How recommendations are ranked

Signals used

PavedIQ 1.0 combines evidence-weighted skill similarity, transferable-skill adjacency, role and seniority progression, career trajectory, compensation alignment, work style, and industry fit.

Signals excluded

Protected personal attributes are not accepted or scored. An automated test verifies that adding age, gender, or ethnicity fields cannot change a ranking.

What the numbers mean

The match score is a directional, deterministic ranking aid produced by unvalidated heuristic parameters — never a probability of hiring. Signal coverage (previously labeled confidence) measures whether the inputs needed for each weighted ranking component were present; it does not judge their truth and is not statistical confidence or interview likelihood. Growth scores are directional comparisons, not earnings promises. Every ranking response discloses its heuristic parameter version and known limitations, and each match lists the signals PavedIQ could not see.

User control and exploration

Goals and explicit feedback can rerank results. One in four visible curated positions is reserved for a higher-growth adjacent role when the catalog allows it, reducing over-narrow personalization.

Limitations

PavedIQ has not yet been calibrated against real candidate relevance outcomes. Its weights, thresholds, synonym table, and adjacency values are versioned expert choices, not validated estimates; skill similarity is lexical and may miss semantic relationships. Sparse, non-standard, non-English, frontline, and career-break profiles are known blind spots. Before general availability it requires segmented relevance evaluation against strong baselines, fairness monitoring, drift detection, and an appeal/correction workflow for parsed evidence.