Reading results
Result fields, verdicts, starred contacts, and the project status lifecycle.
Result fields
| Field | Type | Notes |
|---|---|---|
resultId | string | Stable within a project |
fullName | string | |
title | string | Current job title |
company | string | Current employer |
location | string | City / country |
linkedinUrl | string | |
email | string | null | When available |
emailStatus | "verified" | "unverified" | null | Deliverability confidence |
verdict | "relevant" | "irrelevant" | "unknown" | |
verdictReason | string | Why this person matches or doesn't |
starred | boolean | Starred by a user in the Orbit UI |
surfacedBy | string | Search strategy label — never names a provider |
Verdicts
relevant— matches the brief. This is the signal you want to act on.irrelevant— surfaced but doesn't match. The reason explains what was off.unknown— not enough data to decide. Review manually; borderline cases land here.
Filter on verdict === "relevant" for your short list. Review "unknown" for edge cases.
Project status lifecycle
| Status | done | Meaning |
|---|---|---|
running | false | Search in progress |
done | true | Complete |
awaiting_input | false | Orbit has a clarification question |
error | true | Failed — error field populated |
Counts
{
"resultCount": 42,
"relevantCount": 35,
"irrelevantCount": 5,
"unknownCount": 2,
"requestedTarget": 30
}requestedTarget is the count extracted from the brief if you included a number. relevantCount is the usable signal set.
Pagination
Results are paginated with limit (1–200, default 100) and offset. The SDK's listAllResults(projectId) auto-paginates through all pages.