---
title: "Get Worker Impact"
method: GET
path: "/workers/impact"
tags: ["workers"]
---

# Get Worker Impact

`GET /workers/impact`

Measured impact of the caller's organization's Workers, by month and use case.

Takes no window parameters; the series resolves server-side so every client
reports the same months.

## Response `200`

Successful Response

- WorkerImpactResponse — Measured Worker impact for an organization, by month. Clients start the chart axis at ``available_since``, and can use ``hours_saved_total_points`` as the shared month axis since it is always dense. ``downtime_averted_minutes`` is wall-clock time and is never added into hours saved.
  - `available_since` string, date-time, nullable, required
  - `hours_saved` ImpactTile, required — A headline figure for the current month, with its change. Either field is ``None`` when nothing was measured, so clients show unavailable rather than zero. The change compares the month so far against the prior month cut to the same elapsed days; against the whole of it, the 1st would report a collapse that never happened.
    - `value` number, nullable, required
    - `delta_vs_previous_month` number, nullable, required
  - `downtime_averted_minutes` ImpactTile, required — A headline figure for the current month, with its change. Either field is ``None`` when nothing was measured, so clients show unavailable rather than zero. The change compares the month so far against the prior month cut to the same elapsed days; against the whole of it, the 1st would report a collapse that never happened.
    - `value` number, nullable, required
    - `delta_vs_previous_month` number, nullable, required
  - `hours_saved_series` ImpactHoursSeries[], required
    - `mode` 'incident' | 'alerts', required — The kind of channel the agent lives in — its standing mission. Drives the swappable mode block in the system prompt (and, later, the toolset). ``incident`` and ``alerts`` have mode blocks today; ``escalation`` and others land as localized drop-ins (their standing block + tooling) when that work begins.
    - `points` ImpactMonthPoint[], nullable, required
      - `month` string, required
      - `value` number, required
  - `hours_saved_total_points` ImpactMonthPoint[], required
    - `month` string, required
    - `value` number, required
  - `downtime_averted_points` ImpactMonthPoint[], required
    - `month` string, required
    - `value` number, required
  - `assumptions` ImpactAssumptions, required — The constants these figures were priced with, returned so the client's copy reads the same numbers rather than restating them. Why each value is what it is: - ``incident_duration_hours`` (4h, from the ROI framework) is assumed, not measured. The only signal for when an incident ended is the agent's own last post, so a measured duration would grow whenever the agent is slower to notice resolution, and every figure derived from it would improve as the product got worse. - ``incident_active_fraction`` (30%) is the share of an incident a responder is hands-on rather than watching. - ``incident_efficiency`` (10%) is what Traversal removes from that hands-on time, and also drives downtime averted. - ``incident_avoidance`` (10%) covers responders never pulled in at all, so it is added to efficiency rather than compounded with it. - ``minutes_per_alert_response`` (10 min) sits below the alert-stories figure the ROI framework quotes, because an alerts Worker answers every fire where alert stories counted only the ~7% opening a fresh investigation. - ``responder_window_hours`` (4h) is a measurement bound, not a price: how long after a Worker joins a poster still counts as responding. Separate from ``incident_duration_hours`` because one number doing both jobs made the figures scale non-linearly in it — raising it admitted more responders *and* raised the per-responder rate. Uniform across organizations, since comparing two customers' hours only means something if both were priced the same way.
    - `incident_duration_hours` number
    - `incident_active_fraction` number
    - `incident_efficiency` number
    - `incident_avoidance` number
    - `minutes_per_alert_response` number
    - `responder_window_hours` number

---

[API](https://skmtc.net/traversal/apis/fastapi.md) · [All operations](https://skmtc.net/traversal/apis/fastapi/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/traversal/fastapi/revisions/2134ebffd1ef/schema)