?

Implementing Data Governance for Business Results

Learn how mature data management practices helped one organization better address competitive business needs.

Challenge

A global multi-unit restaurant company was looking to accelerate its digital transformation journey to remain competitive. The client, like many businesses in the franchise model, has long relied on a franchisee-centric data governance to support lines of business (largely focused on reporting). 

The model was not built for scale, lacked federation or data aggregation and contributed to issues such as IT inefficiency and lack of confidence in data quality. The absence of a mature data foundation blocked the client from developing internal data science capabilities, resulting in missed opportunities and loss of market advantage to data-savvy competitors.

Solution

WWT designed and executed a data management strategy to improve analytics efficiency and ultimately accelerate the client’s digital transformation. We also collaborated on an initial data science use case directly tied to adding business value.

Assessed current state

WWT started by evaluating the client’s data maturity based on WWT’s Data Maturity Curve to prioritize specific use cases that deliver data value right from the start. 

data maturity curve

We surveyed and interviewed different lines of business, as well as IT. At the end of that process, we arrived at the following list of pain points that were most common and impactful across the company. 

  • Data infrastructure
    • Separate database instances for each store make consolidation of information challenging
    • Difficulty integrating data between new initiatives and historical systems
    • Lack of a centralized data repository for advanced analytics and data science
  • Data management
    • Low trust in data due to quality issues (e.g., point-in-time report discrepancy for tax rate)
    • Business units work with data in silos and don’t speak the same language because of a lack of shared definitions and process
    • Limited connection between decision making and data (e.g., analytics/data science capability)
    • Need for strategy around data security model for identity and access management, as well as GDPR considerations
  • Data governance
    • Limited data ownership that is not empowered to enable data quality and analysis (e.g., no data stewards, data governance council)
    • Limited strategy around data retention and replication

Built the analytics architecture roadmap

Data architecture specifies which data is collected and how data is stored, arranged, integrated and consumed by different parts of the organization. 

While WWT was designing the data management strategy, the client decided to move to a cloud-hosted data platform. However, the client did not know which cloud components to choose for long-term benefit or how to migrate legacy data systems while minimizing impact on daily business operations.

Our Business and Analytics Advisors team engaged WWT’s cloud subject matter experts (SMEs) to help the client build out multicloud architecture solutions. The team was able to build a phase-by-phase analytics architecture roadmap to help navigate the client through the cloud migration.

Current state

current state
  • Current state architecture faced challenges with user authentication. User identities and access are defined separately at store, franchisee and enterprise levels. 
  • The client's current state architecture also lacks data consistency across systems even at the same level, which is one major reason for low data quality.
Cloud architecture with basic data science capabilities
Cloud architecture with basic data science capabilities
  • Proposed design advances the client's architecture in the cloud with operational data science capabilities using Cloud Spark/Hadoop and Aloud Analysis tools.
  • Third party information such as weather and social media data is acquired, which further supports the client's data science team in predictive analysis.

Defined data governance organizational framework and processes

Data management is not a one-time project – it should be built as a key part of the business’ daily routine. To achieve that, WWT helped the client to define the responsibilities and identify data owners, data stewards, SMEs, a data administration team and a data governance council from lines of business. 

In order to start putting this new framework into practice, WWT also helped establish the data governance processes for several high-impact data use cases.

initial data governance organizational framework

Created business glossary and source mapping

WWT emphasized the urgency of shared data documents, since business departments had little data confidence due to issues such as inconsistent definitions of core business terms (e.g., gross sales, net sales) and limited or a lack of shared understanding on sources of key data elements. Together with data users from different lines of business, WWT created the client’s first Business Glossary and Information Source Mapping.

data documentation comparison

Implemented first data science use case

Given WWT’s success of helping the client on data management strategy, WWT was involved to build the company’s first advanced analytics use case – a real-time product demand prediction model that accounts for complex real-life situations. Our team of data scientists collaborated with the software development team building a store system to ensure seamless integration to the overall environment. 

The model will be deployed in thousands of stores for real-time inventory guidance, which is critical to the client’s business success. It was also the client’s first data science initiative which built data confidence and promoted the data culture across the company. 

product demand prediction model

Impact

The work unlocked data potential with a solid data management strategy, allowing the customer to:

  • empower the business with timely access to credible data supported by a single source of truth for each type of information;
  • inform key technology decisions such as what database types are selected for microservice containers; and
  • promote a culture of fact-based decision making enabled by more accurate analytics in every part of the business.
Immediate value to business
model drives store success
Long-term benefits
client's data science journey

Call to action

This engagement shows how important it is to value and emphasize collaboration between IT and business. Data management is a company-wide effort that will create direct and tangible business benefit while creating basic efficiencies and higher order value for IT. 

Prioritize analytics use cases that can deliver value from the start to get buy-in from different levels of the organization, and continue to iterate and innovate to create a data-driven culture. Develop an actionable roadmap for organizational capabilities needed to execute on analytics use cases.

For more information on how to implement data governance for your organization, request a workshop from one of our data management experts request a workshop with our data management experts.

Share this