Unveiling the Core of Data Strategy Aligning with Business Goals
Executive Summary
‘Unveiling the Core of Data Strategy Aligning with Business Goals’ focuses on the critical elements of building a comprehensive data strategy. Emphasising the importance of integrating business architecture in data strategy, Howard Diesel highlights the role of data management, implementation of the Porter model, business capabilities, and enterprise data modelling. He addresses identifying critical data elements and outlines strategies for automating data sharing and integration. Specifically tailored for the healthcare sector, the webinar delves into the significance of value streams in healthcare and the challenges associated with streamlining healthcare processes. Moreover, Howard stresses the need to quantify data value to optimise healthcare operations effectively.
Webinar Details
Title: Unveiling the Core of Data Strategy: Aligning with Business Goals
Date: 14 June 2024
Presenter: Howard Diesel
Write-up Author: Howard Diesel
Contents
Building a Data Strategy.
Importance of Business Architecture in Data Strategy.
Data Management Strategy.
Implementation of Porter Model and Business Architecture.
Business Capability and Data Management
Business Architecture and Subject Area Model
Developing an Enterprise Data Model
Identifying Critical Data Elements.
Strategies for Automating Data Sharing and Integration.
Building a Data Strategy.
Data Models and Healthcare Value Stream.
Value Stream and Model Building in Healthcare.
Value Stream and Optimization Metrics.
Data Modelling and Business Architecture Notes.
Value Streams in a Hospital Setting.
Changing Landscape in Healthcare.
Streamlining Healthcare Processes.
Challenges in Data Problem Solving.
Challenge of Upstream Value Definition.
Importance of Quantifying Data Value.
Building a Data Strategy and Identifying Value Streams.
Building a Data Strategy
Understanding the importance of industry models in aligning a data strategy with use cases, AI, BI, and data product development is critical for connecting with business goals. Howard Diesel, who is presenting, notes that this includes using a healthcare industry reference model to guide the delivery of a data strategy. Additionally, the concepts of data architecture, business architecture, and data value architecture, which come together under the term "data value architect," are significant. Conducting ideation workshops with the business and aligning with business goals, priorities, and strategy is crucial when creating a data strategy. It's also important to have a business strategy and business architecture document before conducting ideation sessions. Finally, conducting a maturity assessment to gauge the capability of fulfilling the business's data initiatives, considering policies, procedures, people, and technology, is essential.
Figure 1 DG Plays - Data Strategy
Importance of Business Architecture in Data Strategy
The assessment of maturity determines technical feasibility, while ideation examines commercial feasibility. Howard focuses on developing a data strategy that aligns with the business and information architecture, as well as the vision, mission, and goals of the business. The business architecture outlines value propositions, drivers, capabilities, and information concepts. Understanding data creation, reading, updating, and deletion is crucial for the data strategy, as is establishing a data estate to identify areas that deliver business value. Aligning with the business architecture and understanding data domains within the data mesh is also essential. Without business architecture, a data strategy may devolve into a mere data management strategy, lacking valuable statements on data quality, governance, and technology.
Data Management Strategy
The data is categorised into four layers: reference data, identity data, transaction data, and behaviour data. A clear vision statement for the data strategy is essential to facilitate prompt and effective decision-making within the business. It's important to note that data management primarily functions as a support service until it can contribute to revenue generation. Aligning the data strategy with the digital and business strategies is crucial for attaining the business's vision and goals. The data strategy's value proposition and delivery capability must be straightforward and focused on supporting the business's objectives. Operational excellence is the organisation's primary value driver, making it the central focus of its data management strategy.
Figure 2 Ensure Business Relevance
Figure 3 Vision Mission Goals & Objectives
Figure 4 Value Driver
Implementation of Porter Model and Business Architecture
The Porter model focuses on secondary activities, IT services, and value generation for stakeholders. The primary data flow is linked to developing industrial cities and technology zones aligned with Industry 4.0, enabling factories to use shared data centres for process control. This reduces the need for individual process control systems. Ministries have seen increased engagement and value delivery within these cities. The business architecture aims to evaluate business capabilities, utilising capability heat mapping to pinpoint areas for improvement. A value stream example is the challenging process of onboarding employees. The capability heat map can also assess business architecture maturity model areas such as onboard tracking, employee information management, and resource matching.
Figure 5 Value Stream
Figure 6 Business Capability Heat Mapping
Business Capability and Data Management
Howard covers various aspects of data management and business capabilities. He emphasises the importance of critically evaluating proposed actions in ideation sessions, documenting business capabilities by identifying challenges, working closely with an architect, and driving value through supporting decision-making and defining value. Howard also discusses different use cases, such as revenue generation, business process optimisation, and business capability enhancement, aiming to deliver workable data products and provide value to customers, as demonstrated by enabling wealthy customers to analyse their portfolio data.
Figure 7 Business Capability Components
Figure 8 Power by Data-Value Driven Way
Business Architecture and Subject Area Model
Wealth portfolio management involves individuals controlling their financial decisions rather than relying solely on wealth managers. Business architecture outlines a company's strategy and value proposition by understanding business capabilities and value streams. Instead of focusing on volatile business processes, the emphasis is on capabilities, data state, and creating value data products. Data architects can create a data model using verbs and nouns related to value actions and objects, aiming for 10-20 terms. A broader definition of terms allows for including multiple aspects in the subject area model. When creating an operating model, data owners should be the owners of the subject area model to ensure their involvement and understanding.
Figure 9 Deriving Informational Concepts
Developing an Enterprise Data Model
The starting point for an Enterprise data model is the business architecture, and it can be created by analysing the capability map and then developing a subject area model or data estate. The subject area model is a key resource for identifying data owners, the data management operating model, stewards, coordinating stewards, domains, and data mesh. Referring to Martin Fowler's domain-driven development is crucial when developing a data strategy. Creating a data estate involves defining reference data, master data, and behavioural aspects of data management. Preparation, such as creating a data estate, is essential before engaging with business users to address their challenges in the realm of data strategy.
Figure 10 Business & Information Architecture
Identifying Critical Data Elements
In the use case ideation process, Howard underscores the significance of identifying critical data elements distinct from a subject area model, which may consist of multiple critical data elements. This identification is crucial for analysing various data sets to develop a data product. Howard also stresses the importance of obtaining data from business architecture and strategy to establish data governance and kick-start data management initiatives. Additionally, he highlights that business requirements and use cases play a vital role in identifying the critical data elements necessary for different capabilities and opportunities.
Figure 12 Data Strategy Reference Data
Figure 13 SWOTs (Data Management Challenges)
Figure 14 SWOTs (Business Requirements)
Strategies for Automating Data Sharing and Integration
Howard notes that the use case involves automating data sharing for a business that previously relied on manual data management. Implementing a cost-effective data integration platform is beneficial for operational excellence. The approach includes identifying value drivers, value stream, capabilities, attention and required data for a comprehensive data management strategy, incorporating business architecture, data architecture, data strategy, data SWAT analysis, action plans, and a roadmap for implementation. The focus is on addressing data problems and friction within the business while ensuring resource and talent availability. A visualisation demonstrates the flow of data from Excel to visualisation and maps the digital strategy to the data strategy across multiple pillars.
Figure 15 Data Strategy Inputs - Architecture Mapping and SWOT List
Figure 16 Business Data Vision Dictates Strategy Composition
Building a Data Strategy
Use cases need to align with data strategy elements, including strategic objectives and linking various components. The execution involves identifying necessary personnel, determining salaries, creating a budget, and developing a project plan and roadmap. The prioritisation process involves business value ranking, data value realisation, implementation feasibility, and use case prioritisation. The communication plan and approval process are essential for socialising the data strategy. The industry reference model for healthcare providers also defines licensed entities providing medical care and treatment.
Figure 17 Building a Data Strategy
Figure 18 Building a Data Strategy continued
Figure 19 Reference Data
Figure 20 Data Strategy Inputs - Prioritisation
Figure 21 Building Data Strategy continued
Data Models and Healthcare Value Stream
The distinction between a hospital as a healthcare provider and a doctor and data models for patients, practitioners, and medical facilities are important considerations. An example of a value stream called "treat condition" outlines steps to admit, treat, and discharge a patient, highlighting the need to measure the effect of changing a patient's condition and the time to discharge. The financial aspect of hospitals, including the importance of patient turnover and occupancy ratio, is crucial. Compliance with statute requirements for patient care, decisions around patient discharge, and transportation arrangements for discharged patients are also significant. Lastly, the value stream emphasises the need for data collection to measure effectiveness and the importance of measurement and analysis in ensuring patient safety and healthcare quality and minimising medical errors.
Figure 22 Healthcare Provider Definition
Figure 23 Major Value Stream
Value Stream and Model Building in Healthcare
It's important to have a well-defined treatment plan, prepare the individual, and execute the treatment to ensure proper care and minimise risk. The ultimate goal is to reduce the risk of readmission while also assessing the efficiency and duration of the care provided. Anticipating the cost of chronic care can be challenging, but utilising reusable components within the architecture for ongoing assessment and chronic care can help. Additionally, both common and specialised value streams in healthcare architecture and including common and specialised healthcare elements in the business architecture are crucial. Lastly, the model should account for specialisation in specific industries to ensure comprehensive coverage.
Figure 24 Capability Map
Value Stream and Optimization Metrics
Howard discusses the concept of value stream and optimisation metrics in the context of dealing with HR stakeholders in a business. He aims to understand the value items and metrics involved in patient care to identify areas for improvement. Key performance indicators (KPIs) such as patient safety, quality care, and risk reduction are crucial in measuring the value of admitting a patient and providing medical care and should be monitored. The ultimate goal of flowing through the value stream is to achieve a final value proposition where the patient's condition changes, and assessing this outcome is essential to determine the effectiveness of care provided. Additionally, unique patient identification in different parts of the healthcare industry is essential and may vary based on industry-specific considerations.
Figure 25 Value Stream
Data Modelling and Business Architecture Notes
Cross-industry models make assumptions and reference master data and identity for insurance purposes. Data professionals build subject area models based on reference data and discuss them with the business to confirm accuracy. A baby's legal and commercial identity is derived from its mother, but that identity is not always carried through to the baby's care. APIs and verb-noun analysis are used to map value items into core business concepts for subject area models, preventing an explosion of data. These models are essential for businesses without a business architecture and can be used to build upon existing structures.
Figure 26 Data Estate Subject Areas
Figure 27 Data Estate, Subject Area Model & Domain
Value Streams in a Hospital Setting
Value streams in a hospital encompass concurrent operations, often intersecting with each other, such as oncology and treatment plans. Managing these intersections involves establishing hierarchies and understanding the stakeholders to deliver value without stretching resources. Each value stream is tailored to specific stakeholders and involves defining, registering, and building subject area models using colour and icons for visual representation in detailed data modelling within the business architecture.
Figure 28 Treat Condition Value Stream
Figure 29 Treat Condition Value Stream continued
Changing Landscape in Healthcare
Howard introduces value-based care, emphasising quality of care and outcomes over fee-for-service, and a PowerBI visualisation showcasing different business and data architecture areas. He also discusses Carevoyance, a tool for enhancing healthcare understanding, and the integration of AI by companies like Definitive Healthcare for patient and practitioner selection and AI analysis of patient data. Furthermore, Howard touches on the behavioural elements in healthcare, particularly in patient care and facility scheduling, and highlights the use of AI in managing patient data, diagnoses, procedures, and facility scheduling.
Figure 30 Valued-Based Care Definition
Figure 31 Power BI Visualisation of Vertical Alignment
Streamlining Healthcare Processes
Carevoyance focuses on integrating various healthcare areas, optimising doctor scheduling for surgeries, maintaining proper occupancy ratios, and ensuring efficient use of expensive medical equipment across hospitals, such as robotic systems. Thus, it enhances administration, scheduling, and admittance processes in healthcare facilities. It plans to leverage PowerBI to visualise and link different stages and value items in healthcare processes. Howard stresses that analysing business architecture yields valuable insights to initiate discussions on streamlining healthcare processes, marking this as only the initial phase.
Figure 32 Streamlining Healthcare Processes
Figure 33 Data Strategy SWOT Analysis
Challenges in Data Problem Solving
The dependency on business documentation is intriguing, as it is the nexus between the value stream stage and the information concept. Engaging in a conversation about the information concept is crucial, even if individuals are unaware that they act as data stewards. The challenge lies in mapping the information concept to the value stream and establishing credibility, confidence, and use cases to highlight inconsistencies. In cases where starting with a value stream map is not viable, dividing the process into parallel tracks and commencing with the information concept may be necessary. The overarching problem encompasses ensuring the completeness of the subject area and data state and gauging the impact of enhancements on the business through relevant metrics. A significant challenge in data problem-solving involves aligning with the business to measure the value delivered and determining the success metrics for data improvements.
Challenge of Upstream Value Definition
Howard notes the importance of prioritising tasks based on perceived value, particularly in low-context businesses, and how convincing the CFO to recognise and credit the team for this value is crucial. Aspirational goals may involve adding more business units in the future, but starting at an information concept level could pose challenges in defining and quantifying the value. Quantifying and monitoring the ROI of data initiatives is essential, and mitigating risk in upstream value definition is crucial when joining projects in progress. While downstream initiatives may seem straightforward, the definition of upstream value remains important.
Importance of Quantifying Data Value
The key point is to identify and quantify the impact of data quality fixes on business operations, measuring the value generated by fixing data quality issues and comparing it to other initiatives. The capacity to discuss data value and delivery is crucial in business conversations, as rushing to build a data warehouse without quantifying its impact on the business can lead to negative feedback. Justifying the value of data initiatives is important to avoid such feedback, making it crucial to define and measure the expected impact of data initiatives.
Building a Data Strategy and Identifying Value Streams
During a discussion, Howard emphasises the significance of understanding a company's value streams and architectures before developing a data strategy. He mentions an Excel spreadsheet from the Business Architecture Guild that contains genericised value streams and architectures across different industries. The primary objective is to identify and measure the value before joining a company with no available architectures. An attendee shares their knowledge of deriving information concepts and linking them to the value chain, specifically in the medical industry. The conversation also focuses on seeking a training reference architecture and finding an industry model for higher education that aligns closely with their work.
If you want to receive the recording, kindly contact Debbie (social@modelwaresystems.com)
Don’t forget to join our exciting LinkedIn and Meetup data communities not to miss out!