Here’s my suggestion:
(a) Before persona creation, form “persona hypotheses” by interviewing company folks like sales people, marketers, and executives.
(b) Capture their tacit knowledge into the persona hypotheses (e.g., “our typical customer is a young woman from Helsinki that loves outdoor experiences”).
(c) Do your study. Interview customers and/or analyze quantitative data. Which of the hypotheses hold?
The point is to make the “buyer of the persona” feel like they’re part of the process. This lowers their shield when later on you come back with your findings based on real data to confirm, refute, or add to their understanding of the customer segments.
Want to see the usefulness of personas in action?
Salminen, J., Jung, S.G., Chowdhury, S. Şengün, S., and Jansen, B. J. (2020) Personas and Analytics: A Comparative User Study of Efficiency and Effectiveness for a User Identification Task. ACM CHI Conference on Human Factors in Computing Systems (CHI’20), Honolulu, HI, USA. 25–30 April, 1-13.