Anti-Personas: Writing Down Who You Are Not Building For, and Meaning It
Every roadmap argument has a ghost at the table: the customer who would love this feature, who is not your customer. Anti-personas give that ghost a name and a polite escort out of the room. An anti-persona is a documented profile of who you are deliberately not serving, built from churn, support cost, and failed deal data. Most teams have the data. Few write it down.
What an anti-persona is built from
Start with the expensive evidence. Customers who churned in the first period despite onboarding effort, accounts whose support load ran multiples of the median, deals that closed and then consumed the roadmap for a year. Code what they share: company profile, use case shape, expectation pattern. That composite, named plainly, is the anti-persona.
The pattern is usually a mismatch of fit, not quality. A self serve product keeps attracting enterprise buyers who need audit trails and procurement processes it will never have. A deep workflow tool keeps attracting casual users who need three clicks and a template. Both groups churn angry, and both were honestly sold a product built for someone else.
The decisions it is supposed to change
An anti-persona earns its keep in three rooms. In marketing, it kills targeting and keywords that reliably attract the wrong fit, however cheap the clicks look. In sales, it legitimises the early disqualification that reps currently feel guilty about, converting a lost deal today into saved quarters of support and churn work.
In product, it ends a specific kind of feature debate. When a request maps to the anti-persona and to nobody in the served personas, the answer is no, quickly, without relitigating strategy each sprint. That speed is the point. Teams waste more time on maybe-later features for non customers than on any actual competitor.
Writing it without cruelty or fiction
Name the profile by situation, not by insult: the compliance first enterprise, the one off project user, the discount only switcher. Describe their needs respectfully and state why serving them would break the product for the people it fits. Many anti-personas are excellent customers for somebody else's product, and saying so in the document keeps the exercise honest.
Review it with the same cadence as your personas. Markets move, a segment you cannot serve today can become viable after a pricing tier or an integration ships. An anti-persona is a current decision, not a permanent grudge, and it should carry the data and date it rests on.
The pushback to expect
Someone will always produce the exception: the one anti-persona account that stayed and pays well. Exceptions exist in every segment. Policy is written for the pattern, and the pattern, across churn and cost data, is what the document records. Handle the true exception as an exception, with eyes open about its support cost, rather than letting it veto the category.
Keep reading
Related guides on this site: Persona Confidence Ratings: Labelling What You Know and What You Guessed Personas in Roadmap Reviews: Using Them to Kill Features, Not Decorate Slides Practical answers about personalization.
Frequently asked questions
What is an anti-persona?
A documented profile of the customer type a company deliberately chooses not to target or serve, based on evidence like churn, support cost, and poor fit. It guides marketing, sales qualification, and product decisions by making exclusions explicit.
How is an anti-persona different from just a bad lead?
A bad lead is an individual. An anti-persona is a recurring pattern across many accounts that share a profile and predictably poor outcomes. Defining the pattern lets teams disqualify early and consistently instead of rediscovering the same misfit one deal at a time.
Does having an anti-persona mean refusing customers?
Usually it means not targeting them and qualifying honestly, rather than refusing outright. Some anti-persona customers will still arrive; the document shapes how they are sold, onboarded, and supported, and whether features get built for their needs.
How often should anti-personas be updated?
Review them alongside regular personas, at least quarterly against churn and support data. A segment that was unprofitable to serve can become viable after product, pricing, or integration changes, so the exclusion should rest on current evidence.