The 4 pillars of product excellence
A practical framework for organizations to enable data-driven product management in pharma, banking, healthcare, and other compliance-intensive sectors.
Scroll to read
Table of Contents
In a rush? Download the PDF for later
Download the e-BookIntroduction: The promise and reality of data-driven product management
Behavioral analytics platforms promise transformative insights for digital product teams. Vendors describe “plug-and-play” implementations that deliver near-instant value. The pitch is compelling: instrument your applications and immediately unlock insights that reduce onboarding time, lessen the administrative burden around products or processes, and identify workflow bottlenecks. Behavioral data is the foundation of these insights: signals that reveal how users actually interact with software in real time, from where they click and hesitate to where they abandon a task entirely.
But for enterprises, particularly ones in highly regulated industries such as pharma, banking, or healthcare, the reality is far more complex than what vendors promise.
Whether it’s pharmaceutical companies managing clinical trial workflows, banks orchestrating compliance-heavy financial processes, or healthcare organizations navigating patient data requirements, these organizations face a unique set of challenges that no tool, however sophisticated, can solve on its own. It’s no surprise that digital transformation in pharma has only a 4-11% success rate historically, according to McKinsey research findings.
Consider the typical environment at a large pharmaceutical R&D organization: A single business process may span 20-30 distinct applications. The same user (a clinical research associate, a data manager, a regulatory affairs specialist) must navigate between these systems daily, dealing with highly cognitive workloads involving scientific data, blinded clinical information, and strict regulatory requirements.
What’s more, regulators like the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) require ongoing updates to processes and rules — for example, new reporting forms, added safety checks, revised clinical trial protocols, and enhanced post‑market monitoring — and these requirements vary by region, so companies must track them carefully.
At the same time, the digital product teams building and configuring internal tools cannot easily measure how employees actually use the software in their daily work.
Data-driven product management requires building something far more foundational than a dashboard. If you don't have a core common language and product management standards that drive consistency around what success looks like—around usability, around process, around shared taxonomies and metrics—then every time the business reorganizes or turnover happens, you lose ground.
This paper was developed in collaboration with Pendo and Novo Nordisk and focuses on internal digital product and operational capabilities. It does not address clinical efficacy, patient safety outcomes, or clinical decision making.
Developed by Novo Nordisk's R&D Product Excellence function, the framework includes the four interdependent capabilities that must be established before behavioral analytics investments can deliver meaningful value.
These pillars complement the product operating model established as part of a R&D digital transformation mandate and sequence enablement and structured activation of digital product teams spanning early-stage to late-stage drug development workflows and processes.
While born in pharma, its principles are directly transferable to any organization operating at the intersection of complex processes, heavy regulation, and digital transformation, including in banking, insurance, and public-sector institutions.
The starting point: Why ‘plug and play’ fails
When Novo Nordisk's digital product operating model was established roughly three years ago, the organization was already partway through its digital transformation. An agile transformation had already taken place earlier within the IT organization with best practices on delivery established. And while many new tools and products had been re-configured, built and deployed, the extent of value realization had not yet fully reached its potential.
On the ground, clinical research associates working in affiliates reported significant friction: difficulty navigating across systems, disenchantment with new technology that had been promised as an improvement, and a growing sense that headquarters did not fully understand the day-to-day realities of running clinical trials. At the same time, headquarters had a clear vision for enabling more efficient workflows and reducing administrative burden. Both sides wanted the same thing. The gap was in the “how.”
The organization had access to back-end data and process mining tools, which provided useful signals about task completion times and system performance. But this data alone could not answer the most critical questions: How are people actually working within their workflows? Are employees struggling due to insufficient training, confusing design, or poor processes? Where is the real friction and what’s causing it?
Without the behavioral signals from product analytics that reveal how users interact with software in real time, these questions remained unanswerable. And without the organizational capability to interpret and act on behavioral data, even the best instrumentation would amount to expensive shelfware.
This realization became the founding insight of Novo Nordisk's product excellence function: Selecting the right tool is necessary but not sufficient. Without the organizational capability to act on what the data is telling you, even the best platform will underdeliver.*
Why highly regulated industries face unique barriers to transformation
The challenges Novo Nordisk encountered are not unique to pharma. They are structural features of non-digitally native enterprises or any highly regulated industry:
Multi-application complexity. In pharma, a single end-to-end process such as clinical trial management spans dozens of applications; some third-party SaaS, some homegrown. In banking, a loan origination process may touch equally many systems. The user is the same person, but each application has its own product or implementation team, its own roadmap, and its own definition of success.
Regulatory-mandated change. The FDA, EMA, and other regulatory bodies mandate process changes that vary by region and must be implemented on tight timelines. If you’re in banking, equivalents might include evolving anti-money-laundering regulations and data residency rules. These changes are non-negotiable and constant, which means product teams spend significant capacity on compliance-driven work rather than user-experience improvements.
Process-led culture meeting product-led ambition. Many regulated organizations have deep process expertise and strong process governance. When product-led thinking is introduced—an approach in which the product itself becomes the primary vehicle for delivering value, understanding users, and driving business outcomes—it creates tension. This is not because the goals differ, but because the language, the methods, and the mental models are different. Everyone is trying to optimize end-to-end cycle time and reduce redundancy. But the process organization and the product organization often approach these goals from fundamentally different directions.
Unfamiliarity with behavioral data. Product teams in regulated industries are often staffed by professionals who have deep domain expertise but limited experience with modern-day product management practices where product data, continuous discovery and observable results are used to support decisions and improve product performance and user behaviors.
Asking them to incorporate product data into their daily routines means both increasing their product instrumentation usage and educating them on how product data can support insights that drive decisions that also deliver real value.
Mindset change must happen across all levels—from product leaders to product managers and delivery teams. Business leaders need education on how this data translates into business impact and can focus product leaders towards being more outcome-orientated.
The Novo Nordisk 4 pillar fundamentals
To address these challenges, Novo Nordisk's product excellence function developed 4 fundamental pillars as interdependent capabilities and relevant to its digital product operating model maturity. Together, they form the foundation for data-driven product management practices that work within the context of both regulated environments and established internal procedures. Each pillar addresses a specific organizational gap. Each is necessary, but not individually sufficient, for the whole system to function.
| The fourCapability | What it enables | Key activities |
|---|---|---|
| Product Leadership |
Standards for how product management should operate—roles, responsibilities, accountability, and outcome-focused thinking | Standardized job titles and levels; product OKRs tied to business impact; leadership coaching and skills development |
| Strategic Design |
UX positioned at the intersection of business process and product development to bridge silos and de-risk investments | Design strategists surfacing cross-functional insights; usability-to-productivity mapping; early decision support for investment choices |
| Product Analytics |
Operationalizing behavioral analytics to create a common language and level playing field across product portfolios | Building metrics literacy with behavioral analytics tools; combining front-end, back-end, and process data; linking insights to product OKRs |
| Journey Management |
User journey as a collaboration framework for cross-functional prioritization across complex, multi-application workflows | End-to-end workflow visibility; shared taxonomy between process and product teams; enterprise-level signal aggregation |
Standards for how product management should operate—roles, responsibilities, accountability, and outcome-focused thinking
Standardized job titles and levels; product OKRs tied to business impact; leadership coaching and skills development
UX positioned at the intersection of business process and product development to bridge silos and de-risk investments
Design strategists surfacing cross-functional insights; usability-to-productivity mapping; early decision support for investment choices
Operationalizing behavioral analytics to create a common language and level playing field across product portfolios
Building metrics literacy with behavioral analytics tools; combining front-end, back-end, and process data; linking insights to product OKRs
User journey as a collaboration framework for cross-functional prioritization across complex, multi-application workflows
End-to-end workflow visibility; shared taxonomy between process and product teams; enterprise-level signal aggregation
Pillar 1: Product leadership
The first pillar establishes the standards, roles, and accountability structures that define how product management should work across the organization.
At Novo Nordisk, this began with a practical problem: The organization had a wide variety of job titles, role definitions, and expectations for product managers. Some came from startup backgrounds with strong B2C product instincts; others had grown into the role from within the company's process-heavy culture. Without a common standard, it was impossible to create consistent expectations, measure performance equitably, or build a shared understanding of what good product management looks like.The function thus converged on two primary job titles—product director and product manager—with clear expectations leveling beneath them.
Beyond title standardization, the leadership pillar emphasizes an outcome-focused mindset: namely the connection between product OKRs and business impact. If product teams cannot articulate how their work influences leading indicators such as task completion rates or support a regulatory requirement that proves uptake for target user group, (and surface that information to senior leadership in a way that connects to funding decisions) then analytics data, however rich, remains disconnected from the decisions that matter most.
Pillar 2: Strategic design
The second pillar positions design not as an aesthetic or UI concern, but as a strategic business capability. In Novo Nordisk's framework, strategic designers act as bridge builders between the process organization and the digital product organization. They sit at the intersection of business process, product development, and user experience, synthesizing signals from all three to surface decisions earlier and de-risk investment choices.
This is a meaningful departure from how design functions typically operate. Rather than working as product designers embedded within a single team, strategic designers in this model function as business design strategists. They create visibility across the product and process landscape, helping stakeholders understand where the greatest friction lies and where interventions will have the most impact upstream and downstream.
The strategic design pillar also directly addresses the tension between process-led and product-led thinking. In many regulated organizations, process owners carry deep expertise in compliance requirements and business rules and tend to be the ones driving prioritization decisions. Product managers, particularly those from B2C or startup backgrounds, bring a different perspective rooted in user outcomes and iterative validation. Without a bridging function, these perspectives can create conflict rather than clarity. Strategic design provides the connective tissue. Process owners, for example, may overcomplicate the UI in order to gain compliance; strategic designers help find an elegant solution that addresses both usability and business requirements.
The function also positions design as a de-risking mechanism for large investments, tying usability directly to productivity and connecting both to the chance of success with future automation initiatives. This aspect of the function is critical to gaining executive buy in.
Pillar 3: Product analytics
The third pillar is where behavioral analytics tools like Pendo enter the picture, but only after the organizational groundwork of the first two pillars has begun. Operationalizing product analytics means creating an equitable measurement framework across the entire product portfolio: common metrics, shared definitions, and a standardized approach to instrumentation that allows leadership to compare signals across different products and workflows.
At Novo Nordisk, this entailed deploying behavioral analytics instrumentation alongside a structured learning program. The two had to go hand in hand. Instrumentation without education produced the same result as no instrumentation at all: product teams had access to data but faced challenges on interpretation, the right questions to ask, or how to connect insights to their product OKRs. It also became clear that behavioral data alone tells only part of the story. Usage signals reveal where users struggle, but not why. Pairing quantitative analytics with qualitative feedback (in-app surveys, session-level sentiment, direct input from end users) gives product teams the full picture they need to prioritize with confidence.
The function took a grassroots approach initially, training product teams and hoping adoption would spread organically. The early adopters (roughly 5% of product managers) engaged immediately and enthusiastically. But for the remaining set, adoption stalled.
The learning: The shift towards becoming data-driven is not a nice-to-have; it is a must-have. Embedding this mindset starts with leadership reinforcing a product-forward set of behaviors by establishing and unifying product standards, so that consistency and interpretation of what good looks like make sense across teams. Whilst equipping product managers with instrumentation, leaders must also be equipped with what to expect from their teams as part of the enablement plan.
One important moment came when the team applied Pendo’s product analytics to a single SaaS solution and quantified how user friction maps to lost hours and reduced capacity for a specific cohort. While this doesn’t yet prove realised business value at scale, it links usability and productivity far more tangibly. That insight has shifted perception: instead of treating usability as an anecdote, this product team can now prioritise product work based on measurable impact and target improvements that are likely to protect delivery timelines and support automation and AI efforts.
The next step is embedding a product‑data mindset across product teams so these signals can be measured consistently, investments can be prioritised, and the true value of usability improvements can be demonstrated.
Pillar 4: Journey management
The fourth and most pillar considers journey management as a decision tool and collaboration model in cross-functional decision-making.In a regulated enterprise where a single workflow spans many applications and involves process owners, product managers, architects, and business stakeholders, no single team has a complete view of the user experience. Journey management can fill that gap by bringing a common view of how work actually flows — both as end-to-end user journeys (for example, clinical research associate) and as core business capability journeys that represent specific, completable business needs.
This capability is evolving within the digital product organization as product teams and journey practitioners embed journey thinking into product management, so decisions in uncertain areas and handover points are guided by measurable outcomes rather than separate advisory teams. By treating journeys as the organizing principle, teams can prioritize product work that reduces friction and demonstrate the productivity gains that follow once a product-data mindset is applied at scale.
This pillar establishes a shared language and shared taxonomy that both process and product teams can use. It maps end-to-end workflows across applications. It aggregates behavioral, process, and qualitative data signals along the journey. And it surfaces enterprise-level insights that inform prioritization. Rather than each product team optimizing its own application in isolation, journey management allows the organization to identify where friction in one application creates downstream problems in another, and to prioritize accordingly.
Journey management is also where the framework begins to address a critical gap that many organizations face: the disconnect between product-level improvements and enterprise-level impact. Individual product teams may demonstrate incremental improvements within their own epics, but without a journey-level view, it is difficult to assess whether those improvements translate into meaningful reductions in end-to-end cycle time or administrative burden.
While journey management drives value at enterprise scale, its impact is best measured through improved prioritisation clarity, reduced process friction, and stronger alignment across product and process teams.
Lessons from the field: What practitioners should know
Change does not happen overnight
The timeline for Novo Nordisk to build these capabilities for data-driven product management was measured in years, not quarters. Leaders agreed in principle that behavioral data should inform decisions, but without an established framework, the shift from agreement to action required sustained investment in education, instrumentation, and demonstrated results.
The hard part isn’t importing a one‑size‑fits‑all model — it’s understanding the specific gaps and evolving the digital product operating model to fit the organization’s needs while staying a few steps ahead. As an enablement function, product excellence doesn’t do the work for other teams; it helps guide what comes next and designs the “how” for meeting those needs. Over time, this builds a structured, planned approach to capability building and enablement.
For some organizations, the lesson is practical: transformation timelines can be shortened, but not by imposing a template. Using a clear framework from the start can help establish solid foundations faster — moving progress from years to quarters — if it’s applied as a diagnostic and design tool rather than as a rigid mandate. The most effective change combines two approaches: top‑down governance and standards to create clarity and guardrails, along with bottom‑up early wins that demonstrate tangible value and build momentum.
The prerequisites are not optional
Organizations eager to deploy analytics platforms often want to skip directly to instrumentation. The four-pillar framework makes explicit what many learn the hard way: Without leadership accountability structures, standardized roles, design capabilities that bridge process and product, and a shared journey-level view, analytics investments will not deliver their potential. Together, these foundations enable teams to make faster, evidence‑based decisions by providing clearer signals and more consistent criteria for prioritisation.
Employee efficiency is rooted in the employee experience
One of the more powerful insights from Novo Nordisk's journey is the recognition that operational efficiency and employee experience are not separate goals. When clinical research associates reported frustration with their tools, they were not just flagging a technology problem, but also describing their daily experience of work.
Reducing administrative burden and improving navigation does not just make processes faster; it directly affects how people experience their professional lives. This reframing has proven effective in building stakeholder buy-in, particularly among leaders who may not naturally think in product management terms but who deeply care about workforce engagement and retention.
Cross-industry applicability
Novo Nordisk's product excellence function has actively studied how other regulated industries, particularly banking, address similar challenges. As in pharma, banks operate across multiple applications with complex, compliance-heavy workflows. They face constant regulatory change from bodies like the FCA, SEC, and ECB. And they must balance process governance with the need for digital innovation. The parallels are direct, and the lessons flow both ways.
Practitioners in any heavily regulated industry facing a digital transformation can learn from Novo Nordisk’s experience. Whether deploying new SaaS platforms, consolidating legacy systems, or rolling out AI-powered capabilities, practitioners across industries will find this framework applicable.
Looking ahead: Why this foundation matters for AI
The four-pillar framework is no less relevant as organizations across regulated industries accelerate their adoption of AI, including agentic AI capabilities being embedded into SaaS platforms by vendors.
AI transformation does not exist in a vacuum. Agentic AI tools that automate parts of key workflows, compliance checks, or regulatory submissions still operate within the same complex, multi-application environments described in this paper. They still serve the same users navigating the same end-to-end processes. And their success still depends on the same foundational capabilities.
The same evidence-based foundation that enables effective product decisions is what ensures AI capabilities solve the right problems and integrate without adding new friction. A strategic design team needs to ensure AI enhances rather than complicates the user experience. Product analytics will be needed to understand how employees actually interact with AI-powered features. And journey management will be essential to prioritizing AI investments based on where they will have the greatest end-to-end impact.
Without the behavioral understanding that comes from established product analytics, organizations risk deploying AI capabilities that address the wrong problems or that introduce new friction into already complex processes. The organizations that will realize the greatest ROI from AI investments are those that have already built the capability to observe, measure, and act on how their people work, while also deepening their understanding of current processes to measure efficacy over time.
Conclusion: The path to real transformation
For digital workplace leaders, the message of this paper is straightforward: In highly regulated industries, there is no shortcut to data-driven product management. A plug-and-play approach to analytics does not account for the organizational complexity (multi-application workflows, regulatory mandates, lack of familiarity with product and behavioral data, etc.) that defines these environments.
Driving real transformation requires building organizational capability first plus a commitment to regular scrutiny through both internal and external lenses. Novo Nordisk’s four-pillar framework provides a practical, tested model for how to do this. Orgs will find in this framework both a diagnosis of the challenges they face and a roadmap for how to overcome them.
The work is not easy. It is not fast. But it is the work that makes everything else—be it product analytics, automation, or AI—actually deliver on its promise.
*Telemetry and behavioral analytics were implemented in accordance with Novo Nordisk privacy policies and applicable data protection laws
