Personas
Overview
Respondents are algorithmically grouped into personas based on their historical in-survey behavior and device-level fraud signals. Each respondent is assigned exactly one persona at any given time.
Personas provide context for why a respondent received a particular data trust score and help distinguish between high-quality and low-quality traffic. As additional data is collected, respondents may move between personas. Personas are recalculated every time a data trust score is recalculated, ensuring classifications reflect the most up-to-date information.
Core Personas
We classify respondents into four primary personas:
Incognito Operators
Respondents with a history of using devices that exhibit fraudulent characteristics. This persona is commonly associated with automated traffic, bots, or deliberate fraud activity.
Keyboard Mashers
Respondents with a consistent pattern of poor in-survey behavior. They have been observed multiple times with high failure rates, such as inattentive responses or rushed completion. While these respondents may bypass traditional digital fingerprinting and fraud-detection tools, they consistently produce low-quality survey data.
Newbies
Respondents who are new to our database. Their devices appear clean, but there is insufficient historical data to make a quality determination. These respondents remain in this persona until enough behavioral evidence is collected.
Gold Standard
Respondents with a strong history of high-quality in-survey behavior and clean devices. This persona represents the highest level of trust and data reliability.
Transitional Personas
In addition to the core personas, we use two transitional personas to represent respondents whose data indicates a directional trend but is not yet definitive:
Building Trust
Respondents who are no longer considered Newbies and are trending toward the Gold Standard persona. These individuals generally use clean devices and demonstrate good or neutral historical behavior.
Losing Trust
Respondents who are no longer considered Newbies and are trending toward Keyboard Mashers or Incognito Operators. These respondents may use clean devices or devices with detected failures and exhibit neutral or negative historical behavior.
Unprocessed Personas
When the system cannot accurately determine a persona due to processing limitations, the following persona is assigned:
NONE
Returned when a request could not be fully processed. This occurs during submissions that were too quick, blocked requests, wrapper tampering, or when the system experiences unexpected traffic and cannot return a complete response.
See: Blocked Requests, Could Not Process, Requests Too Fast, Wrapper Tampering