Pureprofile Adds Synthetic Responses Option

In Australia, data and insights company Pureprofile has launched a synthetic data solution promising business the ability to complete response quotas and deliver 'reliable, timely research'.

Pureprofile's research promises 'rich insights into real human behaviour' - the firm builds in-depth profiles of consumers via proprietary and partner panels, helping clients to understand their audiences, target and engage them. In December, it launched an AI-powered video survey capability.

The new Synthetic Responses solution combines advanced machine learning with real survey responses and panellist profile data, and is intended as a top-up solution, according to the company, working alongside traditional panels to quickly deliver reliable, contextually relevant data that reflects the original dataset.

Powered by Livepanel, the solution takes identified quotas to complete, along with real survey responses and de-identified profile data and analyses this dataset via a training and prediction process. Machine learning models are tailored for each training dataset to accurately reflect real panellist responses and ensure they are 'fresh'.

'Synthetic Responses offers a reliable solution that efficiently fills respondent quota gaps' says Chief Product Officer Niamh Fitzpatrick (pictured), 'ensuring comprehensive, representative data collection without compromising quality or exceeding budgets. The speed with which Synthetic Responses fills these quota gaps is also unparalleled. While it once used to take several days to a week to fill a respondent gap, our product can generate synthetic data in as little as 24 hours'.

The company says the new tool is 'especially valuable' for clients concept testing new products or uncovering the opinions of hard-to-reach audiences.

The company was founded in 2000, is based in Surry Hills, NSW, Australia, and is online at www.pureprofile.com .

All articles 2006-23 written and edited by Mel Crowther and/or Nick Thomas, 2024- by Nick Thomas, unless otherwise stated.