Mandatory filters, taxonomy match, semantic retrieval, evidence assessment and weighted scoring stay five separate layers — so relevance is never mistaken for qualification.
When a supplier selection is challenged, the reasoning usually lives in a specialist’s head, which is exactly where it cannot be inspected. SpiderNet keeps matching in five distinct layers and shows the result as a requirement-by-requirement matrix with evidence counts, gaps, contradictions and risk flags, so the shortlist can be opened up rather than asserted.
Five separate layers: mandatory filters, taxonomy, semantic retrieval, evidence assessment, weighted scoring.
A requirement-by-requirement matrix per supplier.
Evidence counts, gaps, contradictions and risk flags shown.
Pool origin is visible — approved, known potential or external.
An approved shortlist is immutable.
Yes, with a recorded reason — then the approved shortlist becomes immutable.
No. Ranking neutrality is an architectural constraint: supplier-side monetisation never affects qualification, matching or ranking.
No, and keeping them separate is the point — a relevant-looking supplier can still fail a mandatory requirement.
Extracts each requirement from the SOW with its type, importance and a citation to the page.
A requirement excludes a supplier only after a person confirms it is mandatory.
Specialised agents widen the candidate pool beyond the suppliers you already know.
Every material attribute carries its source, date, method, confidence and verification status.
See the complete product: problem, features, how it works and deployment.
Request a demo to turn a real statement of work into an evidence-backed supplier shortlist.