Patients arrive at health services already misinformed. We tell you what they've been told.
We read what patients are reading. YouTube comments, Reddit threads, podcast transcripts, AI chat answers, all of it. Then we tell you which misconceptions they've absorbed, which viral claims they're repeating, and what they walked in believing, so your clinicians, marketers, and onboarding teams don't have to guess.
Why we started Dialogue.
Dialogue began with a series of conversations our founder, Ben, ran with health providers, nutritionists, and clinicians. The same theme came up in every one: their patients and customers were turning up shaped by something they'd seen online, and the team had no reliable way of knowing what.
Misconceptions were costing consultation time, distorting onboarding, and quietly steering clinical conversations. It was clear that any business operating in health needs to be far more prepared for what's already out there, in their customers' heads, before the first appointment. Dialogue is what we're building to make that possible.
Health information has never been this loud, or this wrong.
The volume of health information a patient absorbs before they reach a clinician has multiplied in a decade. So has how much of it is wrong. Influencers, generative-AI summaries, podcast guests with audiences in the millions, even mainstream broadcasters rushing the science: the reputable sources are part of the problem now too.
Until recently, reading all of that at scale was impossible. Now an LLM can read every public comment as carefully as a person would, and we can afford to do it. The hard part is no longer reading. It's deciding which corners of the internet to read, and which claims, misconceptions, and confidently-wrong patterns to look for. That's our job.
A patient is not a sentiment score.
A single sentiment score is the wrong unit of measurement. Patients aren't positive or negative about a drug. They're hopeful about the effect, frightened about the side effects, sceptical of the manufacturer, exhausted by previous attempts, and embarrassed about needing it. All at once.
We hand all of that over, comment by comment, so the person who meets that patient already knows what they've been told.
Want to talk?
We're a small team in London. If you run a clinical service, a pharma brand, or a health-tech product, and you want to know what your patients already think, get in touch.
ben@dialogue-ai.co