More than ever, customers are interacting with our brands and voicing their opinions. They’re sharing on social media, leaving comments on review sites and talking with others in forums. There’s a goldmine of insights but oftentimes companies feel overwhelmed. Advances in machine learning and AI have been instrumental in sifting through this goldmine of insights. Here are some of the frequently asked questions:
Q: What research questions are best answered with machine learning?
A: At its core, machine learning is a data reduction technique. It can analyze thousands, even millions of records. We commonly recommend machine learning for a company that has broad questions to answer. For example, what needs exist in the marketplace? What insights in our category could we have missed with prior research? What insights can we gather from adjacent categories that might help inform our strategy?
Q: What sources of data work well for machine learning?
A: Large sources of text-based data are the best sources. For example, public data sources could include product reviews on e-commerce sites, product review sites, and online discussion forums. Proprietary data sources could include answers to open-ended survey questions, customer call center or chat data.
Q: How is Applied Marketing Science’s approach to machine learning different from other approaches?
A: First, our approach to machine learning is cutting-edge and proven. Second, our approach combines the power of machine learning and human analysis.
Developed by researchers at MIT, Applied Marketing Science’s approach rests on rigorous academic research. You can read the full academic article about the approach here.
Machine learning works by using an algorithm to identify key insights. First, we train the algorithm to distinguish between informative and uninformative content. The algorithm uses convolutional neural networks (CNN) and clustering to return a sample of that contains an accurate representation of the total number of insights in the dataset. Our algorithm identifies hidden insights that are frequently missed by human analysts. We’ve tested it on a range of categories, from vehicles to fast-moving consumer products.
We combine machine learning with thoughtful human analysis. Machine learning uses algorithms, but algorithms are not a black box. Humans must train the algorithm to identify what’s important and unimportant to each study. With this model, you benefit from what machines and humans do best: the machine is systematic and can sift through a large volume of data, and humans train the machine to focus on the right topics. Once the data reduction is complete, highly trained analysts convert the output of the machine into a full database and insights and can develop a report of its implications.
Our new video breaks down the mystery behind machine learning and explains its applications to market research. Learn how machine learning compares to traditional qualitative research, the method’s unique advantages, and how companies are using it to uncover critical insights.
For more information, check out this helpful resource guide.