In Part 1 of this 3-part series, I covered the building blocks of a Defensive Data Strategy and when they should be employed - todays article will focus on the art of attack.

Defense may win games, as many a sports coach will tell you, but attack is what sells the tickets. Most data professionals and organisational executives alike want to talk about, and hear about, how we can maximise the value out of the data that is sitting a their feet. “Data is the new oil”, as Clive Humby proclaimed almost 20 years ago, people have been repeating it ad infinitum in the 2 decades since, and most organisations strive to be the next oil baron.

Let’s scale back the ambition and inject a balanced dose of realism. Your Attacking Data Strategy is probably not going to net you the revenue lines of Alphabet or Meta, but it does stand to significantly change the way you do business, interact with customers and can unlock quantities of value to your organisation that are not to be laughed at. Indeed, an effective Attacking Data Strategy can evolve your organisation in ways that were not previously imagined.

The Basic Elements of your Attacking Data Strategy

Our Defensive Data Strategy elements were designed to keep us safe (to keep us out of the headlines for all the wrong reasons). Complementing that, our Attacking Data Strategy elements are aimed at exploring and exploding the value in our data. This comes from a process of firstly ensuring that our data is nimble, repeatable and explainable enough to be useful (Data Commodification), followed by the capabilities to maximise our data internally (Data Democratisation) and externally (Data Commercialisation).

Once these building blocks are in place the true fun begins as we explore and educate the organisation on The Art of the Possible with our organisational data which will result in uncovering those first Killer Use Cases - Your Moments of Magic. And finally, with some demonstrable value unlocked, and the base building blocks in place, you will be laying down the path to Becoming a Data Inspired organisation.

Time is money people, so let’s delve into these elements…

Data Commodification

The very basics of being able to weaponise data (for good!) is first to be able to ensure that your data has been commodified in a way that it can be easily packaged and consumed, initially within the organisation but quite possibly in the future outside of the organisational walls as well.

A few years ago the indefatigable Zhamak Dehghani published her now immortalised paper courtesy of Martin Fowlers site on Data Mesh Principles and Logical Architecture. Don’t worry, this isn’t another Data Mesh article, but Zhamak also brought light to the already known (but little practised) concept of Data as a Product. Her explanation captures succinctly the benefits and components of commodifying data into discrete and atomic quanta - in plain English, it calls out the need for us to build reusable, reliable, repeatable, self-contained data products with a focus on easy consumption.

You must have a strategy (data, architecture, engineering) that enables you to build these components. The patterns must be there, the boilerplate code, the DevOps guardrails - it doesn’t have to be perfect, but the basics (and a plan) must be in place that enable you to build data products that are geared towards reuse and consumption in a predictable way. These will be the building blocks for your attacking data plays - you cannot skip this step (and trust me, people will want you to, so be ready).

Data Democratisation

The words have been so overused in the last decade that they’ve almost lost all meaning and have most certainly lost all impact, but the foundational message is still of extreme importance. Data is not “an IT thing” - say it with me, data is not technology. There is nothing more “business” than data - it is the descriptor of your business, your customers, your processes, every event and intent that takes place throughout the full business lifecycle and customer journey. There could be nothing more “business” than that.

For too long now data has been seen as the play thing of IT and Technology departments, and this has often been due to necessity as the technological barrier to entry for data has just kept growing (we had to learn spreadsheets, then tables, then databases, then warehouses, then lakes, machine learning, data science, statistical algorithms, mapreduce patterns, cloud hyperscale, distributed computing…it was all too much, so easy to baffle).

Data Democratisation is the process of ensuring data is consumable by those who benefit from the data. This may be “the business”, it may be the customer, it may be the market, it almost certainly isn’t IT. Your Data Democratisation strategy needs to focus on the processes, technology and practises (including Data Literacy uplift) that needs to happen to enable functional and practical data democratisation. You need to get the data in the hands of the people who will turn it into insights and therefore value.

Data Commercialisation

Once you have commodified your data and enabled the effective internal consumption in order to unlock value (and not before), it is time to consider the latent value in your data to others outside of your direct focus so far. This is typically the realm of a Data Commercialisation strategy.

There has been significant progress in the world of data sharing standards (sharing API standards, vendor-based marketplaces such as Azure Data ShareAWS Data Exchange or Snowflake Marketplace) which make the process of data commercialisation significantly simpler than it was half a decade ago.

Your strategy should therefore not focus on the interfacing technology, but more so on the internal policies, consent frameworks and legal terms you need to establish to enable the commercialisation of your data. If you’ve developed a solid Data Commodification strategy then the data products and their value propositions should already be clearly defined, what you need to enable are the mechanisms to safely monetise those products - this is likely to be something that has not been done before by your organisation (in relation to data assets and products) so will require some significant pre-work.

Focus on the enabling frameworks and policies that you need to jump through the hoops of selling data products - the technology and the marketplace are largely taken care of, and the former from experience takes a lot longer than anticipated as it’s akin to launching a completely new product line and often a new business model.

The Art of the Possible

You’ve laid the plans for the key foundational pieces of an Attacking Data Strategy, you have commodified data products and methods to leverage them internally and externally. The next piece of the puzzle to work on is laying the educational groundwork for expanding what people think of as “the art of the possible”.

One of the key questions to ask:

“What is the one thing you wish you knew the answer to that you can’t answer today?”

The answers to this question will be extremely varied - “Why do our customers prefer our more expensive products?”, “Why are we selling more in Market X than Market Y?”, “What is the key decision maker in why our customers walk away from us?”, “How can we reach Demographic Z?”, “How do our competitors run at half the operational cost of us?”.

That’s your starting point for the educational strategy to extend the art of the possible. When most non-data people think about “data” and “data strategy”, they think about technology, about data platforms, and lately about LLMs/GenAI (how many conversations have to start with ChatGPT…really??!!) - they think about data as a thing as opposed to their data. Data is an enabler, it’s not a thing. It enables us to answer questions, to mine insights, to discover and understand in ways we never imagined - but to get the most out of data we need to unlock the part of our brains that asks the questions we previously thought were impossible to answer.

Your Attacking Strategy needs to specifically work through an approach that will foster the permission to reimagine every individuals “art of the possible” thinking. If you use data to ask the same questions you’ve always asked, you won’t be able to find new ways of utilising data to unlock value. In the classic analogy given by Henry Ford - don’t ask for a faster horse.

The Killer Use Cases - Your Moments of Magic

The most pivotal part of an Attacking Data Strategy is your plan to unlock those tangible moments of magic. This is when we translate from foundations to actual tangible quantifiable value.

Before you enter into these conversations, define a robust prioritisation framework that emphasises the need for value return - it can be something as simple as Weighted Shortest Job First (WSJF) or many of it’s offshoots but it must allow you to facilitate the prioritisation conversation across your organisation in a way that produces a prioritisation stack rank that holds up to scrutiny and is devoid of prioritisation by emotion (ie the squeaky wheel).

Armed with your prioritisation framework, and often focusing on a single data domain, define your strategic plan for workshopping and canvasing your way to a small set of “killer use cases” that will measurable show how they have unlocked and returned value. Ideally focus on use cases that show a simple value metric - an increase to a bottom line, or a decrease in operational costs.

These use cases are going to give weight to your organisational data strategy, to help facilitate continued investment in data and insights, and will help in future “art of the possible” expansion conversations. Perhaps more importantly they will allow you to iron out the kinks in your strategy (and potentially invalidate parts of it - failing fast is good) whilst still delivering value.

Becoming Data Inspired

The final piece of your Attacking Data Strategy should lay out the plan for your organisation to take the steps to become a truly data inspired organisation.

You may have already heard of being data driven, or data informed - but what does it mean to be data inspired? The simplest analogy is as follows:

Setting the scene, think of the process of driving your car to go to dinner in a restaurant several suburbs away.

Data Driven - You’ve printed out a map and turn by turn instructions (anyone remember the MapQuest days? Or even a Street Directory?). They detail each turn that you need to take to get to your destination and you must follow it to the letter. If there is a deviation, you’re on your own. Follow. The. Map. You will get to your destination, but you’d better hope that the path is an easy one and everything runs to plan.

Data Informed - You have a very solid plan to get from A to B and have detailed all the twists and turns along the way. The information has been tailored to your situation when you begin and it is clear and easy to follow. You understand the route before you take it, you know how to get to your restaurant, but you will also respond to queues as they present themselves. There are roadworks along the way and you are diverted down a detour, you use this new information to follow the detour and guide yourself back to the road so you can continue on your way. You’ll arrive at your destination…but there may be surprises along the way that weren’t known ahead of time so you may be late.

Data Inspired - Your phone recognises the restaurant reservation in your email understanding the date/time and how long it will take you to get there based on your current context - it gives you the appropriate notice. You are given multiple modes of transport options based on the time of day, the conditions of your starting point and your prior preferences. Once you’re on your way you will routinely be prompted with alternate routes, avoiding traffic, accidents, tolls and any other less than desirable situation. You’ll arrive at your destination with the minimum amount of fuss with millions of data points used to enrich your journey and make it feel seamless and simple.

Moving from being data driven to data inspired is an evolution and a maturity curve - there is no shame in being data driven, it is an informed position to be in. But not many of us would go back to a life where we had to direct ourselves in that way - the convenience of a data inspired solution is simply too compelling and enriching. Being Data Inspired requires not being prescriptive to the question being asked, and being open to letting data guide the way.

Articulating what the “North Star” of Data Inspired culture looks like for your company is a truly individual journey for each company, and charting the course to get there is equally bespoke, but it’s a needed investment of time and effort so organisations can align on what awesome looks like, and how they continually progress towards that state. As important as it is to paint this north star, the course to get there, along with the required investment, is a reality that should be articulated loudly and clearly - becoming data inspired is not fast and it’s also not an insignificant investment, but for most organisations the value return makes it a worthy place to put time, effort and money.

Stitching it All Together

Covering all of the elements above you will have the structure in place to present a compelling and effective Attacking Data Strategy that paints the picture of what a transformed, data inspired organisation would look like, which use cases it will showcase and how it will grow value internally and externally through data.

What’s Next

You’ve now had a chance to read through and hopefully assemble the base elements of your Defensive Data Strategy and are now armed with the tools to do the same for your Attacking Data Strategy. Your next key decision is how to balance the two in a way that appropriately addresses your organisational risk appetite and want for innovation, evolution and growth. Employing the wrong mix in the wrong organisation at the wrong time can make even an excellently crafted Data Strategy ineffective and, in some cases, damaging.

In the final installment of this series we’ll review how to research and read the climate of your organisation and how to dial in your data strategy to the right levels of defence and attack walking through some real-life case studies of when organisations have got it right, and also a few where they’ve got it wrong.