Today, organisations in every industry are awash with data, but they may not be taking full advantage of it. David Koke, our senior performance marketing manager, writes about how companies can take advantage of the full insights that are contained within text data for Research Live, the Market Research Society’s news and content site.
David’s article, called “Text analytics needs critical thinking”, covers how standard text analytics approaches tend to miss the critical insights within unstructured data streams by simplifying the analysis to basic reporting on common words and phrases. In truth, text analysis can lead to much more robust and actionable findings – if done right.
In order to get the most out of text data, which gives a window into the “why” of consumer insights (beyond the “what” contained in other data sources), organisations must move on from simple frequency-based analysis. This approach doesn’t always identify the most significant findings and doesn’t put the text data in context.
Because text analytics is perceived to be difficult to analyse in an efficient way, David writes that many insights teams “shy away from this potential goldmine.” He also provides a few examples of how words can be misconstrued and miss illuminating any real, usable data unless critical thinking and context is applied – something that has traditionally been difficult to do at scale.
He then gives a few best practices that can improve the market research industry’s approach to text analytics. First, he maintains that we must start with “good data in.” This can be obtained, for example, through carefully crafted open-ended survey questions that will elicit more robust responses. Starting out with a clear objective, including a deep understanding of stakeholder concerns, before analysis begins can help to better shape this data.
He writes: “Teams need those who understand the benefits of qualitative insights and who can work to promote their value and to ensure a considered approach is taken.”
Of course, efficient and effective text analysis requires better technology, based mainly on AI. While new tools are smarter and more accessible than ever before, it’s hard to weed out the hype from real value. Carefully weighing the capabilities of any new tool, and how it integrates with human talent, is a must do. He says to be wary and ask questions like: “Will this yield high quality results? How well can we train AI to understand the needs of stakeholders?”
David concludes the article with: “Ensure critical thinking and a consideration of context, and leverage text analysis in a way which generates insights which help qualitative research shed any out-dated, legacy associations of ‘finger in the wind’ analysis.”