End-to-End Evaluation Support

Evidence that keeps pace with decisions

We provide end-to-end evaluation support, combining deep methodological expertise with AI to deliver faster, rigorous insights for public policy — without sacrificing integrity.

Work With Us Learn More
The Problem

Decision-makers need robust evidence, but current approaches can't keep up

Evaluations are commissioned too late, delivered too slowly, and by the time findings emerge, the decisions have already been made without them.

Too Slow

Conventional evaluation methods weren't designed for the pace of modern policy cycles. By the time a report lands, the window for action has closed.

Static Reports

Commissioners need evidence that evolves with their services — not a single report, but a continuous stream of actionable insight.

Who We Are

Our mission & values

Reflexivity exists to make evaluation faster, more rigorous, and more useful for the people who commission it. We believe public services deserve evidence that arrives in time to shape decisions — not after them.

Transparent

We are open about what AI does and doesn't do in our work. Every method, model, and decision is documented and auditable. No black boxes.

Thoughtful

We don't use AI for the sake of it. Every tool we deploy is chosen because it genuinely improves the quality or speed of the evaluation — not because it's fashionable.

Compliant

We take data governance seriously. From GDPR to sector-specific regulations in health and social care, compliance is built into our process from the start — not bolted on at the end.

Our Services

What we offer

End-to-end evaluation

We design and deliver full evaluations, from framework and data collection through to analysis, reporting and recommendations.

Evaluation capacity building

Already have a team but not sure where you can embed AI? We work alongside your staff to build sustainable AI-supported evaluation workflows.

Our Approach

Transparent, considered, and quality-assured throughout.

We are not a software platform, and we are not another technology company that underestimates the messiness of real-world policy. We understand the scepticism around AI and we share it.

Our model is built on human expertise first, with AI where it genuinely adds speed and rigour — and not where it doesn't. We understand the regulatory landscape around data in health and social care, and we take it seriously. Sensitive data can be processed locally on our own infrastructure — nothing leaves your premises unless you want it to. Where cloud processing is appropriate, we use secure, GDPR-compliant servers with full audit trails.

Bespoke, Not Templated

We work alongside clients to design evaluations that fit their context. Where it adds value, we build and deploy custom models that run on your own hardware — no data leaving your laptop. We don't default to off-the-shelf AI when it isn't the right tool.

AI-Accelerated Analysis

AI supports data processing, pattern recognition, and rapid evidence synthesis — preserving human judgement for interviews, interpretation, and stakeholder engagement.

Compliant by Design

GDPR, health data regulations, and sector-specific requirements are built into our workflows from day one. Sensitive data stays on your infrastructure when needed, with full audit trails throughout.

Real-Time Insights

Faster evidence that keeps pace with policy decision-making, so findings arrive when they're still needed.

How It Works

A qualitative pipeline, step by step

Every step is human-in-the-loop. AI accelerates the process, but our researchers oversee, validate, and interpret at every stage.

01

Interview scheduling

AI-powered

Our AI scheduling assistant reaches out to your list of participants — stakeholders, service users, staff — and arranges interviews at times that work for them. No back-and-forth emails, no admin burden on your team. You provide the contacts, we handle the rest.

02

Compliance and consent

AI-powered

Before any interview takes place, our platform manages the paperwork. Participants receive consent forms, data handling agreements, and privacy notices automatically. Everything is logged and stored in line with GDPR, giving you a clear audit trail from day one.

03

Literature review

AI + Human

AI accelerates the search and synthesis of existing evidence — scanning databases, surfacing relevant studies, and identifying gaps. Our researchers guide the scope, assess quality, and ensure the review is grounded in methodological best practice.

04

Surveys

AI-powered

We design, deploy, and manage surveys tailored to your evaluation questions. Responses are collected and analysed in real time, with findings fed back to you as they emerge — so you're not waiting until the end to see what participants think.

05

Conduct interviews

Human-led

This is where we stay deliberately human. Our evaluation specialists lead every interview themselves — building rapport, following up on nuance, and adapting in real time. We don't believe the most important conversations in an evaluation should be handed to a chatbot.

06

Additional qualitative data

AI + Human

If you have other documents or qualitative data that needs including — case files, monitoring reports, programme documentation — we feed it into the same analytical pipeline. This lets us triangulate findings across multiple sources, giving you a richer and more robust evidence base.

07

Transcript analysis

AI + Human

Interview recordings are transcribed and processed using AI on secure, GDPR-compliant infrastructure. The system runs qualitative coding — identifying themes, patterns, and sentiment across your full dataset. Our researchers review and validate every output, and early insights are shared with you as they emerge to support decision-making in real time.

08

Evaluation report

AI + Human

This is where the human touch matters most. We synthesise everything into a clear, rigorous evaluation report designed for decision-makers. Findings are grounded in evidence, structured around your evaluation questions, and shaped by the interpretive judgement that only experienced evaluators can bring. Delivered in time to actually inform the decisions it was commissioned to support.

Our Team

Built on complementary expertise

We combine what no technology firm and no traditional consultancy offers individually: methodological expertise, domain knowledge, and AI capability in a single integrated practice.

Saoirse Moriarty

Saoirse Moriarty

Academic Lead

PhD researcher in Social Epidemiology at UCL and Research Fellow at RAND Europe. Deep expertise in evaluation methodology, health and social policy research, and evidence synthesis. Brings academic rigour, sector networks, and credibility with commissioners and funders.

LinkedIn →
Oisin Moriarty

Oisin Moriarty

Technical Lead

MSci Physics graduate from UCL, specialising in computational methods and machine learning. Specialist expertise in AI, data science, and applied machine learning. Ensures AI integration is robust, transparent, and appropriate.

LinkedIn →

Let's talk about your evaluation needs

Whether you're commissioning an evaluation or looking for faster, more useful evidence, we'd love to hear from you.

Get in Touch