One sometimes read calls, for a variety of reasons, to remove demographics from personas.
These claims are rarely based on empirically grounded research or analysis.
In truth, persona demographics can be essential for the effective use of personas.
Persona Demographics are Important
One clear example of the importance of persona demographics is Look Alike Analysis.
Look Alike Analysis is when you have identified potential customers but have no behavioral data for them. Lacking actual behavior data, you can use demographic data to identify likely behaviors ahead of time, as demographic data has been shown to correlate with certain customer behaviors.
Examples of Important Demographic Variables
Examples of demographic data that are important for the effective use of personas are:
- Age – certain products and services are targeted (or not targeted) for certain age groups, usually young or older segments. In some cases, products may be illegal to target certain age groups. For example, many products and services are not targeted at minors. Examples: gambling, erotica, alcohol, violent games, and rated movies. Other products are target exclusively at older customers. Examples: housing communities for 55+, retirement plans, and reverse mortgages.
- Cultural – certain forms of dress, food, holidays, and traditions are associated strongly with particular cultural groups. Examples: eating durian fruit, drinking soju, eating/not eating pork, and wearing burkas or face veils
- Income – certain products are outside the reach of certain income levels, regardless of the possible interest of the persons. Example: super expensive homes, premium vacations, and luxury products
- Gender – certain products are targeted to specific gender identities. Examples: bikinis, certain types of make-up, some kinds of dress, types of birth control
In each of these cases and others, the personas SHOULD have the appropriate demographic attribute, as the product or service is targeted or not target for these demographic variables. Also, these demographic attributes are often correlated with other behaviors that may be helpful in the design and development of the primary product or provide insights for gap analysis (i.e., identification of customer segments you are not reaching).
The key is to realize that demographics are a noisy factor — meaning there are individual demographic exceptions and demographic outliers for any product or service!
So, when you do have it, behavioral data trumps demographic data!
In APG Team research, Persona Research from The APG Team , we have found the following concerning persona demographics, persona factors, persona characteristics, and persona attributes:
- Nationality – has a strong correlation with behaviors. The ISO 3166 list of countries is a good proxy attribute for culture or ethnicity.
- Age – age has a medium correlation with behaviors. The US Census age categories list is good for the age attributes, but we find it too granular. We generally combine into three mega-age groupings (younger, middle, older).
- Gender – has a weak correlation with behaviors (it is noisy, but there are trends at times). Part of the ‘noise’ issue may be that most analytics platforms (a) use biological sex as a proxy for gender identity and (b) use gender as a categorical variable when it is a (somewhat) continuous variable. However, there is no other workable measure, at the moment, that we are aware of. So, we generally use Male/Female/Other as a proxy attribute for gender identity.
- Income – We have, as of yet, not investigated the correlation of income with behavior.
Take away Concerning Persona Demographics
In sum, depending on the product or service and the present or lack of behavior data, demographics are a valuable component of your personas.