Four steps. Zero magic. Every decision traceable.
Aurora is a decision-support layer between raw funding calls and your team's application decisions. Here's how each step works in practice.
- 01
Declare the institution
Your thematic priorities, countries, capacity areas, partner preferences, and budget comfort. Aurora uses this profile to drive every downstream match.
- Editable forever — priorities shift, scoring should too
- Used as the reference point for every opportunity matching
- Future-ready for multiple profiles per workspace
- 02
Bring opportunities in
Start with the seeded dataset, add manual entries, upload CSVs, or connect donor portals. Every opportunity ends up in the same structured shape.
- 14 structured fields per opportunity
- Source and last-updated visibility for traceability
- Status flows: new → in review → decided → archived
- 03
Score with transparency
A rule-based engine produces a suitability score with per-criterion rationale and missing-info flags. Weights are editable and auditable.
- 7 weighted criteria, default weights 25/20/15/10/10/10/10
- Major risks and missing-info surfaced explicitly
- Recommendation mapping: 80+ Apply, 60-79 Review, 40-59 Monitor, <40 Not Suitable
- 04
Review as a team
Scan the dashboard, deep-read a detail page, and decide. Notes and overrides accumulate alongside the call, not in someone's inbox.
- Reviewer attribution for every note
- Override with rationale — score is a proposal, not a verdict
- Recommendation badges: Apply / Review / Monitor / Not Suitable / Needs Partner
UN OCHA Anticipatory Action Pilot — total 72 · Review in Detail
This is an actual opportunity scored against Türk Kızılayı's declared priorities.
Product principles
Six trade-offs we made explicitly — and refuse to blur.
- Explainabilityoverblack-box magic
- Decision supportoverfull automation
- Institutional credibilityoverstartup theatrics
- Fast scanningoverdense complexity
- Source traceabilityoverconvenience shortcuts
- Configurable logicoverhidden assumptions