As finance leaders, CFOs understand how leveraging data can improve efficiency, productivity and the ability to turn insights about the future into action. They also understand how data investments can fuel innovation and increase value for customers. But, as they also understand, recognizing an innovation opportunity is one thing; capitalizing on it is another.
Having the assets, tools and a team of data scientists isn’t enough for success. Countless numbers of companies have invested significantly to develop their data capabilities yet failed to get out of it what they hoped. The challenge is that it is all too easy to become distracted by the bells and whistles of data discovery and lose sight of the point: bringing value to customers and end-users.
Our time in the trenches has shown that the challenges CFOs and their colleagues face developing data products and the infrastructure to support them can be overcome if leaders consider a few factors.
Creating business value
Data science innovation is transformative, but it is not the single most important component in developing data products. You can have a robust analytics platform but still miss the mark. Creating meaningful data products means avoiding common pitfalls that can hamper integration, rollout and successful user experience.
Successful companies know the following:
- Stick to core product management principles as you define the product-market problem you intend to solve. Don’t lead with data science principles.
- Find the problems whose solutions will bring value. Don’t just mine the data for insights. Validate value using proofs of concept (POCs), minimum viable products (MVPs) and pilots, and iterate with users, customers and partners to confirm you’re achieving the target outcomes.
At Medidata, a data analytics platform we worked together at, we found that we achieved traction once we began looking at desired outcomes as those that would solve real and significant business problems for our customers. This became our guiding principle. To that end, we worked with our customers to define, design and refine products and features. We invited feedback early and frequently. We did real-world pilots. And we iterated and incorporated customers’ feedback throughout the process. This is foundational for success.
Practical adoption and use
It starts with the customer. Being customer-centric means asking the right questions to understand the potential impact your product can have across your customer’s entire ecosystem.
For example, is your customer able and willing to integrate data solutions capabilities into their operations? What are their account privacy, security, performance and customer support needs at scale?
Knowing the answers to questions such as these will help companies in their adoption of data products and will ensure that products fit the criteria of their customers. Some helpful tips:
- Consider ease of use, effort needed to adopt and the change management required for use.
- Understand if customers prefer self-service discovery (curated data, tools and insights) or full-serve delivery (suggestions, recommendations and conclusions).
- Meet customers where they are and help transition them to a target future state. This may include training, providing services or third-party partners to drive change management.
Working together, we recognized the criticalness of focusing on the customer experience. The solutions needed to be fit-for-purpose, simplified and intuitive enough for users to embrace and gain value quickly and easily — a key goal of ours. This is often why some companies leverage a services-heavy model in the initial phases of development for analytics offerings; they want to accelerate customer adoption and ensure continuous feedback.
Collaboration and execution teams
Building the right culture will help ensure you’re making the best use of the expertise at your disposal. An open and successful mindset from everyone on the team will lead to higher productivity. To this end, it’s important to —
- Acknowledge that it takes far more than data scientists (though critical) to build, deliver and productize innovation and scale data offerings. It takes data security/privacy, delivery, data system engineering, enterprise and data architecture and product management (engineering if integrating with a core platform).
- Prevent silos by understanding that data scientists are ultimately part of a bigger value chain for delivering products to the market and should be integrated with the product, engineering and the broader technology organization.
- Prioritize a culture of respect, trust, transparency and collaboration. We realized immediate benefits at Medidata when we aligned our data science team to the broader organization with shared outcomes and objectives. Bringing the functional teams together enabled us to get the most out of each team, learn from one another, and achieve bigger and more innovative outcomes. This model resulted in higher quality and more fit-for-purpose capabilities with a more rapid time to market. And happier and engaged employees.
Build for scale
The pitfalls of building for scale are many and avoidable. Companies are more likely to enjoy sustainable growth if they take time to build a framework that meets their desired outcomes.
- Ensure automated operations are in place so that data is “fit-for-purpose” for desired outcomes; have a framework for connecting to new data sources and ongoing data cleaning and curation.
- Adapt the enterprise platform architecture and infrastructure to support the innovation life cycle from research to production to provide a secure, controlled, integrated environment that enables experimentation and delivery at scale.
- As work shifts from ideation to productization, implement an effective SDLC that holds engineering and product to a set of quality controls for code, data, and implementation.
- Implement rigorous accountability systems designed by the cultural values of respect, trust, transparency, and collaboration, and test them often.
At Medidata, we developed an innovation to product process that leveraged and became a part of the core software development life cycle process. We ensured that new products were compatible with the broader company architecture and ecosystem; that they were well-governed and controlled, maintainable, and secure (but still easy to access); and that they had the ability to empower high velocity exploration, discovery, and innovation.
These key factors are gleaned from our years of experience in the trenches seeing what works and what doesn’t. Companies that learn from the successes and failures of others will be well-positioned to capitalize on the data opportunity at their doors.
Authors Iskow and DiGiambattista worked together at Medidata Solutions, as CTO and Head of Data & Analytics, respectively.