Anthropic BenDep · IDinsight

Measurement & learning roadmap

A constellation of measurement tools across three stages — Adoption → Programmatic outcomes → Impact — for the four 1:1 partnership verticals and the Scaled Program. Cell shading = priority (decision-relevance, not data readiness — almost none of this data exists yet). Three priorities carry the engagement: what Claude usage looks like in the social sector (Scaled), how Claude changes program delivery (all four vertical leads), and the Education deep-dive (every cell).

Priority — cell shading (decision-relevance, ≠ timeline)
High Mid Lower
Tool type
Platform dataHex · Clio · logs · CSM Survey & interviewClint · intake · interviews Partner / externalpartner M&E · public Evaluationprocess · impact Modelimpact · SROI
Timeline — per tool (the n on each chip)
1Now · 0–2 mo 22–4 mo 34–6 mo 46 mo +
Horizon

Showing the first moves — step through the horizon to sequence the roadmap, or hit All to see everything.

How to read this. Two independent axes. Priority is the cell shading — how decision-relevant a question is to that owner. Timeline is per tool: the small number on each chip (1–4) is when that specific tool, for that question, in that vertical gets done — so a single cell can mix a do-now tool with a later one. The Horizon tabs light up one phase at a time; it opens on Now so the first moves are unmistakable: Clio + Hex on usage for the Scaled Program, and Clint + interviews + partner M&E on program delivery (5a) across the four verticals. Priority and timeline are not the same — Education's beneficiary outcomes are high-priority but a later build; the usage read is do-now even where it's only mid-priority. Rows are learning questions (LQ numbers from the measurement plan), tagged Portfolio-wide (all partners, light-touch) or Lighthouse / flagship (the deep-dive on 3–5 select partners per vertical). Tool chips are grouped by method: platform data, survey & interview, partner / external data, evaluation (process, then impact), and model. Interviews and process evaluations are placed only where they refine — the "why" behind funnel drop-off and low-engaged orgs surveys miss, implementation fidelity, and beneficiary depth. Source: priorities adapted from the BenDep Needs Mapping tab; tools from the Metrics Shortlist. Hover any cell for the indicator detail.