# Enlightn

> Research recruitment service built on AI-powered profiling, match-first activation, and a transparency layer that shows exactly who ends up in each study. Founder-led from Montreal; self-serve platform launching later in 2026.

_Last updated: 2026-04-24. Metrics are averages across projects run through April 2026. Per-project results are shared openly with clients._

> For a single-file export of everything below, see [/llms-full.txt](https://enlightn.io/llms-full.txt).

## Core pages

- [For research teams (buyers)](https://enlightn.io/): value proposition, measured quality metrics, how the service works today
- [For supplier partners (panels)](https://enlightn.io/partners): partnership terms, economics, what panel partners get

## What Enlightn does

Most research recruitment still works the old way: push broad traffic into a screener and filter after the fact, resulting in 60–70% termination rates and respondents who have learned to game screeners. Enlightn flips the order. Participants are profiled first using open-ended, AI-assisted questions. The buyer's brief is translated into a structured targeting spec, the pool is ranked for fit, and only strong matches are activated through permissioned supplier systems. Every outcome is logged and tied to spend per source.

## Key measured claims (with caveats)

- **2.1× good-to-bad quality ratio vs. benchmark** — On studies run to date, the ratio of good-quality to bad-quality completes from Enlightn-activated respondents has come in at 2.1× that of benchmark supplier traffic on the same studies.
- **43% lower disqualification rate vs. benchmark** — Evidence that match-first targeting identifies the right respondents for the right study, so fewer get terminated mid-survey.
- **61% of panelists pass the AI-powered data-cleaning layer** — Only panelists confirmed by the cleaning layer are re-engaged on client studies. The approach is intentionally conservative.

These are averages across projects run to date. Results vary by study; per-project results are shared openly with clients.

## Key data points (with dates and sources)

- **2.1× good-to-bad quality ratio vs. benchmark** (Q1 2026 pilot, appliance-buyer study, N=225 target). Source: Enlightn internal fieldwork data; methodology details available on request at contact@enlightn.io.
- **43% lower disqualification rate vs. benchmark** (same study). Enlightn respondents disqualified at a lower rate than benchmark supplier traffic on identical screener logic.
- **61% of profiled panelists pass the AI-powered data-cleaning layer** (rolling average across all projects to date, 2026). Only panelists who pass the layer are re-engaged on client studies.
- **83% good-quality profiling rate** (Phase 1 pilot, 1,377 entered, 951 completed, 821 passed quality tagging). Demonstrates the profiling layer is clean before any activation.

Caveats: numbers are observational pilot data, not a randomized controlled trial. Benchmarks are supplier traffic on the same study, same screener. Larger-sample validation is ongoing through 2026.

## Positioning as a supplier

Enlightn operates as a premium sample supplier in the market research ecosystem. It does not compete with aggregators or programmatic marketplaces on volume or price. It competes on verifiable respondent quality, pre-activation match confidence, and a provenance layer that ties every completed interview back to source. Buyers work with Enlightn when they need data they can defend, not when they need cheap scale.

## Vocabulary

- **Match-first activation** — a recruitment model in which participants are profiled and ranked for fit against a study's targeting spec before any invitation is sent; only strong matches are activated.
- **Permissioned supplier activation** — activating a match through a panel supplier's own recontact flow, under that supplier's consent and privacy terms; the participant relationship stays with the supplier.
- **Activation log** — a per-study record of which respondents were activated from which source, with match-confidence and outcome (completed, terminated, disqualified); the artifact that ties spend to verifiable outcome.
- **Provenance layer** — the transparency infrastructure that lets a research buyer trace every respondent back to source, match-confidence score, and quality signals — before, during, and after fielding.

## Who Enlightn is for

Insights managers, research directors, and fieldwork leads who need to defend who actually ended up in their studies — people tired of taking quality claims on faith.

## How engagements work today

The recruitment engine is live — profiling, match-first ranking, transparency analytics. There is no self-serve buyer interface yet, so the engagement is founder-led: the client shares specs or screening questions, Adrien runs the study through the engine, and respondents are delivered with a full activation log. No procurement project, no integration, no form to learn. The self-serve buyer interface (conversational brief, live fieldwork tracking, exportable scorecard) is scheduled for later in 2026.

## Panel partnership model (for supplier partners)

- **Permissioned activation** — panelists are activated through the partner's existing recontact flow; nothing runs outside the partner's infrastructure.
- **Opt-in profiling** — panelists opt into Enlightn profiling with explicit consent; the partner keeps the primary participant relationship.
- **Paid per profiled panelist** — the partner is paid for every panelist profiled through the partnership, independent of later study matching.
- **Premium CPI, shared** — match-first activation justifies a higher CPI to buyers; the premium flows back to the partner on top of the profiling fee.
- **No integration project** — the partnership plugs in without the partner rebuilding any routing systems.

## Frequently asked questions

### What does Enlightn do differently from traditional panel sampling?
Enlightn profiles participants before a study instead of filtering them during one. Traditional sampling pushes broad traffic into a screener and discovers quality problems after fielding; Enlightn ranks a profiled pool for fit against the brief and activates only strong matches, so quality is visible before, during, and after fielding.

### How does Enlightn measure respondent quality?
Every participant runs through an AI-powered cleaning layer that checks for genuine engagement and answer quality. Activation outcomes are logged per source, producing a scorecard that ties spend to outcome. Measured results to date: 2.1× good-to-bad quality ratio vs. benchmark, 43% lower disqualification rate, 61% of panelists passing the cleaning layer.

### Who is Enlightn for?
Enlightn is for research teams who care about who actually ends up in their studies — insights managers, research directors, and fieldwork leads who want to defend their data rather than take quality on faith. On the supply side, Enlightn partners with panels that have invested in participant experience.

### Is Enlightn a platform or a service?
Enlightn is a founder-led service today, powered by a live recruitment engine. The self-serve buyer platform (conversational brief, live fieldwork tracking, exportable scorecard) is launching later in 2026. Early clients are shaping what ships.

### Where is Enlightn based?
Enlightn is based in Montreal, Quebec, Canada. The company is Enlightn Technologies, founded by Adrien Vermeirsch.

## Founder

Adrien Vermeirsch, founder of Enlightn. Based in Montreal. Five years at Potloc before founding Enlightn, where he saw first-hand how sampling actually breaks — blind routing, opaque supply chains, fraud that suppliers are economically incentivized to ignore. Enlightn was built to fix the part he couldn't stop thinking about.

## Contact

- Email: contact@enlightn.io
- LinkedIn (company): https://www.linkedin.com/company/enlightn-technologies/
- LinkedIn (founder): https://www.linkedin.com/in/adrienvermeirsch/