What is a digital twin of a customer?
A digital twin of a customer is a simulation of one real person at a customer, such as their CFO, built from evidence about them: what they’ve said in meetings and email, and what their company is going through. You use it to prepare for and rehearse conversations with that person. It is not the person, and it can be wrong.
Updated 7 November 2026
Twin, persona or composite
| Persona | Composite twin | Customer twin | |
|---|---|---|---|
| Represents | A made-up typical buyer | A role in a market | One specific person |
| Built from | Assumptions and research | Real people in that role | Evidence about that person and their company |
| Good for | Messaging and training basics | Testing a deck or page with a market | Preparing for a meeting with that person |
How a customer twin is built
From the sources a team connects: CRM records, meeting notes and transcripts, email, and the company’s press releases, filings and news. Signals become readings, each marked Sure, Likely or Guess and linked to its source. With nothing in the person’s own words yet, the twin starts as a first read, an educated guess from their role, company and similar people, and deepens as evidence arrives. How it works, step by step.
Its limits
- A twin is a simulation, not the person. It can be wrong, and its readings say how sure they are.
- It is only as good as its evidence. A twin built from summaries is weaker than one built from the person’s own words.
- It supports preparation; it doesn’t make decisions about people (hiring, credit or eligibility).
- Synthetic people don’t replace talking to real customers. They help you walk into that conversation better prepared.
Questions
Is a customer twin the same as a buyer persona?
No. A persona describes a typical buyer; a customer twin simulates one specific person, built from evidence about them.
Can a digital twin replace customer research?
No. It helps you prepare for real conversations, and real meetings check whether it was right. It doesn’t replace asking the customer.
How accurate is a customer twin?
It depends on the evidence behind it. Tacity shows three measures side by side, Coverage, Evidence strength and Track record, rather than one accuracy score, and shows its odds as bands until real outcomes have checked them.
Is it legal to model a real person?
Tacity builds twins only from sources your workspace connects or uploads, keeps every point’s source, never uses twins to make decisions about the person, and erases a person on request.
Practise your next meeting with the people in it.
A 30-minute walkthrough with a live practice call on one of your own accounts.
Priced in US dollars by team size. We’ll quote on the call.