Marketers love hard numbers – reach and frequency, website visits, multiple-choice questionnaire results – because anything quantifiable helps determine which marketing methods are working. We can crunch numbers into pleasing charts and graphs, and easily manipulate them to tell a truth that matches our preconceived notions.
But when faced with unstructured data – verbatim survey responses, written call center reports and e-mail messages – marketers cringe. Counting things is easy. Gleaning meaning and deciphering substance is daunting. The sheer volume of words makes the task appear insurmountable.
Text analytics strive to derive meaning from the written word. This is massively complex because human communication is so context dependent. Understanding a simple phrase like, “He saw the boy with the telescope,” is tricky. Did he see the boy through the telescope or carrying the telescope?
Junior staff members can read through and code thousands of pages of customer comments to determine whether they are satisfied, if products are performing or if prospective customers are ripe for conversion. Unfortunately, they are the least experienced individuals in the organization and senior staff members simply do not have time to devote to this type of analysis.
Our inability to keep up with the volume of communication is compounded by the rise of social media. Our need to understand conversations taking place out in the sociosphere is critical. Blog posts, tweets, comments on YouTube and more reveal what is on the minds and in the hearts of the public. Absorbing these opinions is not humanly possible, so we turn to technology. Text analytics have been useful for decades and are growing more sophisticated.
Download the white paper Text analytics for social media for a brief description of the underlying technologies from a marketer’s perspective, a quick review of what they do, how they’re now being used in marketing and how they might be used tomorrow.