Synthetic Respondents
Reliable responses, modelled from real human
- Human: 164
- Synthetic: 36
- Human: 133
- Synthetic: 17
- Human: 148
- Synthetic: 27
Overview
Complete research objectives in time without compromising quality
Filling hard-to-reach survey quotas while maintaining accuracy and representation is a challenge. Traditional weighting adjusts data but doesn’t generate new responses.
Pureprofile’s Synthetic Responses fill the gap by generating reliable, high-quality responses that reflect the original dataset and seamlessly top up traditional research.
Key benefits
Faster project delivery
Meet project field time deadlines with synthetic responses without sacrificing on quality
Tailored to each dataset
Machine learning models are tailored for each training dataset to accurately reflect real panellist responses
Reliable and representative insights
Generate high-quality synthetic responses that are more reliable and representative than traditional weighting
Fulfil all target quotas
Complete hard-to-fill quotas with relevant synthetic responses, regardless of incidence rate or audience availability
How it works
1
Identify quotas
Pureprofile identifies the quotas that require completing with synthetic responses
2
Prepare the dataset
Human responses with de-identified profile data are prepared and uploaded to the platform
3
Train and predict
The platform analyses the full dataset and runs the training and prediction process
4
Complete the dataset
Synthetic responses are generated to fill the required quotas with validated, reliable data