Authors of this e-book: Pragyan Subedi and Merishna Singh Suwal
Both of the authors are professional data strategists at Kharpann and have worked with over 10+ companies in long-term data strategy-related projects. The aim of this e-book is to provide a concise introduction for business owners on how to create a proper data strategy based on the authors’ insights. If you wish to receive data strategy consultation for your business from the authors, please write to [email protected].
Welcome to ‘Introduction to Data Strategy for Business’—an e-book on how to build successful data strategies for a business taught by seasoned data strategists.
In this e-book, we provide you a complete overview of how you should think about data strategies as well as how you should go about implementing them for your own business or for an employer/client business.
By the time you reach the end of this e-book, you will be equipped with the fundamental knowledge of frameworks and methodologies required to create powerful data strategies that can change the growth trajectory of a business.
Are you ready? Let us get started.
Contents of this e-book:
This e-book is divided into five different chapters that cover different concepts related to building a data strategy:
- Chapter 1: History and Introduction to Data Strategy
- Chapter 2: Importance of building a Data Strategy
- Chapter 3: Difference between Data Strategy and Tactics
- Chapter 4: Factors behind a great Data Strategy
- Chapter 5: Data Strategy Development Cycle
Chapter 1: History and Introduction to Data Strategy
With the onset of the late 90s, data generation started to grow at an unprecedented rate.
Back then, everyone had the word ‘internet’ at the tip of their tongues and venture capitalists were highly interested in funding new internet-based startups.
These internet-based startups were also known as ‘dot-com companies’ and they spent a lot of time, money, and energy getting people onto the internet. Consequently, the internet flourished as a place for all sorts of things and people started leaving behind a trail of data with every website they visited.
So, even when the dot-com bubble finally popped in the early 2000s and dozens of dot-com companies went bankrupt, the result of the efforts they had made on getting people online still lingered on. At the same time, a technological shift occurred and people started switching from dial-up networks to broadband connections, shifted from chunky hard-built phones to wifi-enabled smartphones, and found it easier to chat online rather than to call someone. So, more and more people got online and started generating tremendous amounts of data.
With all of this going on during the 2000 to 2010 decade, technology-based organizations quickly realized the importance of collecting, storing, and using the massively available data. They created email signup forms for sending out newsletters, integrated contact forms with their websites for getting personal information, and built community forums to fuel social interaction. Furthermore, a few companies like Facebook, Google, Amazon, etc were extremely smart at how they stored and used data, and thus, they were able to test out new ideas and innovate their products at a speed as no other organization could.
Looking back at this from where we stand today, all of the things that such organizations did to collect, manage, store and use data at that period of time is now what we know as ‘building and implementing a data strategy’.
So, a data strategy is a plan of action designed to achieve a long-term data-centric organizational goal. We can also call it a roadmap for guiding an organization on what data to collect and how to store, manage and consequently, use the collected data for achieving organizational goals.
With this overview of what data strategy is, we can now move on to talk in more detail about why it is important to develop one and what are the key components needed to make a data strategy successful.
Chapter 2: Importance of Building a Data Strategy
First of all, let us understand why building a strategy of any kind is important for an organization.
Consider there’s a logistics company that delivers items from one country to another. When a customer provides an item to the logistics company, the company has several questions it needs to answer for making a successful delivery.
The questions can be as simple as how to package the item, how to store it and how to transport it. And, the questions can even be as hard as, how to handle customer refund requests when an item is misplaced or broken or how to monitor the delivery of different parcels going out in different routes.
To answer these questions, the company needs to create a strategy to set a roadmap for what will happen before, during, and after delivery for each parcel. This is why making a strategy is highly important for an organization.
Building on this concept, we can now try to understand why a data strategy is needed. There are several questions an organization needs to answer in order to make efficient use of its data. Some questions that can come up are,
- How to collect data?
- How to store and manage data?
- How to make the data available to different organizational departments?
- How to power decision-making using data?, etc.
All of these outlined questions are just looking at the tip of the iceberg and there are a lot of other questions hidden behind the surface. This is also the reason why organizations prefer to hire data strategy consultants rather than building out an organization-wide data strategy by themselves.
Forming a data strategy helps an organization align themselves with their goals and with a roadmap in place, they can execute efficiently without being in problem-solving mode all the time.
Exercise: Think about the organization you’re working in and write 5 different questions that you think would be answered if there was an appropriate data strategy in place.
Chapter 3: Difference between Data Strategy and Tactics
Oftentimes people in leadership roles confuse data strategy with data tactics but it is very important to differentiate between the two.
A data strategy is created by envisioning a future and then, planning out a path to get there. It clearly defines an organization’s long-term goals and how the organization looks to apply its strengths against the most promising opportunities. Also, since a strategy is made looking into the future, it is susceptible to change when new information or insight comes into light.
On the other hand, Data tactics are created by looking at a much shorter time frame and by setting a measurable goal. They are concrete and are oriented towards how a part of a data strategy is achieved with small progress increments.
By now you must have figured out that there is one common denominator that binds both data strategy as well data tactics together and that is the long-term goal set by the organization. So, the quality of a data strategy or a data tactic is defined by how SMART the goal is.
By SMART, we mean the following – is the goal specific, measurable, actionable, relevant, and time-related.
- Specific means can the goal clearly identify the problem or the opportunity
- Measurable means can the goal be measured using a quantitative or qualitative metric
- Actionable means can the achievement of the goal help the organization towards performance improvement or any other important aspect
- Relevant means can the goal address the actual pain points of the organization
- Time-related means can the goal be looked at from a viewpoint of time and the trends that come with it
Let us look at one such SMART goal and break down what kind of data strategy and data tactics can be created based on it.
|SMART Goal||Increase website conversion by 30% compared to the previous year|
|Data Strategy||Run optimized paid advertisements|
The data tactics for the above SMART goal and data strategy can be as follows:
- Run a Search Engine Marketing campaign with well-researched focus keywords that have less competition.
- Create a retargeting ad for previous website visitors.
- Convert currently running ads to be call-to-action (CTA) focused rather than website traffic focused or brand awareness focused.
- Add a sales coupon.
- Add referral-based benefits.
With this example, you now know the flow of how a data strategy or tactic is formed in relation to an organizational goal. Next time, if you are compelled to perform a move on the tactics level, make sure to first outline and understand what the overarching strategy and the goal is.
Remember: Think strategically, act tactically.
Chapter 4: Factors behind a great Data Strategy
You already know that when creating a data strategy, you should keep the organizational goal at the highest priority. However, there are a lot of other factors that come into play when creating a great data strategy than just the guarantee of achievement of the goal.
This may sound counter-intuitive since why would a data strategy not be considered as great if the goal has been met. Well, in practicality, things are not as straightforward as they seem and there are multiple factors that need to be thought of as constraints when creating a data strategy.
You can understand this more clearly with the help of the following example of creating a war strategy.
Consider that there are three war generals strategizing to occupy an enemy nation. Here, their main objective is to gain total control of the population as well as of all of the resources in the country.
After formulating their individual strategies, they go up to their king and ask him to review the three strategies they’ve come up with:
- The first general suggests that the king should send an assassin to the country’s capital and kill the enemy king. This way the army can march right inside the country and take control in the midst of all the chaos.
- The second general suggests that the king should seek an alliance with the neighboring countries and launch a large-scale attack.
- The third general suggests that the king should ask if the enemy king would like to bow down to him and end this matter in a more peaceful manner.
All three strategies sound good but which one would you personally prefer? You see it isn’t an easy task to come up with a strategy.
Even if the most peaceful strategy works in the given example, that is, the enemy king bows down to the king, the king needs to be constantly worried about the enemy king starting a conspiracy behind his back.
In the eyes of the king, he has to think about his people, his resources, his values, and his own vision when approving any of these strategies. The same is true for data strategies as well.
There are many factors that determine the greatness of a data strategy.
- Alignment with the vision and values of the organization
The first factor, as you may have already guessed, is whether or not the strategy aligns with the vision and values of the organization.
For example, at Google, ‘Don’t be evil’ is one of their codes of conduct and this means that if a data strategy aims to exploit sensitive customer information, this will be regarded as evil and the strategy will not be implemented no matter how well it can be used to achieve organizational goals.
- Consideration about the people and the culture of the organization
The second-factor one has to think about when creating a data strategy is the people and the culture of the organization. If a data strategy doesn’t take this factor into consideration, the people responsible for carrying out the strategy or the people benefiting from the strategy, in general, will not be interested in getting involved and thus, the strategy can never come into implementation.
For example, trying to set up a data-driven decision-making system for an organization will not work if the managers have to go through a long learning curve. They will simply ignore the system altogether after a certain amount of time.
- Data governance
Data governance has to be thought of as an integral part of a great data strategy.
A great data strategy should have a collection of processes, roles, policies, standards, and metrics that ensure the effective and efficient use of information in enabling an organization to achieve its goals. Lack of this can only mean that the data strategy introduces more chaos in the existing data practices of the organization and the goal will most probably never be met.
Furthermore, the following are some of the other factors that play a role in determining the greatness of a data strategy:
- Availability of resources
- Overall Data Literacy Rate of the organization
- Choice of technology and architecture
- Operating model of the organization
You now understand how complex of a task it is to create a great data strategy. Also, remember that a founding member of the business is the best person to create a data strategy with since their knowledge about the entire business is the highest.
Chapter 5: Data Strategy Development Cycle
Since you now have an understanding of what data strategy is, let us dive deep into the go-to process of creating a data strategy for any given organization.
Data strategy development happens in a cycle of 5 major stages:
- Step I: Define Organizational Goals
The first step towards building and implementing a data strategy is to start defining organizational goals. We have already talked about how to set SMART goals (in chapter 3), so, let us talk about qualitative and quantitive goals in this chapter.
An organization’s goal should be aligned with its vision and mission and such goals can be either quantitative or qualitative.
A quantitative goal is a goal that can be measured and such goals can be created using an objective approach. For example, a quantitative goal may be to increase customer retention by y%. Such goals are easier to set and the fulfillment of such goals can be easily determined based on whether the target is hit.
However, a qualitative goal is a goal that can be felt and not quantitatively measured. Such goals do not follow the ‘Measurable’ criteria of SMART goals but most organizations have a need for such goals. For example, a qualitative goal may be to increase word of mouth brand awareness. The best way to evaluate the fulfillment of such goals can be by taking a test before and after a data strategy has been implemented (such as a Net Promoter Score).
Once you list out all the goals of the organization, make sure to create a prioritization matrix where the x-axis is the implementation feasibility and the y-axis is the business value.
Using this matrix, you can easily figure out which goals should be high in your priority list and which goals should be either skipped completely or pursued with minimal effort.
- Step II: Current State Assessment
The second step in the data strategy development cycle is to perform a current state assessment of the organization. You can think of the assessment as a collection of questions that you may want answers to before building a data strategy of any kind for the organization’s high-priority goals.
Please note that you should not go on building a data strategy before knowing all the current processes, policies, and infrastructures in an organization. We’ve seen multiple cases where strategists spend months building a data strategy just to realize that none of the employees are skilled enough to implement it. This has happened way too often.
A proper approach to performing the current state assessment is by taking a top-down approach. First, indulge yourself in conversation with the senior level managers and make your way down the organization. Senior-level managers can give you pointers on what they think of the organizational goals you’ve defined on step 1 and can help you introduce the right people down the organizational chain.
At this step, you should also identify the available data sources, current data-related practices, and points of inefficiencies.
- Step III: Data Strategy Roadmap
Once you know the organization’s high priority goals and the current state of where the organization is, the next step is to create a data strategy roadmap.
The data strategy roadmap is typically a 3 months to 5 years plan of how an organization goes from where they are at to where they want to be. The time-bound of the roadmap is dictated often by the goals of the organizations and the time needed to achieve them.
A data strategy roadmap should contain actual data strategies for all of the organization’s high-priority goals. It should define what tactics need to be employed for each data strategy as well in order to go from point A to point B. Furthermore, a data strategy roadmap should not have any determinable uncertainties such as what would be the timeline, budget, or required number of personnel and their work responsibilities for a given data strategy and tactic.
This step is the hardest and the most crucial for any organization. Senior data strategists are highly required at this point so if you’re new to this, we suggest that you either get an advisor or a consultant to help you through the task. A well-defined data strategy roadmap will serve as your codified playbook once completed.
- Step IV: Earn approval and Implement
The fourth and final step of the data strategy development cycle is to earn approval on your data strategy roadmap and start implementation.
At this step, your job is to finalize the strategy by discussing what you’ve made so far with the senior managers. When doing so, please understand that earning approval for a data strategy roadmap is like getting approval for a finished logo design – you’re finalizing a draft and not presenting a final design.
Almost all of the time, most organizational leaders will want some changes in the roadmap even if you’ve spent weeks making it. This is actually a positive response since it shows that the leaders are thinking about the roadmap as hard as you are and want to create the best outcome during implementation.
Once you make the necessary changes and the data strategy roadmap gets approved by the stakeholder, the required budget for the first or a few milestones in the roadmap is naturally approved as well. This is when you start hiring the necessary personnel for the implementation, start building the infrastructure, and lead/hand over the project to success.
- Step V: Introspection and Revision
The fifth and final step for a data strategist in the data strategy development cycle is to see if the organization is staying on its course as per the data strategy roadmap and if yes, is the data strategy roadmap working or not? This means that your job is to patch out any holes or problems that are uncovered during the implementation.
We suggest that you hold weekly or bi-weekly meetings to make sure things are progressing at the right speed and all difficulties that arise with the implementation are being addressed. Also, since a data strategy roadmap is a long-term plan, there may be different factors that come into light due to economical or leadership changes in the roadmap. Your job is to navigate the rough seas and bring the ship to land.
We hope this e-book was helpful for you in getting a good understanding of how to build successful data strategies for a business. There are many things to learn and you will uncover them along the way as you work up your way to become a professional data strategist.
If you have any questions or suggestions to edit this e-book, please write to us at [email protected].
If you wish to receive data strategy consultation for your business from the authors of this e-book, please write to [email protected].