Docs

Overview

Dialogue is a consumer intelligence API for UK personal finance companies to understand the language real people use to describe financial products. It provides structured, searchable access to millions of comments, threads, and posts from across social media platforms, community platforms, and traditional media.

Every consumer opinion is tagged with:

FieldExamples
Life stage”Coming up to retirement and terrified I haven’t saved enough”, “Just got my first proper job and no idea where to start”
Life events”Lost my job last month and can’t keep up with the mortgage”, “Had a baby and suddenly thought about life insurance for the first time”
Emotional register”I’m genuinely scared I’ll never be able to retire”, “Finally feel like I understand my pension”, “This government has completely failed ordinary people”
Products referenced”The workplace pension I’ve ignored for ten years”, “Life insurance I only got after my dad died with nothing”, “Fixed-rate mortgage we locked in just before the rises”
Behavioural signals”I’ve been putting this off for years and I know I need to sort it”, “I’m actively comparing providers right now”, “I didn’t even know I had a protection gap until I watched this”

Who it’s for

  • Marketers — write website copy, advertising, and social content in the words your audience actually uses, not the words your compliance team invented.
  • Product managers — understand real user pain points, life stage needs, and emotional triggers before writing a single line of spec.
  • Financial advisers — recognise the emotional states and life events your clients are navigating, and communicate with them accordingly.
  • Insight and research teams — access qualitative data at scale without the cost and lead time of focus groups or commissioned research.

Corpus channels

ChannelComments
Damien Talks Money1,619
James Shack756
PensionCraft759
Total3,134

Base URL

https://api.dialogue-ai.co

Authentication

Contact ben@dialogue-ai.co to request an API key.

Rate limits

All endpoints are rate-limited to 60 requests per minute per API key. The /api/v1/search and /api/chat endpoints call OpenAI embeddings and the Anthropic API respectively, so responses may take 2–10 seconds depending on query complexity.