evidence layer

Evidence that compounds

The evidence layer underneath every Askable study. Findings, scored and traceable, growing stronger with every study you run.

Every study feeds the layer

The transcript captures what participants said. The recording captures what they did. Vision analysis for unmoderated tests, event data for Figma prototypes. The pipeline processes both streams in parallel and ties every quote and screen event back to a verified participant.

verbal evidence

Every quote, structured

What participants said, extracted from the transcript and tied to the participant who said it. Every utterance scored the same way, whether it came from an interview or a website test.

Maya R • 34, Melbourne • Grocery shopping study
Behavioural evidence

What they did, not just what they said

Vision analysis of every screen recording surfaces where users paused, hesitated, or backed away. For Figma prototypes, event data comes directly from the API.

A finding is a structured insight, not a summary.

Every finding carries the same shape, regardless of which study or method produced it.

01

Title

Behaviour-oriented, descriptive.
02

Synthesis

Prose summarising the pattern across sessions.
03

Mention & participant counts

How many people, how many times, across how many studies.
04

Source date range

When the evidence was captured.
05

Quality score

Five dimensions: spontaneity, novelty, specificity, clarity, behavioural grounding.
06

Highlights

Playable participant clips, each with attribution and quote.
07

Source links

Links to the studies and sessions the finding came from.
finding
quality:
95%

Trust in AI drops as the trade size grows

First-time investors trust AI advisors for general guidance but reject them for high-stakes decisions. The same users who happily ask the AI about diversification refuse to take its recommendation on a large trade without speaking to a person first.

mentions
204
ppts
121
source range
Aug 5 — Sep 12

“ I trust it for the small stuff but not for anything over five grand “

Maya R, 28 • New York, US • First-time investor

context:

When asked how she decides whether to act on an AI-generated recommendation, Maya described a personal threshold she applies before taking advice.
View source
“ If it tells me to move my super, I want to talk to someone. “
Kim J, 35 • Chicago, US • Experienced investor

Every new study compounds the last

Findings live across studies, not inside them. A pattern that surfaced once becomes one finding with deeper grounding the next time it shows up. Confidence climbs. Outliers become visible. Cross-study patterns emerge that nobody was looking for.

1 month
finding
quality:
64%

Trust in AI drops as the trade size grows

ppts
11
AI Advisor adoption
4 months
finding
quality:
78%

Trust in AI drops as the trade size grows

ppts
42
AI Advisor adoption
New investor onboarding
High-value trade flow
Retirement planning interviews
1 month
finding
quality:
95%

Trust in AI drops as the trade size grows

ppts
121
AI Advisor adoption
New investor onboarding
High-value trade flow
Retirement planning interviews
Robo-advisor switch evaluation
Voice of customer Q3
Brokerage app usability
Pension product comprehension

One layer. Four ways to reach it.

The layer shows up wherever decisions get made.

Study results

Open the results of an AI moderated interview or live website test. Findings extracted from that study's sessions appear in a built-in results view: scored automatically, with playable highlights, participant attribution, and source links.

Explore AI Moderated interviews →

Ask AI

Ask a question. Findings come back with inline citations, supporting quotes, and source clips. Every claim traces back to a verified participant.

Explore Ask AI →

In docs and reels

Generate a written doc or a video reel from any set of findings. Editable in place. Shareable by link, email, or team access. Comments, embedded clips, source links.

More about artefacts →

Connectors in your stack

Connect Askable to Claude, ChatGPT, Cursor, Copilot, Figma Make, Lovable. Query the layer from where you already work.

Explore Connectors →

Every session.
More evidence.

Book a demo