Growing digital sophistication means that healthcare practitioners (HCPs) are increasingly expecting more tailored and engaging experiences. However, proliferation of channels and touch points is making conventional experience-based engagement with HCPs more challenging: a one-size-fits-all approach will not fulfill HCPs’ expectations, and sales reps’ experience alone is not enough to determine the optimal engagement plan across channels for individual clinicians.
Simultaneously, conventional face-to-face visits to HCPs have reached the limit of their effectiveness: investing in more face-to-face visits is costly and may not effectively contribute to disease awareness or more appropriate product use for the right patients. In response, pharma company players around the world have started to develop a vision to transform their HCP-interaction models via digital and advanced analytics and many have developed initial solutions for select products in priority markets. Recently, the COVID-19 pandemic has provided further impetus for companies to transform their commercial models.
Additionally, there are numerous misconceptions around digital, analytics, and omnichannel. Many industry players remain unconvinced by the urgency of the situation and are unsure of the feasibility and impact of applying digital and analytics in transforming their commercial model. Indeed, from our discussions with pharma leaders in Asia, we commonly hear five key questions asked in relation to adopting an analytics-enabled omnichannel commercial model (Exhibit 1):
- Why does it matter, especially in Asia where the in-person relationship would likely remain essential?
- What data do we need and how does the new approach differ from what we have done before?
- What are the potential insights from the new commercial model and use cases?
- What sort of talent do we need to develop this model?
- How do we ensure adoption and trust in the new model across the organization?
In this article, we aim to demystify the concept of an analytics-enabled omnichannel commercial model by answering these five key questions. Further, we share the steps that pharma companies in Asia need to take to transform their commercial model and the related stakeholder experience using advance analytics.
Why does it matter: An analytics-enabled omnichannel commercial model can create value; companies should start now
Transforming engagement experiences by adopting analytics in commercial models and engaging stakeholders across multiple channels has become increasingly important. At the same time, conventional access to HCPs has become more challenging as the COVID-19 pandemic has accelerated this trend.
Sectors such as media, retail, and banking have benefitted from using advanced analytics to improve customer engagement with data positioned as a strategic asset. In Asia, analytics-enabled customer engagement has been changing consumers’ daily lives: for example, tech players are enabling seamless online/offline shopping experiences by diversifying customer touchpoints and enriching customer journeys; mobile payment apps are replacing cash and regular trips to the bank; the daily push of personalized content in the form of news and blog updates via people’s phones is replacing conventional media.
While pharma lags other industries, channels to engage HCPs are proliferating, and the line between online and offline engagement is rapidly blurring. Being accustomed to the variety of digital channels and information outside their time in patient care, HCPs are increasingly expecting content from pharma companies that is directly relevant and tailored to their practice and patients. Managing this imperative has become more and more overwhelming for sales reps, who have traditionally relied on their “instincts” to build relationships with HCPs; it is becoming impossible even for the most experienced reps to tailor and optimize today’s complex mix of channels, content and frequency of interactions for individual HCPs.
Since the emergence of COVID-19, many hospitals across Asia (as in other parts of the world) no longer allow physical visits by sales reps. Any return of physical engagement with HCPs is expected to be patchy and slow in the near term, limited by the speed of vaccination rollout and the continually mutating virus. As a result, the industry is calling for a reinvention of stakeholder-relationship and interaction models. Historically, many companies have a vision for how they might transform their engagement model. Yet, all too frequently, such plans have suffered from “pilot paralysis” or a lack of willingness to risk near-term sales disruption to achieve long-term benefits. Today, transforming the existing commercial model is considered both inevitable and urgent (Exhibit 2).
Pioneers that have adopted analytics and omnichannel approaches as part of their commercial model have garnered significant rewards. Globally, we have observed that, when implemented well, commercial transformations can create 5–10 percent revenue uplift, a 10–20 percent increase in marketing efficiencies and cost savings, a 3–5 percent increase in prescribers, and 5–10 percent higher HCP satisfaction (Exhibit 3). These improvements are driven by the differentiated insights enabled by advanced-analytics that help guide commercial activities: examples include the creation of personalized messages and channels for individual HCPs, and the allocation of resources for greatest impact. The omnichannel commercial model would fundamentally upgrade ways of working, including skills adapted for a more tech-enabled world. HCP experience would also be transformed as personalized engagement would lead to more meaningful interactions. Importantly, continuous iteration with feedback from the market strengthens the predictive and optimization power of the analytics system.
Inevitably, building the infrastructure and capability required for a successful analytics-enabled omnichannel commercial model takes time and effort while companies also face a shortage of digital and analytics talent. Moreover, changing the way marketers and sales reps work is not a trivial undertaking. For pharma players, the time to act is now—being late to the game could mean a rapidly widening gap with peers that becomes more and more difficult to close.
What data do we need: Data can be a challenge, but most have sufficient data to get started
Most industry players have already started accumulating product data, HCP characteristics, details of HCP interactions, content, sales data, and market-context data (including demographics and information on the competitive landscape) via their CRM system, market-research surveys, and through third parties. In the development of an analytics-enabled omnichannel commercial model, these very same datasets can be sourced and linked to form a 360-degree view of HCPs such as who and where they are, their market context, how the company has been interacting with them, and how successful these interactions have been. (Exhibit 4). Analyzing such data allows the company to derive insights and define what are most impactful touch points with individual HCPs in determining their prescribing preferences and ensuring the right treatments reach the right patients at the right time. These insights can in turn drive improvements in HCP prioritization, channel mix, product positioning and messaging specific to individual HCPs.
From our experience, most players in Asia have sufficient data to get started for at least some, if not all, products. Companies need not wait to have best-in-class data quality in all dimensions to get started. Data quality can be assessed across several dimensions, including HCP coverage, granularity, duration and frequency, linkability, and production readiness. As long as the minimal data criteria are met for select products, organizations can proceed with analyses and then continuously optimize data quality over time to improve accuracy. Six data criteria, and at least a year of data in many cases, are necessary for meaningful analytics:
- Sales data for each HCP or to a small group of HCPs by product. This enables estimation of sales at the individual HCP level and thereby analytics modeling of sales impact by changing engagements.
- HCP interactions data specific to the HCP and channels. This enables frequency and channel recommendations.
- Messaging/content data also specific to the HCP and channels and linked to the interaction data. This enables content modeling and associated business recommendations.
- HCP characteristics, including basic information of HCP, penetration and patient estimate preferably at brand or indication level. This enables estimation of sales at individual HCP level, as well as micro-segmentation of HCPs.
- Product data that matches the HCP universe and can be mapped across interactions and sales.
- Market data such as competitive landscape, market access, market events, epidemiology, which can provide additional context, enabling richer insights.
When collecting and managing various types of data, it is essential to follow regulatory and compliance requirements of individual countries. The rules would likely vary by country and are important considerations when tailoring a data strategy.
What are the potential insights and use cases: Analytics can transform the HCP experience to enable highest-impact engagements
Through linking and analyzing multiple datasets, advanced analytics enables four levels of insights that deepen understanding of HCPs and can transform the engagement experience:
1. Descriptive/explanatory analyses of markets—answering what has happened in the past, and why. Specifically, mining historical data with advanced analytics can enable micro-segmentation of HCPs across channels (versus conventional survey-based attitudinal and demographic segmentation). This allows a deeper understanding of HCPs and their journeys and guides personalization of content and messages that can help HCPs serve their patients better.
2. Predictions on initiation potential, growth potential, and impact of interactions—answering what is likely to happen in future through machine learning–powered predictive analytics. Importantly, predictive performance can be boosted over time through data updates. Specifically, for an individual HCP or a microsegment of HCPs, the model can help predict a variety of scenarios at specific moments in time, such as the likelihood to initiate first patient, expect increased patient numbers with the prescription of interest, respond positively to reps’ face-to-face visits that are focused on benefits for a particular patient population, stem declining share, or start prescribing drug in the next six months. Such insights guide optimization of sales-force deployment across HCP segments and prioritization of HCPs within a territory.
3. Agile content and delivery optimization—content personalization at the HCP level, building on insights from HCP micro-segmentation and predictive modeling, plus real-time A/B testing for optimization of content and the form of delivery, and rapid deployment across channels (Exhibit 5). This can enable an impactful “test and learn” approach. The effectiveness of content delivery is evaluated in real time and refined as fast as every one to two weeks through A/B testing in the field. The most recent interactions are also taken into account to shape recommendations in a fast-evolving environment.
4. Optimization—addressing what could be done to influence the future. Specifically, analytics simulates the impact of different interactions for each HCP or HCP segment, and it develops an optimized sequence of interactions for each HCP or segment to maximize the predicted sales return based on a total budget of sales-force resources (Exhibit 6). For example, analysis could show that inviting HCP A to an upcoming peer-to-peer program with other HCPs in the shared patient network, then paying two visits to HCP A within the next two weeks using “efficacy message sequence A”, followed by a further visit in the following two weeks using “safety message B” would provide the more meaningful engagement. In other words, advanced analytics enables optimization of HCP engagement for the right HCP, with the right frequency, using the right channels, and with the right messaging in real time via sales-force activities that can maximize impact, while also applying constraints to reflect real-world feasibility and brand priorities. Additionally, sales reps would need to continuously provide feedback and updates to the system about the outcome of their interactions and changes in HCP characteristics so as to keep updating the system to improve its recommendations over time.
The power of an analytics-enabled omnichannel model in practice
Pioneering pharma companies in Europe, the United States, and Asia are starting to use analytics-enabled omnichannel commercial models to address a variety of business needs.
Case 1: Uncovering new opportunities to drive new growth—increasing commercial effectiveness through AI-driven segmentation and improved HCP prioritization
In Germany, a major global pharma company wanted to accelerate growth of an immunology biologic that had already been in the market for over eight years. The company pooled 14 different datasets and developed a methodology that could align sales to individual HCPs, enabling physician segmentation and prioritization at the individual level. Through analytics, the company developed a new predictive method to identify HCPs who are prescribing immunology biologics and gain a quantitative understanding of patient treatment decision – and the implications for product messaging—to then uncover new segments of HCPs that would be interested in immunology biologics for their patients. This effort has led to 30–40 percent of calls being reallocated and 7–15 percent estimated increased interest in prescribing the biologic. Additionally, the company has continued its efforts by replicating analyses across other EU markets, developing scripts for rapid ramp-up of the analytics cloud environment, standardizing data engineering and analytics code, mobilizing and training internal project teams, and codifying the optimum engagement experiences.
Case 2: Accelerating awareness and adoption for new launches with precise initiation—predicting at individual HCP level the likelihood of initiation so as to guide launch planning
A multinational pharma company in Japan launched an analytics-enabled omnichannel initiative designed to accelerate awareness and adoption for an upcoming new-indication launch of a primary-care drug. The company linked multiple datasets, including market-research surveys, CRM, and third-party data, to create a 360-degree view of HCPs with more than 100 features included. Advanced-analytics models were developed to predict at individual HCP level the activities most likely to drive initiation and growth, as well as the likelihood of initiation and expected prescriptions. These efforts identified some 8,000 HCPs with the highest likelihood of prescribing at launch, of which approximately 30 percent were initially classified as not being early adopters. The analytics model also generated additional insights to facilitate launch planning: for example, at least two face-to-face visits a month are needed for initiation, while explanatory meetings are most effective early in the launch. Through optimization of face-to-face visits, the company achieved an estimated 7 percent uplift in patient coverage.
What sort of talent do we need: Talent shortages can be challenging but an initial project team can be set up using global and external support
To drive analytics-enabled commercial transformations, the following roles are typically required locally: product owner, translator, data scientist, data engineer, change manager, together with functional representatives such as a marketer, sales director, and medical adviser (Exhibit 7). Some of these roles can be covered internally using existing roles, such as product owners, while some require significant upskilling, leveraging global/ regional support, or hiring externally, before establishing them in-house.
A shortage of digital talent has been a major concern for pharma companies involved in analytics-enabled commercial transformations, especially in Asia. According to a LinkedIn analysis of data on their platform over the period from 2016 to 2019 (“APEC Closing the Digital Skills Gap Report” 2020), the digital talent hotspot remains in the US, where average penetration of digital skills is almost twice the overall average for Asian countries. Within the healthcare sector specifically, the US also has the highest digital skills penetration, followed by Singapore. Although these data may be partially shaped by cultural differences in LinkedIn platform usage, the analysis is largely consistent with perceptions of the global digital-talent landscape—job profiles such as data scientist, data engineer, designer, or translator are rarely present in healthcare and life sciences companies. Indeed, hiring tech talent can be especially challenging because data-science or data-engineering talent with healthcare experience is rare, and even more so if local language skills are required. Nevertheless, this should not stop companies from pursuing an analytics-enabled commercial transformation. Capability building can take a step-by-step approach: start from a minimum viable product (MVP) team of four or five internal full-time employees with extensive external and global or regional support; then evolve into a full brand team with new roles gradually established in-house through upskilling and recruiting. Scale up is possible by replicating a similar team structure to more brands within the country. When recruiting tech talent, however, companies should not hesitate to hire data engineers and data scientists without healthcare experience. Given the shortage of digital talent in the healthcare sector, it’s more realistic to hire digital talent and cultivate their healthcare experience on the job working with those who have that experience.
How do we ensure adoption and trust: Buy-in and adoption is critical and can be achieved by a comprehensive change-management effort
Adopting a new way of working across the organization is often the most challenging hurdle for those undergoing a transformation towards an analytics-enabled commercial model. In fact, according to McKinsey’s Global Survey on Digital Transformation, only 16 percent of digital transformations have fully succeeded in achieving their business objectives. The new commercial model challenges many conventional ways of working and requires a comprehensive change-management effort that incorporates four best practices (Exhibit 8)
- A different operating model with much shorter cycle time. Conventionally, a long cycle time (six months plus) is the norm in commercial activities. In contrast, an analytics-enabled commercial model requires an agile way of working with fast decision-making and actions rather than strictly following a plan. This usually involves splitting projects into “sprints” for rapid testing and iteration of content and its delivery, and analytics solution, combined with quick decisions on new engagement plans, and associated resource deployment based on analytics outputs and market responses. Agile decision making with simplified and streamlined processes—for example, reducing the content-approval timeline from months to within a week—is needed to sustain the pace.
- Linking tech and business functions. The analytics-enabled commercial transformation should not be driven solely by the tech team. The project team should be cross-functional comprising both tech members and participants from different business functions to ensure mutual understanding of the analytics methodology and development of actionable insights for the business.
- Understanding and adoption by the field force. The field force represents the last mile of implementation and impact delivery. They can be often skeptical of the insights and recommended actions delivered by advanced analytics—the analysis can often be perceived as a “black box” that is difficult to understand and can also lead to confusion when sales managers have different interpretations of the outputs. Companies that have been successful have been able to shift mindsets and behaviors via the classic influence model: role-modeling, fostering conviction and understanding through proactive communication, reinforcement via formal mechanisms, and capability and skill building. In additional successful companies also loop in seasoned sales managers often and early on as part of the project leadership and closely involve them in insight generation, target setting, and leading communication to the broader set of business stakeholders. The sales managers are also tasked to organize and drive implementation at the front line with the sales force, including engaging some in the role of change agents. Sales reps are the core end-users of the system; therefore, the project team must actively listen to and quickly act on their feedback, in order to quickly adapt to new insights and solutions.
- Governance and implementation ownership. Analytics-enabled commercial transformations often fail when there is no formal implementation oversight by senior commercial leadership after analytics development, and where clear organizational responsibilities are lacking. This situation often leads to limited implementation support, inconsistent or partial implementation of actions leaving value on the table, unclear linkage to performance, and growing resistance to change. Those companies that have been successful have emphasized end-to-end ownership and quantified financial objectives for each analytics and commercial initiative; have dedicated roles to track and report implementation progress and impact across initiatives; and have assigned clear ownership of deliverables from leadership to sales reps. To ensure transparency and accountability, implementation progress, adoption, and financial impact should be tracked from day one with indicators such as qualitative feedback, adherence, interactions with engagement/conversion, model accuracy, and sales tracking. A clear performance-management system should also be in place to incentivize adoption and impact delivery.
The STAR journey for analytics-enabled omnichannel commercial transformation
Companies can follow a four-step STAR journey—set up strategic vision, test, adopt, and replicate—to achieve an analytics-enabled omnichannel commercial transformation (Exhibit 9).
S: Set up the Strategic vision and business objectives.
What is the company trying to optimize with advanced analytics in commercial? The best practices we have observed have clear definition and alignment on the strategic vision and business objectives to pursue. This then results in the right prioritization of the most suitable use cases and expected product-specific solutions to deliver. This is achieved by mapping out the specific plans for growth, such as by reaching more HCPs, developing new indications, maximizing awareness and adoption of new products, standardizing actions of sales reps, improving content via personalization, and optimizing content, message and channel deployment. The scope of the advanced-analytics transformation—including the types of data, modeling approach, the features involved, end-users and user experience/user interface design, talent required—will also vary by the specific use case and priorities. Establishing a clear vision and the business objectives up front guides realistic planning and resourcing.
Example use cases:
A global pharma company wanted to transform its commercial model for the “next normal” in Asia Pacific markets by deploying analytics-enabled omnichannel engagement. Significant efforts were invested to align on the prioritized used cases to be pursued by the analytics team. Improving orchestration of touch points and content personalization were the most common asks across different business units. The alignment on use cases allowed the company to guide scoping of efforts, define product-specific solutions, and ensure resources were in place for the analytics team.
In another scenario, a global pharma company was seeking to transform content, message and channel deployment to HCPs by the local commercial organization and encourage adoption of analytics recommendations. Advanced analytics were applied to develop segmentation of HCPs, identify successful messaging, and develop insights to guide next actions for sales reps, and ultimately, successfully transform conventional experience-based sales and marketing activities to data-driven rapid deployment and optimization.
T: Test with MVP on a carefully chosen product and region scope.
Starting from a value-adding MVP for an individual product or even a single indication within a pilot country in Asia (or even selected regions within that country) can build momentum in the organization by providing a tangible showcase for a data-driven personalized approach to engaging HCPs. Specifically, the choice of product or indication should strike a balance between impact and ease of implementation: for example, a product or indication with high value, large sales force, more customer data available, strong competition with more room to grow share, and flexibility to tailor content and its delivery can be prioritized for MVP development. The rich data involved would simplify development of impactful analytics. However, the selected country or region should be an important market with good data available and strong willingness to change within the local team.
Starting the journey with a small MVP team of four or five individuals put together internally, backed up by extensive external support or support from the global organization, can overcome any talent shortages in the initial stages. The new digital and analytics muscle can gradually grow locally or regionally as the company progresses along the transformation journey.
Example use case:
A global pharma company was initiating analytics-enabled omnichannel commercial transformation across Asia Pacific markets. Several key brands were chosen to pilot in one market and then scale the MVP to the broader field force. This then set up the capabilities needed to further expand the MVP to other products in other markets as well. The market was chosen based on the maturity of current solutions and capabilities and the products due to the potential to achieve considerable financial impact early on so as to ensure sustained momentum to scale beyond. Strategic importance in the portfolio plus the intense competition in the market which drove the need for more effective differentiation were also taken into consideration when selecting the products.
A: Adoption by the full brand team in the pilot country.
Once the minimum viable product (MVP) is up and running, the commercial model can be expanded to the full project team for a flagship brand in the pilot country, with clearly aligned roles and responsibilities and adoption of new ways of working. At this stage, a full transformation of the entire brand organization should be considered, including setting up specific financial objectives; redefining workflow, governance, and implementation ownership from leadership to the front line; conducting change management from line management to front line plus field support to end users; establishing rigorous discipline for relentless execution with dedicated roles to track progress; performance management; and redesigning incentives and performance management to reinforce impact delivery. Capability building should be undertaken in parallel, including upskilling from existing roles in the organization together with any necessary external hiring of top talent for core roles.
Example use case:
Upon completing development of analytics for its flagship products (using various datasets of product data, interaction data, and third-party data in a select Asia Pacific market), a multinational pharma company initiated transformation of its sales organization to adopt the new way of working. A transformation office was set up to orchestrate communication from top management to line managers and the front line. Initiatives to upgrade process, operating model, organizational structure, and governance for the new way of working were developed and implemented with disciplined cadence and rigor. This was perceived by top management as an essential step to institutionalize the capabilities and mindset for the new commercial model after the considerable efforts to develop the underlying analytics function.
R: Replicate and scale to additional brands and countries.
Once full transformation of the flagship brand and full adoption of the new way of working is achieved—with visible impact from the analytics-enabled commercial model—the approach can be replicated to additional brands and countries. A clear rollout roadmap should be in place. Investing effort early in the transformation journey to codify protocols or implementation plans for the general methodology and developing modular codes that can be easily tailored for different business needs, can smooth rollout down the road. Setting up a central analytics team in Asia to work in collaboration with local commercial teams to deploy analytics methodology across countries can also facilitate scale-up within the region. Investment in automation and data management can be done stepwise to ensure initial value capture prior to large upfront investments.
Example use case:
A global pharma company tried to implement global frameworks for omnichannel commercial transformation within local markets but found there were challenges and unexpected resistance. After months of negotiation and alignment efforts with local teams, the company learnt that local countries often have their own omnichannel framework and digital systems—and force-fitting a global framework would not work well. For example, many global digital systems could not run smoothly on the local network or might even violate local compliance regulations. When rolling out omnichannel commercial model in local markets, it is important to conduct an inventory of what is already in place, and then seek to support unmet needs among local teams via the global framework and solutions, as opposed to simply imposing the global framework locally.
Many pharma companies in Asia have most of the ingredients needed to pursue an analytics-enabled, customer-centric omnichannel commercial model—even taking into account variations in data maturity among brands and countries. Because it may take years to realize a full-scale analytics-enabled commercial transformation and capture its full potential, the time to begin the transformation is now.