data strategy; data science; big data; data analytics; IoT

What is a data strategy and why do I need it?

Data, it is said, is the “oil” of the 21st century. By getting lots of it, you can better understand how your customers will behave as well as predict their needs and requirements, building a virtuous cycle of growth, allowing you to obtain yet more data, and so on. Unfortunately, most businesses do not make adequate use of the data they have, because they do not have a data strategy in place. Indeed, as much as 70% of enterprise data remains unused, and 35% of companies do not even see the value of a data strategy.

In a world where social media, IoT and connected devices means unprecedented connectivity, making optimal use of the data your company has is an urgent imperative. All businesses must have world-class data strategies to remain relevant, particularly if they want to make best use of data-services such as predictive analytics.

To be sure, to make the best use of your data is not about the data; it’s about the strategy you have for making sense of how to use that data.

The challenges of creating a data strategy

A data strategy is an integral part of digital transformation and most companies face major challenges in this respect, from understanding how to transition from legacy IT systems and maintaining data privacy and security protocols or creating personalised customer experiences to knowing how to implement a data strategy.

For companies to overcome these challenges, we propose the following four steps:

  1. Establish a clear vision
    • All departments will be (and should be!) impacted by data, so they all need to develop data-driven cultures
    • Create data-centricity to all facets of your organisation
  2. Establish data accessibility
    • Break down silos and enable all people within their departments to have access to all of the relevant company or department data so they can optimally harness it
    • Ensure your people are trained in data analytics
  3. Interdepartmental collaboration
    • Ensure that there is true cross-collaboration between people from the business e.g. marketing, sales, operations, etc, with data engineers, data scientists, project managers, data analysts, etc, so that the data-side understand the needs and requirements of the users of the data and so the business-side can develop the right context and narratives from the data
      • This will improve decision making and decrease inefficiencies
  4. Keep data well maintained
    • Of critical importance is that data is clean, structured and easy to digest (whilst your workforce should have data analytics training, investing in a simple visualisation interface will allow you to strike the right balance so that less technologically-versed employees can still benefit from superior insights)
    • Data scientists need to consistently develop effective collection and maintenance methods

How to develop a successful data strategy

  1. Data Management: ensure you have the right structures in place so that you can leverage your data in a way that provides you with a competitive advantage
  2. Governance: create the right governance structures, so that your data is processed & stored correctly, consistently across all teams and that all teams follow the same policies
    • Develop a governance framework that includes processes, keep in mind local laws and understand how you share data without compromising privacy
  3. Infrastructure: Migration from legacy systems
    This will involve building new data locations and systems, including cloud migrations, storage, extracting & loading migrations
  4. Data Attributes: Focus on Quality
    • Data is only valuable if its trustworthy; it therefore needs to be reliable so it’s fit for analysis
    • GDPR: encrypt your data; be compliant
  5. Maturity: Understand where your company is in the data-maturity timeline and create a roadmap to become mature
    • It’s not a race against others: it’s about understanding where you are today
      • Descriptive analytics is the most basic level and creates standard reports – is this where your company is?
      • Predictive analytics (looking forward; forecasting; modelling; statistical analytics) – is this where your company is?
      • Prescriptive analytics: prescribing to get actions – is this where your company is?
      • Cognitive analytics: mimicking the human brain (NLP, reasoning, ML) – are you this advanced, perhaps?

Ultimately, commitment from the top is key: they should visualize the outcomes. Data today is critical and companies that don’t harness the power of data will fall behind. By 2020, insights driven business will steal $1.2 trillion annually. Even today, companies with true data driven cultures have 7% more annual revenues.

At present, this means most companies will lose: According to HBR: 80% of analysts still waste time simply discovering and preparing data whilst 70% of people have access to data they should not have. This is why it is so important to have a robust data strategy, and ensure this data strategy is in-line with your core strategy. For this to happen, it must be led by the CEO & executive team.