Case Study: Survey Analysis Automation

Alexandra Strauß, Senior Consultant at Kantar Public
It used to take me several hours to days to do this, now I get all possible variations done for me in 15 minutes!

To drive their decisions, policymakers and business leaders rely more than ever on accurate and timely insights into public opinion. Innovators in the survey and polling industry, such as Kantar Public, are keenly aware of their clients need for faster turnaround times and deeper insights and thus embarked on the bold journey to deliver even more comprehensive results, but within a fraction of the time!

Leveraging the capabilities of Inspirient's Automated Analytics Engine, the Innovation Hub at Kantar Public Germany has succeeded in automating its established survey analytics process end-to-end, from initial validation of panel data to preparing key insights to be shared with their clients. A process that previously would keep an expert-level team busy for weeks can now be run within hours – and even iterated as new panel data comes in!

Automated SPSS survey data to deliverables conversion
Automated SPSS survey data to deliverables conversion

Innovation and Benefits for Kantar Public

In detail, the automation of survey analytics achieved at Kantar Public comprises...

  • Validating and cleaning up gathered survey data, incl. measuring data quality and detecting issues such a speeders, straightliners, and interviewer effects
  • Checking aggregations for significant patterns across relevant cohorts, incl. calculating all contingency tables with significance testing and the compilation of all relevant cross tabulations
  • Prioritizing and executing statistical deep-dives such as network analyses and multivariate regressions with odds ratios and average marginal effects
  • Prioritizing identified results, visualizing key insights, and compiling preview documents for their clients

Usage of Automated Analytics

For the experts at Kantar Public, Inspirient's Automated Analytics Engine generates comprehensive intermediate results that allow them to double-check and prioritize insights, e.g., a summary of all contingency tables with significance testing, outlier detection and other metrics such as Cramer's V. And to support Kantar Public's delivery of insights to their clients, the engine also pre-renders and annotates relevant insights as presentation-ready slides. All automated data processing steps have been fine-tuned to intelligently handle missing data, appropriately weight cases, and carefully document which assumptions and parameters are used in which analysis.

Automatically generated cross tabulations
Pre-rendered annotated aggregations
Automatically generated cross tabulations and pre-rendered annotated aggregations

With this level of automation now in place, Kantar Public no longer needs to differentiate in their workflows between "easy" standard deliverables and "complex" analytical deep-dives. All analyses are pre-rendered automatically if Inspirient’s Automated Analytics Engine deems them appropriate for the input dataset. To ensure that our engine interprets their data correctly at all times, Kantar Public's experts can provide detailed data annotation to guide the analysis. And even better, our engine learns from these annotations over time to get better and better at interpreting new survey data autonomously.


With Inspirient, the experts at Kantar Public can now focus on what matters most: Exceeding their clients' expectations on timely and rigorous analysis, with a curated and carefully contextualized selection of the most impactful insights.

Selection of automatically deduced insights from a real-world survey dataset
Selection of autonomously deduced insights from a real-world survey dataset, the Berkeley IGS Poll
Fully automate the analyis of your survey data and deliver more comprehensive results in less time!

The Inspirient Automated Analytics Engine automates the entire data analytics process end-to-end: From the assignment of input data, pattern and outlier detection, automated visualization of patterns, weak points and opportunities to automatic generation of textual explanations and recognition of the underlying relationships and rules. Most other analytics solutions rarely include these textual explanations and observations regarding the underlying data relations, which are both critical to provide a deeper level of analysis and more actionable conclusions.

Copyright © 2024 by Inspirient GmbH. All rights reserved.