THE 'PLATFORM MINDSET'- HOW ENTERPRISES CAN EVOLVE DATA AND ANALYTICS TO DELIVER BUSINESS VALUE

Companies are fired up to launch new products and services and advanced analytics is their fuel, according to PwC’s recent Global Data and Analytics: Big Decisions survey. Out of the 2100 global executives we surveyed, 61 percent said their companies are somewhat or rarely ‘data driven’ and still rely on descriptive and diagnostic analytics. What makes the difference between the companies that use data and predictive analytics capabilities to drive disruption and those at risk of being driven from the marketplace by those more adept at the data game? I strongly believe that the next generation of value creation through analytics will be fueled by enterprises that innovate and deliver analytics products and solution adopting the ‘platform mindset’—a set of 5 unique traits.

1. Multi-Domain, Integrated Approach

Companies fail to maximize the full potential of analytics when they organize their analytics initiatives, investments and resources as vertical silos around business functions namely sales, marketing, finance, supply chain, human resources, etc. The result is a plethora of tightly coupled analytics solutions that extract the same data sets multiple times and redundant capabilities attained through different technologies/tools and execution teams, which hampers integration. In stark contrast to companies that organize their data and analytics vertically, companies with a “Platform Mindset” take on an integrated, horizontal approach by connecting and coordinating data and analytics across business unit initiatives, investments and resources. For example, in a leading retailer, a cross-domain team of marketing, strategic sourcing and supply chain analytics domain experts coalesce to discuss the supplier-product-customer lifecycle. They ask questions like: how do the strategic suppliers need to be on boarded, managed and assessed? How will this impact the company’s ability to source quality products to serve their customers? What assortment of inventory should we store to improve turnover, reduce operating costs and invest in customer experience?

2. “Outside in” Participation

Besides supporting their internal stakeholders and functions for improving their core business, companies with a “Platform Mindset” think outside the four walls of their enterprise when it comes to value generation activities from its data and analytics assets. They include suppliers, end-users, potential consumers, suppliers to their suppliers, strategic partners and peers from adjacent industries etc. One of our large global financial services clients interacts with a large and diverse set of internal and external data assets: customer, credit transactions, income, preferences, employment, wealth, demographics, credit and lifestyle data. The company not only drove efficiencies in their internal operations with these datasets but also developed a set of data services for their retail partners that delivered insights on customer spending, behaviors, consumption patterns to promote their goods and services. The participating retailer partners paid a service fee and customers had a choice to “opt in” to the service with no charges to receive promotions and coupons from their preferred retailers. The retailers observed a 4x times sales lift and redemption of coupons from subscribing to this data service.

3. Multi-layered Technology Talent Structure

Companies with a “Platform mindset” manage to get their organizational set-up and agile delivery model, right.

Companies with a “Platform mindset” manage to get their organizational set-up and agile delivery model, right


They tend to create a well-integrated and coordinated three-layered analytics technology team structure with clearly delineated decision rights, roles and responsibilities between them.

The 3 layers are:
1) Business Technology team: - These teams are embedded within the business units. They tend to possess domain, data and analysis skills relevant to a specific business function. They are comfortable in developing code and configuring simple components necessary to prepare data and self-service BI processes for business users.

2) Platform Engineering team: - This team is responsible for developing the common software components library. They closely collaborate and coordinate activities with the solution teams within the business units and IT team to define and execute on their component library development roadmap. This group possess core software engineering, SDLC skills and has talent recruited from predominantly software companies, which has shipped and supported a commercial product.

3) IT team: - This is the company’s central information technology team, which has control over the company infrastructure, back office systems, applications, network, security and data. They are a key enabler to both the platform engineering and analytics solutions technology teams to ensure they have the right support.

4. A Next-Generation Platform that is used to deliver analytical products and solutions

Companies with a “Platform mindset” focus on the development of their Next Generation platform with modern, scalable, open source distributed technologies like Hadoop, Spark and in-memory databases as a foundation to deliver analytics across multiple domains to both internal and external stakeholders. The platform must also be developed as a modular, micro-services based components architecture. The importance of modularity and component-based architecture are the bedrock of scalability and robustness for these platforms.

The Next Generation Analytics Platform:

1. Consists of clearly delineated layers that can support the end-to-end lifecycle of information management and analytics activities from data ingestion to data integration to analytics, visualization and applications development. It also includes all the enabling layers like data security, data lineage, metadata management and governance, etc.

2. Consists of a set of core and utility components at each layer implemented as microservices to support multi-domain analytics products and solutions. Core components are assets of high value (e.g. Machine learning algorithms), less volatile in their functionality and proprietary in nature whereas utility components are variable in their functionality and are a commodity. Core components are normally built whereas utility components are bought.

5. Common software components library

Development of a library of common software components that can be assembled on-demand, shared and re-used across the enterprise is a critical for the Next- Generation Platform. The platform governance team must centralize this platform engineering function to ensure that there is consistency and standardization of technology components and no duplication of efforts among various technology teams.

The common software components must:

1) enable the various data and analytics capabilities identified by the various business teams within the enterprise and external stakeholders

2) be meta-data driven and flexible enough to take a standardized set of input parameters and deliver standardized outputs from different analytics solution groups.

3) facilitate seamless integration with external vendor components, leverages open source technologies as much as possible and avoids vendor lock-in and high total cost of ownership.

Many companies may use the turn-key services available with major cloud vendors as a foundation for developing these common components to accelerate development.

Forward thinking enterprises are adopting a platform mindset to move from a vertical, siloed approach that limits data’s value to an integrated, horizontal, platform centric mindset that maximizes data’s potential. Irrespective of industry and size, all companies can capitalize on the business opportunities and enhance firm value by adopting a “Platform Mindset and benefit from improved analytical insights, increased shareholder value and reduced costs.”