Data quality and consistency come from a stable survey environment and significance testing built into the dashboard at 95% confidence. Surveys are designed with a focus on respondent experience which means only measuring what matters to keep survey length short (ideally within 10 minutes) and engagement levels high. But even before a survey is designed, questions are tested to ensure respondents understand what is being asked – context is an important consideration. Attention checks such as trap and trick questions are also incorporated to measure engagement levels. Once a survey is in field, quality scoring is applied with more than 100 in-app checks on every response to maintain confidence in the data.

During processing, our data engineers apply a series of manual checks – interrogating any aspects of the data that require closer inspection. Nothing gets reported or goes in the dashboard without our team's complete confidence in the integrity of the data.

And it doesn’t end there. This is an ongoing process of refinement and improvement. As technology becomes more sophisticated, so will our approach.

We have established partnerships with data platforms and panel providers. All partners undergo rigorous due diligence to meet our quality standards, and a partner is selected for its ability to meet a program’s specific requirements. Panels are vetted and we only use the very best samples. Where respondents are removed through quality control checks, they are replaced to maintain sample sizes.