Research Intelligence Platform

The research layer your
whole org uses — and your
team controls.

Turns every interview, review, and customer signal into searchable decisions. Built for research teams on Notion, Google Drive, and OneDrive.

Book a discovery call
We're opening this to a small group first. Book a 20-minute call and we'll show you what the system looks like for your team's specific stack and research workflow.
What you get on the call
A live walkthrough of the interactive demo — built around real research data
A clear picture of what the system would look like inside your stack
An honest answer on whether this is the right fit for your team right now
See How It Works →
Apply for Early Access →
Notion · Google Drive · OneDrive supported · No commitment · 30 minutes
60–70%
less synthesis time per study
AI handles extraction & formatting
3 days → 6 hrs
study complete → insights live
From transcript to searchable card
−35%
duplicate research requests
Teams self-serve before filing
5 teams
with structured research access
PM, Data, Sales, Marketing, Research
The situation at most companies

Research doesn't disappear.
It becomes unfindable.

The bottleneck isn't research quality. It's what happens after the study ends.

01
Insights rot in folders
Studies are completed and filed. Six months later nobody can find them, let alone act on them.
02
Teams duplicate work
PMs file research requests for questions already answered. Researchers run the same study twice.
03
Decisions outrun evidence
Roadmap calls get made on gut because pulling the relevant research takes longer than the meeting.
Today — manual ops
Studies complete but insights live in Drive folders nobody navigates
Synthesis takes 3–4 days of manual extraction and formatting
PMs file research requests for questions answered 6 months ago
No way to know which decisions research actually influenced
Sales, Marketing, CS all fight over the customer voice
Admin time per study~11 hrs
With Research Intelligence
Every study in one searchable repository — findable in 60 seconds
Transcript → structured insight cards in under 6 hours, not 3 days
PMs self-serve research queries — research team unblocked
Decisions log tracks every decision research influenced
All teams contribute signals — research curates and quality-gates
Admin time per study~45 mins
How it works

Every transcript becomes
a decision-ready insight.

Five stages. AI handles extraction and formatting. Your researcher handles judgment. Nothing publishes without human review.

01
Drop the file
Upload a transcript or doc to Notion, Google Drive, or OneDrive. Detected automatically.
02
Claude extracts
3-stage pipeline: cleanup → insight extraction → JTBD mapping. Structured JSON in minutes.
03
Researcher reviews
Draft insight cards enter a review queue. Researcher edits, approves, and publishes.
04
One source. Five teams.
PMs query by product area. Marketing searches emotional JTBD. Data cross-references quant signals.
05
Decisions get briefed
PM types a product question. Claude reads the repository and returns a structured decision brief in seconds.
Who benefits

One source.
Every team.

Every team gets access to research in the format that fits how they actually work — without adding load to the research team.

📊
Product Managers
Self-serve research queries before filing a new request. Ask questions, get evidence.
PM Decision Brief
🔬
Data Scientists
Connect behavioral signals from your analytics tool to the qualitative why.
Functional signals view
💬
Sales
Real customer language for objection handling and discovery calls.
VOC feed · deal signals
✍️
Marketing
Voice-of-customer quotes for copy, positioning, and campaigns.
Emotional JTBD view
🔭
Research Team
Full system control. Automated synthesis, recruiting, and reporting.
Full admin + automation

Currently accepting teams using

In this first round we're working with teams whose research already lives in Notion, Google Drive, or OneDrive. Dovetail and Confluence integrations are in active development.

📚 Notion✓ Live
📁 Google Drive✓ Live
☁️ OneDrive✓ Live
🎯 DovetailQ3 2026
📄 Confluence · SharePointQ4 2026
Request early access →
Interactive demo

See the full system working —
with real data

We built a fully working demo using real public VOC data — synthesised from reviews, Reddit threads, and community feedback across a real SaaS product category. Every insight card, quote, and decision signal in the demo was extracted through the same pipeline we deploy for clients.

Interactive demo · real VOC data
SaaS product team · B2C platform · ~200 employees
Stack: Notion · Google Drive · Slack
"I never know if the other person has actually seen it. Sometimes they say they didn't receive anything — and I have no way to prove otherwise."
— P3, Workflow Friction study · Signal 1 — Direct solution fit · Strong evidence
The research signal
140+ public data points from reviews, Reddit threads, and community forums — synthesised through the pipeline into structured insight cards with JTBD mapping, evidence scoring, and decision signal ratings.
What the demo shows
The full system working end-to-end: searchable insight repository, PM Decision Brief with live AI, social listening feed, cross-team signal submission, and an impact dashboard.
What this proves
The pipeline works. Real quotes, real JTBD statements, real decision signals — extracted automatically and searchable in seconds by anyone on the team, not just the research function.
20 insight cards · real VOC data
5 cross-team audience views
8 JTBD taxonomy tags
Live PM Decision Brief
Book a 30-minute demo →
Get early access

See if Lumen
is a fit for your team

We're working with a small number of research teams in this first round. If it's a fit, we'll scope it together on a 30-minute call — no pitch, no pressure.

Apply for Early Access →
No commitment. We'll scope it together on a 30-minute call.