Act faster with predictive voice of customer

Use predictive synthetic data to gain granularity when analyzing quantitative surveys.
Header image
1
Map quantitative inputs
2
Identify sub-segments
3
Assess boost coverage
4
Implement new workflows

Rely on predictive data to reduce friction when polling your customer base

Significantly reduce your sample requirements when collecting data by applying predictive augmentation

Smart

Use reliable predictive intelligence that is trained and monitored on your data

Economic

Reduce sample sizes needed to run your ongoing consumer trackers and surveys

Flexible

Works with any workflow, whether in-house, agency or digital panels

Powerful

Analyze segments once considered too hard or slow-to-reach
Technical validation

Internet Service Provider can reduce collection friction and streamline feedback loops

Fairgen's synthetic niche dataset performed better than 2x across segments, questions, and answers — and it was trained on only 16% of the ground truth.
Could Fairgen reliably augment all underrepresented segments?
Yes, Fairgen could reliably augment all 23 niche segments (+2x) using predictive synthetic data.

In order to conclude that, a large dataset was provided as the ground truth (6,240 rows), and we measured the synthetic dataset against it. Fairgen was trained on a fraction of it (1,000 rows), and 23 niche segments of interest were analyzed.

The augmented dataset presented a very low MAE, right below 3%, and the quartiles were close together, showing that Fairgen could remove edge cases.

It performed better than a real sample twice as big as the trained one, showing that Fairgen could effectively predict and augment each niche by at least 2x.
What results did Fairgen provide when zooming in on one sub-segment of interest?
When looking only at one sub-segment of interest, in this case "consumers with very low income", all questions maintained low distances to the ground truth (below 5%), much better than 2x.
What if we zoom in on answers of one specific question?
No problem! We also compared the performance of the augmented dataset on one specific question, of one specific sub-segment.

Here we see that Fairgen could successfully predict, being much closed to the ground truth than the real sample.
What is the business impact?
It was clear that synthetic panels, as long as trained on real data, can be a permanent solution to reduce friction and accelerate time-to-insights.

Insights team decided to reduce weighting variables across all ongoing consumer trackers, trusting Fairgen to support their niche market analysis.

Interested in new use cases?

Fairgen Labs is your ideal partner to co-build products that will modernize research practices. Our team of AI experts are currently working with leading research teams globally.