Results of aKPMG report ‘The Data Imperative’ reveal that 57% of large enterprises do not have a defined data strategy. An accessible, actionable data strategy is a core component of successful digital transformation, aligning your data assets and capabilities with your business ambitions and identifying the initiatives that will truly drive change.

The purpose of a data strategy.

The main driver for developing a data strategy is to create clarity around the role of data in a business’ future success. Many organisations have so much data they don’t know what to do with it. At the same time, thanks to the ongoing hype around AI and analytics, organisations know that there is potential or value in their data assets, but they simply don’t have access to the expertise that could help them identify the best opportunities for their business. This is where a data strategy can help – done well, it creates a clear narrative around an organisation’s opportunity and a roadmap to inform investment as part of a broader transformation.

Creating an effective data strategy that drives action can be a complex task.  Here’s some of the common questions we get about data strategies:

What should you consider before writing a data strategy?

If there’s one aspect you need to get right with a data strategy, it’s that it must align to (and help deliver on) your wider business goals and objectives. Without this, your data strategy will lack relevance – you won’t be able to engage your senior stakeholders – so inevitably it will become expensive shelf-ware.

We must also remember that delivering a data strategy is all about change – transforming and reframing your business by putting data at the core of your operations heart. If the organisation is not ready for change, then you may not be able to realise your data-driven potential, however well-described in a strategy.

Why might I need a data strategy?

Firstly it’s important to note that every organisation uses data – you can’t not in this increasingly digital landscape. If you hire a single employee, send a single invoice, or simply send an email, you are using or creating data.

However, the role of data has changed dramatically in the last 20 years – from a by-product of operations to a strategic asset that can be used to create a leaner, smarter, more relevant business.  Because of this, an organisation needs a vision and a plan around its use of data and the role that data will play in their future success.

It comes down to Darwinism: how will you fare if your competitors are able to fundamentally make better decisions than you?  Or if they’re able to cut 25% of their processing costs to deliver a product or service?

What should go into a data strategy?

Fundamentally, your data strategy should create a clear narrative around:

1. Your ambitions: An understanding of why you’re creating this data strategy, and what you hope it will achieve in the context of your business objectives - is it to become more smart or efficient for example? Since many people in the business won’t know what a data strategy is supposed to accomplished, we should describe why we’re creating it and what it will enable the business to do that it cannot do today.

2. Your opportunity: A vision for the way in which data will deliver on your business ambition, including a set of specific activities and the impact these are expected to deliver. For example, consider the following:

• What does a future data-driven version of our business look like?

• What value could data deliver to our business?

• What is the cost of transformation? What is the cost of not transforming (or not transforming now?)

3. Your journey: A clear definition of your current maturity levels, and what will need to happen to support the value you are looking to deliver:

• What is our current starting point in terms of maturity?

• What does a practical and balanced roadmap look like?

• What are the immediate next things we should do?

To describe your journey, the data strategy should present a summary of the current levels of data maturity across an agreed set of pillars which include People & Culture, Technology, Data Management & Governance, Skills & Change. This ensures your data strategy is well balanced (and not just a business case for tech investment, or a suggested data management process).

How should we communicate our Data Strategy?

When your data strategy is created, you will likely find that you will need to create alternative versions to communicate key aspects to a variety of stakeholder groups. For example, you may want to create a version for the board to support an investment case, aversion for broader internal stakeholder engagement and even a public version for external stakeholders.

It is important to remember that a data strategy is a live working document, so establishing a clear plan for ownership, and an approach to communication is vital - forming a dedicated project team from all represented areas of the business is essential, rather than being conducted in siloes.

In summary, an effective data strategy can create alignment in the business about the role data will play in meeting your ambitions, helping everyone understand the reason for your journey and the steps you will take towards becoming a more data-driven organisation.

By Rich Pugh, Chief Data Scientist,Ascent.

#datadrivencompany #datastrategy #digitalstrategy

Does your company have the right technology platform for a successful data transformation? At first glance, this seems like an innocent enough question. Surely if you have a modern data platform comprising best of breed technologies, then the answer is a resounding ‘yes’.

In actual fact the answer isn’t quite that simple. Technology is just a tool after all, and for any tool to provide value, you need the skills to use it, and it needs to be right tool for the job (which means that you also have to know what that job is). Having a DEWALT DW729KN does not make someone a master carpenter, and conversely, an experienced carpenter with the right tools and skills still needs to know what to make before they can create something valuable.

It’s the same story with data transformation – before you even start thinking about what technology platform you need, you must make sure that you understand the job that platform has to do. Ideally, you need a data strategy which first sets out the roadmap for how data can help achieve your goals and objectives, and then describes the role that technology needs to play. At a minimum, you need a set of business goals and objectives as your starting point. In any event, you should avoid driving your data transformation journey from a pure technology angle – there are too many examples of organisations who have implemented excellent data platforms without a clear view of the role the data needs to play. The result is a delayed – or worse, a failed – data transformation journey, with no value to show for the effort put in.

Once you have worked out what you want to achieve with data and why you want to achieve it, you are finally ready to start thinking about technology. But how do you pick the right platform to make sure you succeed? In the third of Ascent’s Data Conversation webinar series, I spoke to Alex James, Ascent’s Chief Technology Officer, to look at the key do’s and don’ts in this area, and signpost the path to success.

Where to start: those all-important initial decisions.

The initial decisions you need to make revolve around the key points in the data lifecycle. What technologies are you going to use to acquire data, transform and store it, analyse it and finally serve up information and insight derived from that analysis?

In the last few years, there has been a move towards cloud-based data technologies (think Azure, AWS and others) offering increasingly similar capabilities through commoditised ‘as a service’ solutions. So, deciding which platform to use is no longer about finding the one that can do what you need - usually all of them can.

Instead, you need to focus on your existing technology ecosystem, and look at the potential points of integration with the data platform, like extracting data from key operational systems.

Once you have this big picture view of the data platform and its role, you can then start deciding on the approach to take at each integration point, asking questions like:

• How can we make best use of the technology we already have and so minimise spend on anything new? How can we make sure everything will fit together – legacy and new? Should we integrate with, replace or modernise components in the existing ecosystem?

• If we have multiple sources for the same data, why is this? Can we combine or remove any sources? Can we get rid of any operational silos?

• How can we shorten the path to value? What initial objectives should we focus on, and what slice of the data platform do we need to implement for data to be able to help achieve them?

• Are we set up to develop our technology platform in an agile, iterative way, ensuring regular delivery of value and ROI and making sure we don’t spend a lot of time and money building the wrong thing?

Focusing on integration instead of capability in this way helps make sure that the key technology decisions that you make are more likely to be the right ones (in other words, that you are picking the right tools for the job).

How to start: moving from decision to action.

Once you’ve decided on the technologies you’re going to use, how do you turn those decisions into reality?

Historically you would probably have spun up a team of software or platform engineers to implement the technology platform, then let your data engineers loose on the platform to start moving data through it from source to target. Each team would have their own distinct jobs to do, with project management oversight making sure that everything came together successfully.

This siloed approach to developing technology platforms and data capabilities sounds reasonable on paper, but in practice it has several drawbacks. Each team only feels responsible for its own part of the whole. The relationship between the teams is that of customer and supplier, which naturally leads to a non-collaborative way of working. Even worse, it’s all too easy for one team to blame the other when things inevitably don’t work out right the first time. Siloing teams in this way turns out to be one of the biggest reasons why the technology aspects of data transformation fail to deliver the value they should.

The trend now is to have a single team to develop technology platforms and data capabilities comprising the complete set of roles needed to successfully implement the technologies and then push data through them to deliver information, insight and value. These teams typically include software engineers, data engineers, data analysts and scientists, and business analysts.

Having all these different disciplines working together towards a common goal, with each person understanding how their role fits into the bigger picture, solves the drawbacks of the siloed approach. Everyone understands why they are doing what they are doing, everyone feels a responsibility for the success of the whole solution, and a collaborative way of working becomes the natural approach.

How to avoid a governance nightmare along the way: encouraging data citizenship.

If you get the technology platform right, then you open up a world of data capability that enables everyone in your organisation to more easily use data and insights to achieve their goals and objectives. Low code, no code, self-serve and data democratisation all play their part in removing the historical barriers to data use.

But there is a cost. More data capability leads to more data, and more data requires more data governance. As organisations improve their technology platforms, they move from having data they can’t govern (for example, information in a locally produced spreadsheet shared through email) to data they can govern (for example, information in reports and dashboards shared online). Ultimately this is beneficial, of course, as the end state is a set of governed information and insight resulting in more confident decisions based on trusted data.

Reaching this end state for data governance is challenging though – there is no magic bullet to take you there, and technology is only part of the answer. Technology can help you identify what you need to govern (for example, by generating a list of dashboards and who has access to them), but it can’t (yet!) tell you how to govern it (for example, which dashboards duplicate information, or whether the people who can access them are the right people).

To fill in this gap, you need to look at process and, more importantly, people.

The first step is establishing a data governance framework, comprising policies and procedures together with a team who can help ensure that these are followed. This takes care of the process part of the equation, but to get the people part right starts with recognising that human nature is to take the path of least resistance when it comes to data governance, and that enforcing policies and procedures is probably the least effective way of establishing the governance that’s needed.

That’s where the idea of data citizenship comes in. Show people where they fit into the organisation’s data landscape, how they rely on others upstream to provide them with the right data of the right quality, and how what they do with data in turn affects others downstream. In other words, show people what it means to be a good data citizen, how it benefits them and why it’s important to others. This completes the data governance equation, and helps ensure success by working with human nature instead of against it (encouraging rather than enforcing).

Bringing everyone along for the ride.

Finally, having the right technology platform able to serve up the right data to the right people in the right way is still only part of the story. To achieve value, people need to be actively engaged with, and using, the technology and the insight it provides.

Remember that not everyone will be excited by the in-depth details of how everything works. As we said at the outset, technology is just a tool, and too much focus on the technology aspects of your data platform risks building apathy towards it, which is far from ideal.

Instead, focus on how the technology benefits people - by making sure they know how to use it, and what to use it for. Once people see how data and insight helps them achieve objectives and goals, they’ll buy in to the platform and ultimately become champions and advocates for it. Word of mouth anecdotal evidence will start to augment official communication and messaging, leading to a virtuous circle that guarantees that the technology platform delivers the intended value from the data it provides.

So, do you have the right technology platform to succeed?

To close, let’s revisit the question we started with: does your company have the right technology platform for a successful digital and data transformation?

If you’ve followed the above recommendations then you’ll have the right technology implemented in the right way. The data in the platform will be governed effectively, and the technology will be actively and consistently used by people in your organisation to provide them with tangible value.

And the answer will be a resounding ‘yes’!

Most organisations today recognise that data will play a role in their future success. Many have also learned the hard way that achieving this success requires more than just getting the data and tools right. They have also realised that people and process are equally important, and that establishing a data-driven culture lies at the heart of unlocking the potential value in their data. Taking a ‘build and they will come’ approach just doesn’t work – after all, it’s human nature to resist change and, left to their own devices, people will often ignore any new data capabilities that require them to change the way they work.   

Cultural change is required to overcome this challenge; people need to change the way they work with data because they want to, not because they are forced to. Achieving this cultural change is often the hardest part of any transformation involving data. Getting it right is about accepting that some people may not be willing to change, especially where data is concerned, and proactively working with them to bring them on board.

In the second of Ascent’s Data Conversation webinar series, Microsoft’s Chief Data Officer Robin Sutara, joined Rich Pugh, Ascent’s Chief Data Scientist to discuss how, by getting the culture right, you can overcome this barrier to data-driven success. This article presents six key takeaways, inspired by that conversation, that will help organisations to establish a strong data-centric culture wherever they are on their transformation journey.

1. Overcome people’s natural resistance to data.

People are often reluctant, or even fearful, to embrace data and technology. The rapid increase in digital and data transformation over the past few years has left people with concerns over how quickly they are expected to become proficient with new data capabilities. Our rapidly expanding digital footprints also leave people with uncertainty over how to comply with data privacy legislation.

The key is to show people how data and technology can make their jobs easier not harder, and to show them how data capabilities can help them achieve their goals and objectives rather than hindering them. Establishing an organisation-wide data literacy education program achieves this in a practical, structured, and repeatable way. Reassure them that becoming proficient with data is a skill that needs to be learned, and that they are not expected to be instant data experts. This, in turn, helps them overcome their reluctance and fear.

2. Work top-down as well as bottom-up.

Every organisation will have pockets of data excellence: individuals or teams who are already achieving great things with data and need no convincing of its merits. One approach to establishing a data-driven culture is to simply take a bottom-up approach and try to increase the number of people in this category, through education and data capability provision.

Whilst this approach will be successful to a degree, the most effective way of establishing a data culture is to balance this with a complementary top-down approach, ensuring that there is executive support for building data maturity. Leaders must show continual commitment to the organisation’s transformation agenda and communicate the vision and strategy for data effectively at all levels.

3. Build momentum through advocacy.

The best way to build momentum and a sense of community when establishing a data culture is through organic growth. Strategic data initiatives, communication and education also play their part, but the value of having enthusiastic data advocates who can incrementally grow and strengthen the culture from within on a day-to-day basis should not be underestimated.

Data culture advocates, through their business-as-usual activities, can proactively spread the ‘data word’ to everyone in the organisation. Anecdotal evidence of how better use of data has improved their working lives provides a compelling illustration of how it could also make others’ jobs easier, creating onward momentum and another way to help overcome peoples’ resistance to data.

Who makes a good data culture advocate? The short answer is anyone in the organisation who has seen the benefits first-hand, and who is keen to pass the message on so that their colleagues can follow suit. They can be found anywhere in the organisation, and, ideally, should comprise a diverse set of people at many organisational levels and from many business areas, so that everyone can find an advocate they can relate to.

4. Build the culture iteratively.

Establishing a better data culture takes time - little and often is the way to go. Not everyone needs to be convinced of the merits of better data and technology before an organisation can start to build a better culture around data.

Look for where the enthusiasm lies first and build out from there. Support data advocates with new experiences and opportunities to contribute so that they remain eager to promote the culture from within. Implement the more strategic culture building initiatives (such as data literacy education) incrementally, at a pace that ensures the organisation sustains motivation and has the time to adapt and change.

5. Measure progress.

With any project or initiative, measuring progress is vital to keep on track and take corrective actions needed to ensure success. Establishing a data culture is no different – the key is to put in place the right level of progress reporting and then let anecdotal evidence  and storytelling provide the context people need to make it relatable.

Progress reporting should comprise both quantitative measures (like data capability usage and consumption, or process efficiency improvements) and qualitative measures (employee feedback) if possible. The sweet spot is 2-3 measures that provide a directional view of progress - this is enough to show a return on investment and to help inspire people to want to become part of a vibrant, data-fluent community that is keen to explore new opportunities to create value.

6. Communicate success.

The final tip is to communicate success regularly. A data-driven culture has people at its heart, and those people need to know that the culture is working - and that it is worth continuing to evolve and grow it.

So, make sure that successes are communicated - share plans and achievements, report on progress and culture survey results, share anecdotes on how data is making peoples’ jobs easier and help convince any sceptics that this better data-driven culture is something they really should be a part of.

In summary, by establishing a data-driven culture, organisations stand a far greater chance of success in the age of data-driven transformation - and conversely, without strong cultural foundations, organisations will likely be left behind by their competitors and find it increasingly hard to survive.