OR WAIT null SECS
Dan Wetherill offers three priorities for how pharma companies can better use data analytics to engage customers.
Dan Wetherill offers three priorities for how pharmaceutical companies can better use data analytics to engage customers effectively.
Over the past several years, the vast majority of pharmaceutical companies have been investing in improvements to their analytics capabilities. Firms want and need to be better enabled to make data-driven decisions. According to a
recent ZS study conducted by the Economist Intelligence Unit
, companies in all industries are facing similar challenges: 88% of respondents have invested or plan to invest in analytics improvements, and 94% have implemented cloud-based big data infrastructure, yet only 8% have thus far managed to successfully integrate the two. Moreover, while most firms view sales and marketing analytics as “very” or “extremely” important, only 2% of respondents say that their analytics efforts have thus far yielded a “broad, positive impact” for their firms. Companies agree on the importance of successful data analytics but have only succeeded in taking the first steps towards building a more effective capability. According to the research study, pharmaceutical companies face three main challenges in trying to build a leading capability in analytics. The first is the arms race effect: Everyone is investing to improve their analytics performance, which makes it harder for any one company to stand out. One bit of solace is that no industry is really nailing the use of analytics just yet. Pharmaceutical executives often look to retail players or financial services companies as examples of top performers, but our research found that there is no clear industry leader and that other sectors report facing similar analytics hurdles as the pharmaceutical industry. The second challenge is the increasing complexity of the healthcare industry, placing pressure on the analytics ecosystem to keep pace. Payers and providers are consolidating, cost pressures are increasing, and new business models are emerging. For life sciences companies, customers’ needs-and the way in which those customers make purchase decisions-are in constant flux. Similarly, drug pipelines have shifted away from blockbusters, leading to greater competition throughout the pharmaceutical industry in new specialty therapeutics and orphan diseases. However, it’s simply tougher to compete in those categories. They have narrower patient populations, and they require deep expertise of treatment protocols. Companies can still dominate in those areas, but the prize is smaller. The third and potentially most significant challenge is that there’s simply so much data available today-of increasing volume, variety and velocity-that most companies struggle to separate signal from noise. Information streams from many sources, including primary market research, secondary sources, clinical tests, financial reports and more qualitative sources. Assembling information from such a wide range of disparate sources, cleaning it and integrating it all into a single, definitive database-and making sense of the analytics that are based on these databases-can be incredibly tough.
A three-part solution
To improve in this environment, pharmaceutical firms will need to focus on three sets of priorities: their organization, analytics processes and platforms. Regarding the organization, too many firms have data science teams that operate in silos-almost as a back-office function-and thus develop answers and insights that are out of sync with the company’s real problems and realistic solutions. When that happens and business leaders get analytics results that don’t really help them, they lose faith and revert to making decisions the old way: based on gut instinct. To solve this at an organizational level, companies need to break down those silos and foster more direct collaboration between the data scientists (who understand the numbers) and business line executives (who understand what really happens with the end customer). Those are different functions-they have different skills, strengths and objectives-and their leaders often speak different languages. Success will come when analytics and business leaders meet as peers, identify a specific problem and work together to solve it. The second priority area is to improve analytics processes. A key finding from our research is that the analytics functions, themselves, usually aren’t the main problem. The real issues are in how organizations connect the parts of the analytics system, both at the front end (where companies pose questions and decide how to investigate problems using data) and the back end (where they translate results into meaningful actions that change the customer experience in some way). For example, consider a company facing an unexpected decline in sales for a mature product. To understand what’s really going on, the business leader needs to figure out where the losses are the most dramatic, broken down by customer type (both prescribers and patients). It’s a business problem, but it needs to be translated into analytics requirements. Essentially, the problem needs to be framed the right way so that the analytics team can begin to probe for answers and then design solutions based on its findings. Moreover, that kind of translation needs to happen on the back end of the analytics value chain, as well. Once the company develops an analytics-based solution, that solution needs to be contextualized into meaningful actions that change something about the “last mile”: the actual experience that the company delivers to its customers. This is where many companies fall short. Even if they have a strong in-house analytics function that can generate clear insights about the market, they need to apply those insights to improve the customer experience. The third priority area is platforms. Even though most of the companies in our research report are investing in both analytics functions and cloud-based big data infrastructure, only 8% have linked the two. That is an astonishingly low number. Part of the challenge is that implementing big data infrastructure is a major project that takes years to complete. During that time, the company’s priorities can shift dramatically, new data sets can emerge and the overall healthcare environment can evolve. To more effectively link big data infrastructure to the analytics function-in a world of ever-changing data-companies need to create more flexible data infrastructures that can respond with greater agility. As business models among healthcare payers and providers continue to shift, it will become harder for companies to predict what their customer base will look like, or what kind of data they will be able to access to analyze the new industry structure. As a result, companies need to plan for this with the one element that they can control: the structure of their big data platform. To see what improvements in these three elements-organization, analytics processes and platforms-can lead to in the real world, consider an example from the hospitality industry described in our study: Through its customer research, Starwood Hotels found that negative experiences had twice the impact of positive experiences in shaping overall guest loyalty. The company redesigned its customer surveys around this insight to gauge how well it’s meeting customer expectations. It now links social media posts and online reviews from customers to its end-of-stay surveys, and it flags areas that require the company to follow up. That allows Starwood to build detailed profiles summarizing customer preferences, such as whether they prefer sparkling water or would prefer not to be disturbed. Critically, the company created an algorithm that lets it use those insights to improve customer experiences while people are still at its hotels. For example, if someone posts a negative tweet about his stay, the staff at that hotel gets an alert with any relevant information about the customer so that they can fix it in the customer’s preferred way before he checks out. Within a year of adopting analytics across the company, Starwood doubled incremental revenue from targeted guests. The good news for pharmaceutical companies is that this kind of improvement doesn’t require massive investments. Instead, it requires working smart. That means making necessary changes to the organization to put data scientists and business leaders together to collaboratively solve problems. Working smart also requires changes to processes to frame business problems in analytics terms, design solutions, and translate them into actions that improve the customer experience in some meaningful way. Last, working smart requires linking big data infrastructure and analytics platforms in ways that allow firms to stay on top of changes in the industry and emerging data sets. For pharmaceutical companies seeking to build up their analytics capabilities, these steps can help them capitalize on the potential impact of the technology and start using it to improve their performance today.
Dan Wetherill is an associate principal and lead within the analytics process optimization solution area at ZS.