and blurring that line creates real legal exposure.
- Self-identification response rates vary by channel and by how the question is framed at application; a report that treats non-response as a known category rather than flagging it separately will systematically misrepresent the actual pipeline composition.
Frequently asked questions
Does this influence individual candidate decisions?
No — demographic data stays fully separated from any workflow a hiring manager or recruiter uses to advance or reject an individual candidate; it's an aggregate sourcing and process diagnostic only.
How does it protect candidate privacy in small requisition pools?
Breakdowns below a configurable minimum pool size are automatically suppressed, since a small enough pool can make an individual's self-identified response inferable from the aggregate count.
Can we see which sourcing channels are underperforming on diversity?
Yes, the report breaks down representation by channel and by funnel stage, distinguishing a sourcing problem from a mid-funnel drop-off problem.
How often does the report refresh?
On a recurring schedule you set, rather than a manual monthly export, so the numbers are current when a sourcing decision needs to reference them.