Your Data Strategy And Why You Need To Consider It

With increasing globalization and virtualization galvanizing modern economics, Digital Transformation and Digital Disruption have become hot topics for boardroom agendas with executives driving for innovative business solutions, cutting edge customer engagement and optimal process efficiency in a market where existing incumbents are facing ever more pressure from digital first organizations, regulatory bodies and tech-savvy customers. Technologies such as Artificial Intelligence, Internet-of-Things, Cloud Platforms and Big Data Analytics provide significant opportunities to alter the way we do business and create value, how you engage with partners and measure up to customer expectations, but what does it take to achieve these goals? The increasing volume, velocity and variety of data that are generated by organizations ends up wasted or underutilized when strategic data direction is ignored, and benefits are left to chance.

Data Strategy should be formulated in support of business strategy and the organizational vision, in line with long term thinking of value creation, market trends and customer needs. In isolation this effort will be seen as bureaucratic political power scheming with the intent of taking the organization hostage and impeding progress and innovation. The alignment between strategic long-term business goals and an enabling Data Strategy must be facilitated by a simple question. Why? Why is Data Governance and Management important to you? Most fundamentalist would point towards data regulations and hygiene as major drivers, and as true as this is, they are but necessary evils, something that the business cannot do without. Data Privacy and Financial Regulatory reporting is part of everyday business and most users are too familiar with overlapping or incorrect data pain points. But we have survived, so why the fuss, why the big hoo-ha now? Well, Data Zealots have always been there, they have just been locked in the basement behind not now-s and we don’t care-s. But their time to engage has come, their time to emerge from the darkness to step into the light wielding their mystical AI swords, holding high their Cloud Enabled Big Data Banners.

Data now drives business, it drives operational efficiency, it drives customer insights and value creation. It drives new services and market segment exploration. We just need to ensure that it aligns and enables where the business is heading so that it supports rather than impedes progress.

(Roe, 2017)

Why Go Through The Pain

What is it that we are trying to achieve with a Digital Strategy and what is it meant to give us if we do it right?

This talk is all good and well, but what does it look like in practice, how will this benefit the organization in achieving its goals, getting closer to the customer and creating value through innovative insights and capabilities? Not only will a robust Data Strategy provide organizations the ability to effectively manage their data assets, vendors and partners, it will create an opportunity to add to the organizational bottom line through rationalizing and optimizing services and capabilities, by providing transparency into process performance and improvements with effective management information and key metric analysis, and provide opportunity to empower employees by facilitating and elevating organizational knowledge management as a first class practice. In support of regulatory and legal requirements, a strategic Data Management program can facilitate compliance and afford organizations the surety that their information assets are catalogued, tracked and tagged, and managed throughout its lifecycle from creation to archival and end-of-life.

By empowering employees to focus on value creation and business outcomes, organizations create a fertile incubator for collaborative innovation within the organizations, and also positions itself for launching new initiatives by working with external stakeholders such as strategic partners and vendors, community platforms, learning-and-research institutions and open platform contributors.

Where To Start

Let us look at some of the steps required to establish the programme:

As with any change and transformation initiation, Establishing Urgency and Buy-In is imperative to gain executive support for the programme. As per Kotter’s Change Model (and many other change management models) this step is crucial in gaining support and creating an appetite for change. If an organization is unaware of the need for change, the burning platform will not be addressed until it is too late.

Once you have the support of executive stakeholders, the next step is to Create a Data Governance and Management Office to oversee the implementation of the Data Management transformation and ongoing governance. This team should be accountable for setting the direction and pace for Strategic Data Initiatives, Supporting People and Technology Capabilities and ongoing management of Operational Oversight and Continuous Improvement.

The scope of data that is important to the organization and who should be responsible for managing the reliability and quality is pivotal in establishing accountability within the organizations. Many organizations entrust this responsibility to their IT departments, but this should ultimately sit with the Business Data Owners and Data Stewards. Identifying Data Domains, Establishing Accountability and driving quality through these channels provides those closest to the source and with the most insight and understanding the autonomy and responsibility to manage the deliverables and steer outcomes within the governance boundaries.

By establishing consistent and repeatable data ingestion patterns, organizations position themselves to respond to the volume and velocity of data generated in digital organizations and customer engagements. Managing data in flight and at rest with security and privacy in mind, and enabling the use of data in its raw forms to extract new wisdom using models without abstracting away potential insights enables organizations to facilitate flexible value creating practices that can react to customer needs and market changes. Data Collection, Storage and Management Capabilities and practices are foundational practices to achieve the reliability that organizations need from well-established data pipelines.

Once you have identified Data Owners, Subject Matter Experts and Data Stewards, the next step is to establish domain dictionaries to describe terms in a singular, clearly defined and enterprise agreed terminology, create transparency in data lineage and consumption to highlight usage, potential risk and impacts of data quality issues, and to formulate rules for Data Quality and Metrics. These rules must be kept with the domain dictionary to establish characteristics and relationships and drive ongoing quality monitoring, management and continuous improvement, and Vendor and Partner engagement on quality and reliability issues.

The DAMA-DMBOK: Data Management Body Of Knowledge provides a comprehensive strategy and framework on structuring a Data Management Practice with guidance on the development, implementation, oversight and monitoring, roadmaps, policies, processes and practices to Implement Data Management Processes and Practices. It provides a structured approach to capabilities and practices through defined goals and principles, roles and responsibilities, activities and techniques, and tools, inputs and outputs that gives insight into establishing requirements.

Establishing a reliable data ingestion and dissemination pipeline is required to enable the organization to derive benefits from its data investments, and to bring this full circle organizations need to move the capabilities closer to those that can glean insights from the data, those that work with it most and can establish value by utilizing startup like funded projects that can prove value through flexible iterative implementations. By Incorporating Citizen Data Scientists and Data Engineers into the units, establishing cross functional business teams aligned with customer value and business outcomes, incorporating continuous improvement into the process, not just localized to siloed departments or teams but to change how end-to-end process and user journeys collaborate with customers and service user, organizations can benefit from their investments into enabling Data-as-a-Service capabilities, placing the power into the hands of those with the most to benefit from collaborations and data availability.

Rules Of The Road

Data Strategy Principles are the premise and guardrails that create the basis for the system of decisioning and setting direction describing the way in which organizations fulfill their Data Transformation initiative. These principles define the rules that reflect and govern the consensus among various stakeholders within the organization and are used to test new decision requirements for alignment. They do not change often and must be contextualized to the organizational maturity and environment. These propositions could take the form of the below, but can be adapted and extended to each organization.

Garbage In, Garbage Out

Right? No.

Data Management is centered around the establishment, execution and oversight of processes, roles and responsibilities and technology in support of the strategic data initiatives and outcomes. This is where the rubber hits the road, where the organizational commitment will be tested. This is where we change culture through institutionalizing Centers of Excellence and best practice, by changing expectations and behaviors through incorporating the roles and responsibilities into the organizational structure and performance management process, and by enabling data management teams and consumers with tools and technologies to facilitate optimal usage of data assets in line with business outcomes.

(DAMA International, 2017)

• Data Governance – the design, planning, oversight and enforcement of policies and standards, practices and procedures, measures and metrics, and strategy and investment
• Data Architecture – strategically position the organization to rapidly extend existing and explore new value propositions using existing assets and the incorporation of new capabilities in line with business strategy
• Data Modeling and Metadata Management – creating a unified data model that accurately describes the data attributes and relationships through the use of visual models, data lineage and data dictionaries that construct and maintain the authoritative organizational and industry view of what data elements mean
• Storage and Security – technical management of storage mechanisms such as databases, file stores, cloud storage, document repositories. Policy management of data elements throughout its lifecycle in line with organizational, industry and regulatory requirements. Privacy and Security management through encryption in flight and at rest, hashing, hiding and contextualized identity and access management of data assets
• Integration and Interoperability – looking at how data is managed and integrated between Systems of Record, Systems of Engagement and Systems of Insights considering merging, migrating, backwards compatibility and vendor and consumer integration
• Document and Content Management – management of information that does not fit into the standard System of Record, but is still a priority for the business lifecycle and outcome in the form of unstructured and semi-structured data
• Reference and Master Data Management – describes the methods, processes and practices that form the basis for managing information that describes the core entities and attributes that drive cross-functional business processes. This is data about your clients and customers, the services and product that you offer and the vendors, partners and materials you use to provide such a service or product. It’s about ensuring that everyone is on the same page, has the latest copy of information that they need, and that the responsibility of managing the process is enabled and allocated to the appropriate data management stakeholders
• Data Warehousing and Business Intelligence – once you have the data pipe running, you need to store the magic somewhere, and allow users to get to the magic. Provides support for operational functions through clear lineage and transparency of information, and the enablement of self-service off the back of well-defined and clearly described information assets
• Data Quality – defining what quality means, how you should measure quality and how you should go about improving on the scores and processes falls within the data governance area that focusses on Reliability and Consistency, Timeliness and Relevance, Completeness and Comprehensiveness, Availability and Accessibility, Accuracy and Precision, Legitimacy and Validity, and Granularity and Uniqueness. This will incorporate defined quality metrics and breach management, Dashboards and Management Information for process oversight and continuous improvement, Remediation and Workflow capabilities, and Vendor Management
• Knowledge Management – combining tacit organizational knowledge, structured learning, and skills and talent management with enabled data first processes enterprises can leverage codified policies and procedures, and guiding practices to enable optimized ways of working

DAMA-DMBOK: Data Management Body Of Knowledge describes these processes in structure detail including Practice Definitions, Goals, Inputs, Actions, Deliverables, Suppliers, Participants, Consumers, and Techniques, Tools and Metrics which provides enough theoretical guidance for constructing your own flavor that suites your organizational maturity and needs.

(DAMA International, 2017)

It’s not just processes and technology

Accountability, Responsibility and Talent Management will play a major role in the success of the transformation effort. Governance Roles such as a Data Office set the strategic direction, ensures accountability while removing obstacles and break down silos; Data Owners take accountability for the quality and ongoing management of data and facilitates strategic initiatives and continuous improvement efforts; Data Stewards manage data on a day to day basis, executing on monitoring and quality issues; Data Scientists and Engineers embedded within the business units provide the flexibility and expertise to design models and solution that prove value which enables the organization to test business hypothesis before committing to longer term programmes. Operationally this requires investment in skills such as Data Modelers and Business Analysts, Data Architects, Database Administrators, Auditors and Information Security Professionals, ETL and Integration Developers, Quality Architects, and DevOps and Site Reliability Engineers.

Lastly, this requires the ongoing support of a Senior Executive Sponsor to ensure that the project stays on track, keeps traction and is aligned with the vision and organizational strategy. It needs to be executed through a programme office that manages the coordination of the programme and facilitates ongoing change management practices to ensure that all considerations such as Risk, Readiness, Engagement and Communication, Competency and Efficacity are taken into account.

We have Big Data, How About You

Big data has literally taken the IT and Data Analytics communities by storm, and the deluge of data and its unmanaged collection has caused the degradation of Data Lakes to Data Swamps. But what does it mean and what are you supposed to do with it?

Let’s first start with how Big Data is loosely defined. Big Data is defined as the practice, process, technology and people that manage large sets of data, received and stored in a variety of formats, generated and consumed at high velocities from batch to real time integration and messaging. In short it is storing, retrieving and managing data received in any format such as file (PDF, CSV, XML, etc.), Database (RDBMS, NO SQL, Document Store, Analytical) and Real Time Integration (API, Message Queues).

The basic model refers to the 3Vs as Volume, Variety and Velocity but more Vs have been added which extends to dimensions such as Value, Veracity, Visualization, Viscosity and Virality.

(m-brain, 2019)

But what is often missed from the hype is that since data is stored in the Data Lake in its raw formats to provide for agility, flexibility and insights, what is not considered is that the information is typically unformatted, unstructured and thus not well documented, understood , clean or necessarily reliable. It requires an in-depth knowledge of the business domain, and data analytical and engineering skillset to engage with this kind of data to analyze and extract valuable insights. It’s not just a point and click, but rather a scratch, dig and uncover.

Here again, the Governance and Management of Data, the alignment with the Organizational Strategic intent and the empowerment and training of Data Scientists and Engineers are imperative to unlock the potential of the Data.

Baseline Architecture, Target Architecture and Roadmap

Once you have established the urgency and momentum, it is down to the next step in your Data Transformation journey. Establishing a baseline or As-Is provides you with the current lay of the land, highlights burning pain points and catalogues existing data assets and processes. When you have established a list of data sources, services and capabilities, and partners, vendors and data stakeholders it creates an opportunity to rationalize and consolidate in line with process optimization, capability centralization and business strategy. This process will quickly illustrate the gap in data quality monitoring and management, the deficiencies in data management processes and governance and the shortcomings in vendor and partner management.

When you have this completed, you can move onto formulating an envisioned target state that accommodates future user needs, establishes architectural principles and best practices, and builds in agility and flexibility while ensuring reliability and quality. The target state provides a blueprint for the organization to work towards, provides a unified view of what is required to get there and creates a centralized point for conversation and transparent decisioning (Desfray & Raymond, 2014).

Transitioning from As-Is to To-Be can be daunting, especially if the programme encapsulates multiple domains, business organizations and implementation teams, but it does not have to be this way. Implementing interim Transitional Architectures affords enterprises the opportunity to test out specific hypothesis while delivering iterative value on their way towards the future state. As opposed to multi-year large scale fixed programmes, agile implementation and delivery programmes offers continuous feedback and value delivery, and flexibility to change direction as the market and industry changes or organizational and customer needs change.

(Shuster, 2014)

(Morrison, 2018)

Your Anchor In The Storm

Establishing objectives for any strategic initiative is imperative to ensure that you have a north star (Aghina, et al., 2018), that you have a set direction to move in and that you have something to hold onto when the going gets tough.

This needs to be something that you can recite in your sleep, and as a transformational leader it is something you should live, breath and act. It should be the WHY to your initiative, the purpose and foundation for embarking on this journey of transformation. Formulating SMART goals, measures and metrics to describe the value-add and ensuring that you regularly measure your progress towards these goals support an agile feedback loop with built in agility to change direction and facilitate new information as the program unfolds.

The magnitude of this transformation can seem daunting, but the benefits in the end will far outweigh the effort.

If we have data, let’s look at data. If all we have are opinions, let’s go with mine.

Jim Barksdale

This forms part of an evolving series exploring Strategic Business and IT Alignment, Digital Enablement and Transformation, IT Governance, Organizational Design and Agility, and Strategic IT Management.

• A picture paints a thousand journeys
• Me, Myself and Mr. Miyagi.
• In Pursuit Of Strategic Business Alignment
• Why Workplace Relationship Management Matters
• A Fascinating Case Of The Conflict Of The Orders
• Your Data Strategy And Why You Need To Consider It
• Commanding The Tension Between Jekyll and Hyde
• A Ghost In The Machine. Robotic Process Automation
• Know Your Triggers. 5 Things That Make You Flip?

References

Aghina, W. et al., 2018. The five trademarks of agile organizations. [Online] Available at: https://www.mckinsey.com/business-functions/organization/our-insights/the-five-trademarks-of-agile-organizations [Accessed Nov 2019].

DAMA International, 2017. DAMA-DMBOK: Data Management Body of Knowledge. 2nd ed. USA: Technics Publications.

Desfray, P. & Raymond, G., 2014. TOGAF?: General Presentation. Modeling enterprise architecture with TOGAF : a practical guide using UML and BPMN, pp. 1-24.

m-brain, 2019. Big Data Technology with 8 V′s. [Online] Available at: https://www.m-brain.com/home/technology/big-data-with-8-vs/ [Accessed Nov 2019].

Morrison, A., 2018. Create a Plan for Establishing a Business-Aligned Data Management Practice Make sure the right information gets to the right people, at the right time. [Online] Available at: https://slideplayer.com/slide/13472462/ [Accessed Nov 2019].

Roe, C., 2017. gds framework. [Online] Available at: https://www.dataversity.net/data-management-vs-data-strategy-a-framework-for-business-success/gds-framework/ [Accessed Nov 2019].

Shuster, L., 2014. Implementing Effective Enterprise Architecture. [Online] Available at: https://www.slideshare.net/LeoShuster/implementing-effective-enterprise-architecture [Accessed Nov 2019].