Inattentive responding
Patterned or unusually fast responses can weaken the quality of the data.
AUTHENTICITY SCORE
Not all feedbackdeserves thesame weight.
Authenticity Score uses machine learning to assess response reliability, information value, and noise risk across multiple types of feedback data.
Structured responses, response behavior, open-ended feedback, and relevant contextual data are evaluated together rather than in isolation.
Response ReliabilityInformation ValueNoise Risk
AUTHENTICITY SCORE
Structured responses
Response behavior
Open-ended feedback
Context
MACHINE LEARNING
91
- Response Reliability92%
- Information Value88%
- Noise Risk9%
More responses do not always mean better data.
Inconsistent feedback
Ratings, item responses, and open-text comments do not always tell the same story.
Low information value
A completed response can still provide very little useful information for analysis or decision-making.
Authenticity Score evaluates the quality of the feedback, not the person providing it.
HOW IT WORKS
It evaluates multiple types of data together.
Structured responses
Ratings · scale responses · response patterns across items
Response behavior
Completion time · response patterns · consistency
Open-ended feedback
Content · text quality · alignment with structured responses
Context
Relevant experience data · other appropriate data sources
MACHINE LEARNING
AUTHENTICITY SCORE
- Response Reliability
- Information Value
- Noise Risk
Machine learning helps evaluate patterns and relationships across these indicators rather than relying on a single rule or signal.
What does the score tell a manager?
The score is not a final judgment. It is an additional quality indicator that helps managers see which feedback warrants greater confidence and which should be interpreted more cautiously.
Response reliability
How much confidence should be placed in this feedback?
Information value
How much useful information does the response provide for analysis and decision-making?
Noise risk
How likely is the response to contain inattentive, inconsistent, or low-quality patterns?
ACROSS EVERY METRIQORE PLATFORM
Managers see the feedback and how much confidence to place in it.
Authenticity Score is built into the management experience across Happy Guests, Happy Customers, and Happy Students. It gives managers a clear quality signal alongside feedback, helping them see which responses provide stronger evidence and which should be interpreted more cautiously.
Happy Guests
Helps managers interpret the quality and reliability of guest feedback.
Happy Customers
Helps track response quality and noise risk across recurring customer research.
Happy Students
Helps assess the quality of student feedback without labeling individual students.
REAL-WORLD APPLICATION
Once noise becomes visible, data quality can be managed.
45%low-quality feedback
Low-quality or unreliable feedback identified in the original customer pool
93%reliable feedback
Reliable feedback after the panel was rebuilt and ongoing quality monitoring was introduced
A response-quality review for a large fashion retailer found substantial noise in its existing customer pool. The panel was rebuilt and response quality began to be monitored continuously.
Method
It evaluates feedback quality, not whether a person is telling the truth.
Authenticity Score does not decide whether someone is truthful, and it does not draw conclusions from a single indicator. It combines multiple data-quality signals to provide an additional measure of how confidently the feedback can be interpreted.
Facial expression, voice, and movement data are not required inputs to the core Authenticity Score. They may be explored separately in appropriately designed Multimodal Experience Fusion research with the necessary permissions.
LET'S TALK
See the quality behind your feedback, not just the volume.
Tell us how you collect feedback today. We will show you how Authenticity Score helps make response quality visible across Metriqore platforms.
Keep reading
Related pages
Technology
Survey Health Score
Surfaces risks in question quality, scale use and survey length before fieldwork begins.
Explore TechnologyMultimodal Experience Research
Examines surveys and reviews together with environmental, behavioural and operational data.
Explore MethodsScientific Measurement and Data Trust
Psychometrics, data quality controls, multi-signal analysis and explainable management output.
Explore