Constructing an Analytics Capability Roadmap, Part 1

You have built a pipeline of analytics projects responsive to your company’s needs today, factoring in the returned business value for each. In building that pipeline you successfully reprioritized a number of efforts, including offloading several that provided little business benefit or looked more like requests for a BI reporting team. The revised pipeline accurately reflects your team’s current analytics ability and your goals of driving cost reduction or revenue generation and are aligned with the business. You’ve graded the projects along a complexity scale, from descriptive analysis to embedded applications. You note most of the projects are on the descriptive end of the scale. Although not where you want to be eventually, the chosen projects reflect an accurate representation of your current demand and capabilities. You get sign off on the investment necessary to undertake the work. It will be a busy and productive year. Then, as the steering committee meeting comes to an end, an influential executive asks, “How do we do more? How do we get more advanced?” You now need an analytics capability roadmap, or ACR.

An ACR defines the development of competencies and technologies over time to address known and expected business needs in a future state. IIA designed this framework to help data and analytics leaders think through the transition to advanced analytics and AI. Companies undertaking an ACR effort tend to be early in their analytics maturity, trying to shift the balance of their work beyond descriptive analytics to more predictive or prescriptive projects. For these early maturity companies, we can divide three key areas for consideration:

1. Use cases or projects
2. The underlying platform
3. Team and individual competencies

IIA RAN clients have access to the full ACR framework and supplemental resources here.

In IIA’s experience, team and individual competencies are the hardest to address. Most of the competencies that a good analytics group needs to develop are not actually technical. Instead, many data and analytics professionals are weakest at working with nontechnical people on technical problems or managing the expectations of key stakeholders.

Collectively, these skills enable your team to both shape and meet the expectations of the demand-side constituencies within your enterprise’s information economy. Your team will then be in position to [a] effectively set internal client expectations, [b] keep your internal clients appropriately involved in the project, and [c] not only make them happy with what you have given them but convert them to advocates, to brand ambassadors.

We suggest looking at the effort of an ACR over a two to three-year horizon. This timeframe allows for strategic planning without delving too deeply into tactical execution. The goal is to execute a strategy that shifts the demand pattern in the user base for analytics. In year one, the focus is on selling the capability to skeptical constituents. By year two, these constituents should become believers and advocates, exerting influence elsewhere. Year three aims to document more business-valuable demand than can be satisfied, indicating a successful shift in analytics demand.

The Use Cases

For companies in early stages of analytics maturity developing ACRs, it's common to start with basic data availability or first-order BI use cases to address immediate needs. The goal in the first year is to transition toward more advanced analytics like predictive and prescriptive, ultimately aiming for automated and embedded analytics. If there are no advanced analytics projects underway by the end of the first year, it may indicate issues with the ACR or a significant gap in BI sufficiency.

Managing use case expectations:

In the first year of managing use case expectations, the focus should be on achieving a balance between basic BI use cases (80%) and more advanced analytics (20%). By the end of three years, the goal is to shift toward 35% to 40% of project work being predictive, prescriptive, or embedded analytics, supported by appropriate platforms and capabilities. Transitioning to 80% advanced analytics, particularly for early-stage analytics maturity companies, is typically a five-year endeavor, influenced by factors like data availability, decision-making skills, and the ability to derive insights from analytics. To succeed in year one, efforts should concentrate on reducing reactive reporting tasks, enabling direct data access for business users, and refocusing the team on driving advanced use cases. This transition also involves developing competencies to identify and articulate advanced use cases, often facilitated by roles like a "catalyst" who bridges the gap between technology and business, shaping opportunities with business groups and presenting them to the analytics team with clear business benefits.

Selecting use cases:

In the first year of your ACR, prioritize a few advanced analytics projects to showcase their value and gain support from stakeholders. Focus on delivering these projects exceptionally well to establish them as reference implementations. Emphasize use cases that can be completed and deliver testimonials from beneficiaries. Additionally, tackle one multiyear problem, highlighting its long-term value and starting work early. Manage the risk of project sizing carefully, ensuring commitments are realistic and achievable. Initially, prioritize short-cycle projects, then aim for a balance between short and long-cycle projects in the second year. By the third year, successful execution should result in more projects than resources, necessitating careful project prioritization.

Creating leverage with use cases:

When selecting early use cases, avoid solely focusing on quick wins that may lack broader appeal. Consider the potential impact across different business functions; a use case may be valuable to sales, but its relevance and value to manufacturing or product design should also be assessed. Prioritize use cases that not only solve difficult data problems but also resonate with and win over skeptical stakeholders. For instance, an insurance company faced skepticism over accessing and utilizing seven years' worth of call center audio recordings. By successfully incorporating this data into their analytical environment, they could dispel doubts, demonstrate the feasibility of complex data integration, and inspire confidence and enthusiasm among stakeholders. The key is to tackle projects that not only deliver results but also address ingrained organizational beliefs about what is possible with data analytics, thus fostering a culture of innovation and demand for analytical solutions.

The Underlying Platform

Evolving the analytics platform within an ACR presents distinct challenges. While some well-funded companies can strategically invest in a modern data lake platform for advanced analytics, most organizations must build their platforms incrementally, often creating redundant data sets and technological inconsistencies in the process. This piecemeal approach can lead to a fragmented platform that becomes increasingly convoluted with each new use case.

IIA emphasizes the importance of defining a target architecture for the data lake early on, providing a vision for the optimal analytics platform. Evaluating use cases based on the reusability of technology components can help prioritize those that enhance the platform's functionality for future use cases. By aligning use cases with platform evolution, organizations can ensure that their platform architecture evolves to meet the needs of complex, high-leverage use cases, ultimately leading to faster time-to-market for analytical solutions.

Individual Competencies

Over the three-year span of an ACR, focusing on human competencies can help divide activities between business engagement and internal team development. While technical skills are important, the emphasis should primarily be on emotional intelligence (EQ). Analysts should be able to recognize opportunities and effectively communicate them to business colleagues, delivering on these insights effectively. The development of competencies can be structured across the years: year one for relationship-building and understanding demand, year two for creating a Community of Practice (CoP) to leverage skills and knowledge, and year three for establishing a Center of Excellence (CoE) to streamline project execution and drive progress.

Stay tuned for part 2 in this series, where we take a closer look at how capabilities evolve from year one to year three and the work you need to do to get there.