A chief concern among many data leaders is getting value out of data; a major part of this goal is identifying what type of data best suits your business needs.
It’s important to recognise the use cases that your business operates on, and consider the data value chain in your data strategy. Likewise, consider what data would be most liable to your business – and rank this liability against the data’s potential value.
Data Quality and the ‘Deep Economics of Data’ was the main topic at the August CDO Hub get together (a regular event for our exclusive community of senior data leaders in the UK to come together and debate). This discussion was kicked off with a consideration of how to identify ‘valuable‘ and ‘liable‘ data within your data value chain.
One clear example raised in the debate was Facebook’s access to incredibly valuable – but equally liable – personal data, with the potential for breaches or weaponised data a major drawback of this investment. This raises the question: what can you do to increase the value, or reduce the liability?
This is where risk management becomes a real competitive advantage in data; even if your data may be liable in certain circumstances, you can recognise this risk and make amendments to your strategy. If data that holds value to your business could also be liable to breaches, further investment in your governance and management may be necessary. Each business is different and will have different use cases, so it’s critical that your data strategy is aligned with your business goals.
Weighing up the value of data against its potential liability will help to entrench your approach in the long-term. When considering the data value chain, it’s problematic to keep making a quick buck when you can: your data strategy needs to focus on identifying, acquiring and acting on valuable insights. As such, the best approach isn’t only considering value versus liability; you should also consider sustainability.
The ‘data value chain’ approach helps improve data quality by providing a framework for getting long-term value out of the most sustainable data. This is achieved in three steps:
However, from this point you need innovation. Data on its own has no inherent advantage – you need to learn from the results of the data value chain and drive further insight using these learnt lessons. This process helps to pinpoint what it is this data can achieve, and how best you can achieve the best results.
Likewise, once you have created these new competitive advantages in your data, you might want to keep these processes safe from your competition. You can keep ahead of similar organisations with anti-scraping tactics on your own domains, and make porting difficult to prevent the competitions from copying your approach.
Even with your own unique analytical methods, it’s difficult to protect algorithms by copyright or patent. In an open-source world, it’s equally unrealistic that your trade secrets will remain secret for long. At the CDO Hub August discussion, members raised interesting points about what the future of intellectual property rights and rights to control use holds for this debate, as public policy continues to evolve.
Valuable data in the long-term isn’t the task of one person, but of a whole organisation – there’s little merit to acquiring and deriving the insight if the rest of the organisation won’t act on it. Build trust and buy-in for your data through clear communication of your goals, and win over the support of teams across any function that is involved.
This is all to drive your organisation to be better at acting on insight; creating a culture that is both willing and able to utilise this data.
Clarify what benefits will be created by evidence-based decision making, and nurture continual improvement with clear goals for your data – by this point, the value of use cases is more than apparent!