Torch - Monitoring and Evaluation Platform Torch Logo

Qualitative Analysis

Evidence-first qualitative analysis, built for MEL teams

Code text, build themes, generate findings, and trace every insight back to its source. Torch gives your team a structured qualitative workspace that connects directly to your indicators and results.

Join the waitlist

The Qualitative Workspace

Everything you need, in one place

Navigate a structured workspace with dedicated areas for every stage of your qualitative analysis, from raw evidence through to synthesised findings.

Sources
Upload documents, transcripts, and media files into a structured folder tree. Files are processed through a pipeline including text extraction and transcription, ready for coding.
Codes
Build your codebook by highlighting text within sources and applying labels. Codes are created in context and surfaced in a dedicated index for easy management.
Themes
Group related codes into higher-level themes for interpretation. Explore theme patterns and relationships to uncover the story in your data.
Co-occurrence
Inspect how codes appear together across your evidence. Matrix and network-style views reveal connections with drill-through to supporting material.
Trends & Subgroups
Track how themes evolve over time and compare code patterns across project-defined subgroups such as cohorts, locations, or participant tags.
Findings & Evidence Tables
Synthesise evidence-backed findings that link to excerpts and MEL structures. Generate structured evidence tables for reporting and audit trails.

From Evidence to Insight

A clear path through your qualitative data

1

Upload and process sources

Bring in documents, audio, and video. Torch processes your files automatically — extracting text from PDFs, transcribing audio with speaker diarization, and organising everything in a folder tree you control.

Open-text survey responses can also be turned into qualitative sources, so survey data sits alongside your other evidence in the same workflow.

Sources
Interview_Transcript_01.pdf
Focus_Group_Recording.mp3
Survey_Open_Responses.csv
Field_Notes_March.docx
Coding
Source document with passages highlighted in pink and green to show qualitative codes applied to text about regional partnerships and policy.
2

Code and build themes

Open a source, highlight text, and apply codes from your codebook. Group related codes into themes to surface patterns across your evidence. Co-occurrence analysis shows where codes intersect, and trend views track how themes shift over time.

3

Generate findings and link to MEL

Synthesise your coded evidence into findings — evidence-backed statements that connect themes and excerpts to your indicators and results. Build evaluation reports, learning briefs, and structured evidence tables for donor reporting and audit trails.

Outcome harvesting entries let you capture outcomes with narrative, evidence, and verification fields linked directly to your qualitative evidence.

Finding
"Community governance structures show increased female participation across all programme sites..."
Linked to theme: Governance & Participation
Backed by: 12 excerpts across 6 sources
Indicator: 1.2 — % women in decision-making

Optional AI Assistance

AI that supports your analysis, never replaces it

Torch offers optional AI capabilities to accelerate qualitative workflows. Every AI output is a draft for your review — your team always has the final say. Your administrator controls which features are enabled.

Code Suggestions
Get suggested codes from selected text and context, speeding up the coding process while keeping your codebook under your control.
Summarisation
Generate summaries for excerpts or source content to quickly orient yourself within large volumes of qualitative material.
Theme Clustering
Receive suggested groupings of codes into themes, helping you spot patterns and build your thematic framework more efficiently.
Finding Generation
Draft findings from your themes and excerpts. Review, edit, and accept or reject AI-generated insights before they enter your analysis.
Learning Briefs & Reports
Draft learning brief and evaluation report content from your coded evidence, saving hours on document preparation while maintaining quality.
Indicator Explanations
Generate narrative explanations for indicators backed by linked qualitative evidence. Accept or reject with full traceability.

Integrated with Your MEL Framework

Qualitative evidence that connects to everything

Indicators
Link qualitative findings and evidence directly to indicators. View all supporting qualitative material from an indicator's dedicated evidence tab.
Theory of Change
Map qualitative evidence against your theory of change structure. See where your evidence supports programme logic with coverage overlays on ToC elements.
Dashboards
Surface qualitative summaries on dashboards — theme counts, coded evidence breakdowns, and the share of results backed by qualitative material.

Solutions

Where qualitative analysis fits

The same Torch platform described on each solutions page brings qualitative work together with your indicators, results, and strategy.

International development programmes

Link coded evidence and findings to indicators, results, and your theory of change so donor reporting and adaptive management rest on traceable qualitative data across sites and partners.

Learn more

Impact investors & portfolio teams

Ground social performance in coded narratives and themes alongside quantitative metrics—support portfolio reporting, assumptions validation, and a clearer story of change for stakeholders.

Learn more

Social enterprises

Reduce reporting burden by keeping qualitative evidence, themes, and findings in one workspace tied to your strategy and KPIs—easier updates for donors and internal teams.

Learn more

Be the first to try qualitative analysis in Torch

Join the waitlist and we'll let you know when qualitative analysis is available for your team.