Key Principles

Data Leadership Introduction

[ The following is an excerpt from the book, "Data Leadership: Stop Talking about Data and Start Making an Impact!" ]

Data Leadership is defined as how we choose to apply our limited energy and resources toward creating data capabilities to influence our business. There are so many options for how these energies may be applied, and breaking them down and understanding when and how to apply them is the focus of much of the book. I’m not going to claim it’s easy, but at least Data Leadership activities can be directly controlled.

To make it more tangible, think of Data Leadership as the factory where we build the data sprockets and widgets that our customers will use. We may create data capabilities with a lot of potential value, but until these data capabilities are applied, the value in them is unrealized.

Data Value is the outcome we aim to maximize through Data Leadership, and is the most important consideration for data and technology efforts anywhere in our businesses. This is our North Star—if we can find it, we will find our way.

Think of Data Value as the ultimate measure of the true benefits our data sprockets and widgets give to our customers. This depends not just on what we provide to them, but how well they turn the potential value into real outcome.

Data Value is the difference in business outcomes with data versus what it would achieve without it.

Realize that Data Value is not really about data at all. It is about the positive change in business outcomes we are able to make happen by using data. And just as important is understanding the inverse relationship: if our actions create no positive differential in business outcomes, then we have done nothing of value. In more practical terms, if we deliver a report that fails to change anybody’s behavior (and thus creates no change in business outcome), then how could the real resources we used in creating that report be justified?

Data Leadership influences, but does not control, Data Value.

Data Value is the barometer by which we should judge our Data Leadership performance, and it extends to everyone and everything working with information in our businesses, regardless of whether or not they identify as Data Leaders. It is not enough to have information. It is not even enough to use the information to do something. The information must drive a result with consequences (hopefully positive ones!). Like anything we do with data, maximizing our outcomes begins with measuring.

Data Value is measured in three ways:

1 - Increasing revenue

2 - Decreasing cost

3 - Managing risk

These are the only ways Data Value gets created.

[ end of excerpt ]

Algmin Data Leadership tracks everything we do to Data Value. It's the best way we can validate your investment in us, and it's the only way you'll be successful with data once we're gone. Quite frankly, we're baffled that all consulting organizations don't do this. It's almost as-if they don't want to be held accountable for the outcomes resulting from their advice. Ah, now we get it.

Simple Virtuous Cycle

Get Started Right Now

[ The following is an excerpt from the book, "Data Leadership: Stop Talking about Data and Start Making an Impact!" ]

The Simple Virtuous Cycle reduces the complex topic of creating Data Value into its smallest, most atomic form. This cycle can be found inside most business processes, with multiple instances of varying scales working in harmony. As the world around us gets ever-more complex with new technologies and greater demands for data, we can distill them into this foundational cycle:

1 - Measure

2 - Identify Improvements

3 - Improve

Everything we want to do in the world of data breaks out along these dimensions. The patterns are sadly common across all areas, as well:

Measure: Baseline and understand the complexities of the situation around us. Most of the time today this gets skipped completely.

Identify Improvements: Compile and quantify potential actions, and their range of likely outcomes. Most of the time today people advocate for what they intuitively feel is best and just run with it.

Improve: Implement the optimal actions as determined by the prior steps. Since measurements are a crucial part of the Simple Virtuous Cycle, any improvements should both drive Data Value and accommodate future measurements and improvement cycles. Most of the time today people are so focused on hard deliverables that they fail to prepare for future cycle iterations.

The Simple Virtuous Cycle can operate at large and small scales, but when starting out it is advantageous to try to make the cycles as small as possible. Think simple measurements, limited potential improvement options, easy improvements. Remember, this is about building momentum as much as it is about optimizing outcomes.

Why does this work? It keeps us focused on delivering Data Value at a fast pace. We must keep completing the cycle of value delivery to propel our organization forward. Think of this cycle as the wheels of a car spinning. Large efforts are akin to large wheels, taking longer to complete each revolution. At speed, large wheels may operate more efficiently by covering more ground with each revolution—but they are more costly to build and take more effort to get spinning in the first place.

Since a car can’t practically change its wheel sizes, transmissions exist to change the gearing so that the engine’s power can accelerate a car going different speeds. The more gears a transmission has, the more it can finely tune power delivery. A Continuously Variable Transmission (CVT) effectively does so without limit. The output of a CVT is like infinite variations of wheel sizes on a fixed axle.

This is why we start with small iterations that can get going with limited effort and resources. In Data Leadership we can always increase the size of the efforts over time as things get moving, just like a CVT in an automobile does. The Simple Virtuous Cycle similarly creates a self-reinforcing system that gets stronger (and bigger) as time goes on. When we have enough momentum, the full-fledged Data Leadership Framework will help us get wherever we want to go.

[ end of excerpt ]

Algmin Data Leadership was created, in part, because we were so frustrated by people talking about working and never getting to the things that actually matter. We try to make a measurable difference from Day 1, and that is when the Simple Virtuous Cycle™ is most valuable. If you are getting started trying to Maximize Data Value, whether we are helping directly or not, we encourage you to put the Simple Virtuous Cycle™ to use!

Data Leadership Framework

Balance and Scale Your Efforts

[ The following is an abridged excerpt from the book, "Data Leadership: Stop Talking about Data and Start Making an Impact!" ]

For us, it all comes back to Data Value. If we are using data to create real differentials in business outcomes, then hopefully we are on the right track. Can data be used to manipulate the perceived impacts, or show positive net gains when the reality is far worse? Of course. There’s also data that, if analyzed properly, will help us uncover any misleading conclusions.

The Data Leadership Framework™ was born from this premise, and from observing the rampant ineffectiveness of data efforts currently existing in many of our organizations. These efforts are often incomplete, overly subjective, or completely miss some important aspect that if addressed, would lead to much better outcomes. Companies’ typical project-based change management approach amplifies these problems. With finite timelines, resources, and scope, we compromise our ability to react to new information without laborious change requests or approval processes—if we are lucky enough to gain approval at all. The net result to the project deliverables is cutting corners on things like testing, design, or robustness of the solution—but there is another hidden drawback that is even worse.

If we optimize energies solely at the micro-scale (the project), we lose the ability to maximize impacts at the macro-scale (the system).

This means that absent well-coordinated Data Governance and Project/Program Management, our collections of individual projects will not result in the transformative impacts they were intended to have. It’s as if we spent all of our time polishing grains of sand and then were surprised to learn that we had not created a beach, but instead we had created a desert.

While there are countless resources out there to help us get better at individual Data Management disciplines, there is a startling lack of resources to help us put them all together. The Data Leadership Framework™ is designed to help us keep an eye on the overall transformative impact we’re having, and to help us correct course before we waste a ton of effort building something amazing in the wrong place. Context is what makes data useful, and the Data Leadership Framework™ provides much-needed macro-scale context to the detailed activities we spend most of our time doing.

...

The Data Leadership Framework™ helps us achieve balance between the people, process, technology, and data capabilities we must create to maximize Data Value.

First we break down the universe of things we care about in data into five DLF Categories: Access, Refine, Adopt, Impact, and Align. Each of these have a distinct and significant role, and with a balanced approach to address all of them we will be able to create realized Data Value. The underlying hypothesis is that for the system to operate most effectively, these five categories must be in relative balance—that is, their overall output capacity must be of comparable strength to one another.

Within each of these DLF Categories we have five disciplines representing data and organizational change management functions where we can choose to devote energy (in the form of time, money, attention, etc.). The five DLF Disciplines in each DLF Category do not need to be balanced within an individual category, but the disciplines should be used to compare and prioritize the allocation of finite resources toward the categorically-aligned goal.

This will become clearer as we progress through the categories and disciplines, but for now know that the DLF Categories are outcome-oriented, and the DLF Disciplines are input-oriented. We cannot, for example, simply say we want to devote more resources to the DLF Refinement capabilities without at some point allocating them to specific efforts in Data Quality, Metadata, etc.

DLF Disciplines are actionable, whereas the DLF Categories represent the generalized results of those actions.

Another way to think about it is that the DLF Categories are the conceptual-level that typically resonates with business stakeholders. DLF Disciplines are more Data Management-specific, but also necessarily the level of insight we need to get to if we want to know what to do next. This brings up a good professional tip: always aim to know at least one more level of detail than the questions you expect to receive. It not only ensures you have the requisite mastery of a subject, but it also gives you the ability to answer questions confidently. There is a difference!

The DLF is designed to be an aid in assessing existing environments, developing strategic approaches, and most importantly, helping us know what to do next.

Before we get into the specifics of the DLF, it’s also important to understand that the DLF is a framework. It is intended to make complex subjects simpler for us to evaluate, prioritize, and compare with one another. The DLF does not contain specific answers so much as it helps us ask the right questions to determine what we need to do to create Data Value in our particular context. It is most simply a starting point—one which we are all encouraged to use, adapt, and evolve to meet the needs of our individual organizations.

...

Each category overall, and each discipline within them, individually creates potential value. The category with the lowest Data Value creation capability limits the overall system’s potential throughput. To create the category-level balance it is reasonable to invest more or less energy in individual disciplines, even omitting some entirely at times.

Category-level balance is almost always achieved through a wildly varying approach across DLF Disciplines. Since every organization has some things they do well, and others they do not do well, the right allocation of energy to disciplines will depend much on that.  Additionally, over time the allocations will change as new capabilities are introduced, and the efforts necessary to maintain these capabilities are lesser than that what it took to build them initially.

Data Leaders strive to understand an entity’s capabilities across the DLF Categories and Disciplines, guiding the prioritization and allocation of finite resources to maximize overall system balance and throughput.

Keep this mission in mind as we dive into the specifics, and remember that this is just the beginning. Each of the DLF Disciplines represents a subject area where people devote entire careers. There are certainly plenty of additional resources to learn more, and we should try to gain, at minimum, a foundational understanding in each of them.

[ end of excerpt ]

Compared to the Simple Virtuous Cycle™, the Data Leadership Framework™ is admittedly complex. It's also more than a way to start building momentum. The Data Leadership Framework™ provides the foundation of a comprehensive methodology to solve any organization's data challenges, from small businesses to the largest enterprises.

Categories and Disciplines

Balancing Data Efforts in a Scalable Way

Below is a reference to the key components of the Data Leadership Framework, introduced more completely in our book. We use these Categories and Disciplines to break down the complexity of solving data challenges. Our Data Value Guidebook™ and other offerings are built upon these.

Access

Prepare Data for Use

Data Security


Data Architecture


Data Wrangling


Development


Support, Operations, and DevOps

Refinement

Optimize Data Potential


Metadata


Data Quality


Master Data


Enrichment


Curation

Adoption

Acting from Data Insights


Data Modeling and Warehousing


Traditional Reporting


Interactive Dashboards and Visualizations


Systems Integration


Emerging Data Technologies

Impact

Maximize Business Outcomes


Measurements, Metrics, KPIs


Regression Analysis and Predictive Modeling


Machine Learning and Artificial Intelligence


Business Process Automation


Data Monetization

Alignment

Engage Stakeholders


Strategy, Standards, and Policies


Project and Program Management


Marketing and Communications


Organizational Training and Building Quantitative Culture


Regulatory Compliance

Check out the Resources page for tools to learn more about Data Leadership.

Connect with Us

Use the form below or email info@algmin.com.
We'll respond within 24 hours!

digits only
One subscriber each month wins a copy of our Data Leadership book in the format of their choice!