How we turn a conversation into evidence a funder can trust
We collect structured evidence directly from the people your programme reached — built to meet the standards institutional funders and policymakers require. Here is exactly how, from first question to final indicator.
Most methods reach the reachable. We built ours for everyone else.
Traditional approaches rely on small samples — and samples are drawn from the people easiest to reach. Fortell enables direct, continuous engagement with entire populations, including the ones a written survey systematically loses.
What it captures
- A stratified few hundred, generalised to a whole
- Only the literate and the connected
- The answers to questions already written down
What we capture
- Everyone reached — complete enumeration
- The non-literate, on any phone, in any language
- What people raise unprompted — coded and counted
The VoicePrint Framework
Three layers of evidence, captured in a single engagement — structured around what happened, what changed, and what it meant.
Profile
Who was reached. Demographic and outreach data establishing who the programme served, and whether it reached its intended population.
“Who did we reach?”
Outcomes
What changed. Structured indicator data mapped to whatever framework you already report against — OECD DAC, IRIS+, UNEG, a national logframe, or the European Pillar of Social Rights — collected directly from the people your programme reached, in any language.
“What changed?”
Voice
What it meant. Open-ended responses in people's own words — the change that lives beyond the measurable indicators.
“What did it mean to them?”
We map to the framework you already report against.
You don't adopt our indicators. We map to the reporting framework your funders already recognise — so your results read as evidence they trust, not a scorecard they have to take on faith. Wherever you work, and whoever you answer to.
The most demanding framework we've mapped end-to-end is the European Pillar of Social Rights — 41 indicators across three tiers, in a live evaluation of a Spanish microcredit programme. If we can satisfy Europe's own social-rights audit, we can map to yours.
How we process the data.
Before collection begins, we map the instrument to the framework you report against. From there, every response moves through the same pipeline — so raw voice becomes structured, auditable evidence.
Transcription
Speech-to-text tuned for low-resource languages and regional accents. Timestamped, and traceable back to the source audio.
Translation
Translated for analysis, with the original-language transcript preserved. Every language is verified by a native speaker.
Coding
Responses coded against your framework — OECD DAC, IRIS+, EPSR, or your own logframe. AI first pass, human adjudication, inter-rater agreement reported.
Analysis
Closed answers aggregated into indicators. Open answers clustered into themes with verbatim quotes, segmented by the cohorts that matter.
Reporting
A live dashboard while it runs, a written report when it's done. Everything exportable — no black box.
Quality assurance
Every stage spot-checked against the layer below. Discrepancies logged and resolved before anything reaches the report.
Held to an academic standard.
AI-coded voice data matched human expert analysis at κ = 0.75 — substantial agreement under peer-grade conditions, comfortably above the 0.61 academic threshold.
Fortell Impact reports contribution, not attribution — the appropriate evidentiary standard for programmes of this type and scale in the European context.
Built for the people at the other end of the line.
Informed consent
Every participant understands who is asking, why, and how their words will be used — before they answer a single question.
Data protection
Responses are handled to institutional data-protection standards, with anonymised reporting and secure storage throughout.
Dignity by design
Avatars are built to feel culturally close — the right language, the right accent — so people speak freely, without judgement.
Ready to discuss your evaluation approach?
Tell us about your programme, and we'll explain how the VoicePrint Framework applies to your context.