GlaxoSmithKline Case Study: Mining Online Discussions for Deeper Customer Insight

Daniel Faggella

Daniel Faggella is Head of Research at Emerj. Called upon by the United Nations, World Bank, INTERPOL, and leading enterprises, Daniel is a globally sought-after expert on the competitive strategy implications of AI for business and government leaders.

block 1.10

Technology Provider: Luminoso

User Company: GlaxoSmithKline (GSK)

Industries: Healthcare

b>Applications: Text and Language Processing

Problem

GSK sought to understand why some parents vaccinate and others do not. Pharmaceutical regulations limit interaction with consumers, so GSK sought a non-intrusive method to gain insight on vaccination concerns.

Actions Taken

GSK used Luminoso’s experience analytics technology to surface insight and sentiment from online parenting discussion boards, including BabyCenter.com and WhattoExpect.com. The company aimed to discern fears and hesitations that parents have with vaccinations for diseases like measles and mumps.

Text analytics were applied to online message boards in order to define meaningful word associations and clusters related to terms – like “safety”, and “comfort” – in addition to sentiments like “happiness” and “unhappiness.” Pharmaceutical regulations make contacting consumers a difficult tasks (much more so than for other baby products, like diapers and baby food), and GSK hoped to glean a more candid perspective on parent’s concerns than might be expressed in an official survey.

Results

Luminoso surfaced major drivers of parent fear from an analysis of the discussion boards, including questions about vaccines and the vaccination process. Major concerns discovered included parental uncertainty about the timing of vaccinations (could they be put off until a child was older?), and a possible fear that vaccinations might cause developmental problems – namely autism. GSK was able to apply these insights to the creation of new educational materials and communications strategies around parental concerns.

Transferable Lessons:

In this case study, we see natural language processing being used for customer insight. Techniques of this kind could be used to search through a company’s own data (customer support tickets, call center transcripts, email records, etc…) or other data sources (discussion forums, blogs, social media, etc…).

In the case study above, GSK uses this approach to improve their customer and prospect communication around a specific product line. In other circumstances, a company might use customer service data to detect prevalent shipping and delivery issues. Hedge funds and financial institutions use strategies of this same kind of glean insight and sentiment around topics that they believe will influence financial markets (such as media headlines about a specific country, election results in foreign countries, etc…).

Stay Ahead of the AI Curve

Discover the critical AI trends and applications that separate winners from losers in the future of business.

Sign up for the 'AI Advantage' newsletter:

Subscribe