By Jonathan Westley, Chief Data Officer, Experian UK&I
The growth of data creation and replication skyrocketed in 2020, reaching 64.2 zettabytes. Considering that a single zettabyte is enough to store 30 billion 4K movies, or 60 billion video games, really puts it into perspective. That’s a lot of data.
The volume, speed and variety of data increases every minute; our world is full of opportunities for data. If businesses didn’t know the potential of data before the pandemic, they do now. As businesses shift to working from home and consumers see examples of how data is being used in our pandemic response, we’ve really started to appreciate the reach and power of data.
But we overlook the mass of data at our doorstep. Less than 2% of the data created in 2020 was recorded and retained until 2021. And while most companies now collect data in one form or another, many struggle to appreciate the potential of this data or to store in formats that make them impossible to scan.
Organizations are not getting the most out of data because they are missing two-thirds of the pieces of the puzzle. To optimize data processes and extract insights that will lead to better decision-making, companies need to build a “three-legged stool”. Those legs are data, technology applications (like cloud computing), and analytics powered by artificial intelligence (AI).
Obstacles remain when it comes to putting this structure into practice. Everything from poor quality data and people’s mistrust of data sharing, to lack of data management skills, are preventing businesses from accessing critical data insights.
These barriers are not insurmountable. But leaders need to make a conscious effort to overcome them if they really want to optimize their use of data. For a few (4) tips that will kickstart your journey to building that “stool”, read on.
Invest in data leaders and talent
For companies looking to build strong databases, appointing a Chief Digital Officer (CDO) is a good place to start. The researchers found that companies with a CDO are twice as likely to have a clear digital strategy than those without. A CDO leads the charge by embedding a culture where data is viewed as an asset that helps the business make informed decisions. They can also spearhead business responses to changing data privacy regulations and policies, as well as cutting-edge strategies that will protect sensitive customer and business data.
Additionally, businesses should spend time understanding their current data handling, processing and management situation. Performing a data audit is one of the most effective ways to identify issues and shed light on how data assets are currently being used. It highlights levels of data literacy in a business, where money is wasted, and how data could be better used strategically to increase profits.
Identify opportunities for using data
It’s not easy to know exactly where the next market disruption will come from, but it’s a challenge businesses must overcome to stay prepared for the future. Understanding the magnitude of this disruption is not only good for business performance; it’s vital to staying ahead of competitors and adapting business models to meet customer demands. Be ready to get confused.
Currently, many companies underestimate the range of applications for machine learning (ML) and AI in their field. Instead, many digital transformation projects focus on improving an existing business model. The most productive – and long-term, profitable – path would be to explore how better engagement with data enables your business to operate, innovate and grow in new and more efficient ways. A customer-centric digital strategy powered by ML and AI applications can help companies uncover customer trends before their competitors.
Get company membership
Without company-wide support, data teams will likely fall at the first hurdle. At times like this, we can fall back on the wise words of Chinese philosopher, poet and politician Confucius: “Tell me and I’ll forget; show me and I may remember; involve me and I will understand. A little work to demonstrate exactly how data applications like AI and ML can benefit different business divisions may be needed to get everyone on board. It’s not just buzzwords, it’s technology that will change outcomes.
A simple way to do this is to look at a particular business function and highlight benchmarks that demonstrate the effectiveness of AI and ML technologies against the status quo – or, indeed, solutions. alternatives. For example, in financial services, it is straightforward to show that AI does a much better job of determining credit risk than humans. Experian’s own analysis demonstrates that applying ML in credit decision making can lead to a 25% reduction in bad debt compared to traditional linear regression models.
While it can be tempting to paint a bright picture of the data-driven future of business, data teams should be wary of it. Setting realistic expectations with teams and leaders is key to gaining long-term buy-in to a data strategy. After all, AI and ML applications probably won’t perform perfectly immediately; optimizing them is a process of trial and error.
Build trust with your customers
In addition to involving internal stakeholders, a data strategy must also consider external stakeholders, such as customers and consumers. Customers will be among the first to wonder how data is used, stored and protected by a company. A key part of building trust with them is enforcing rigorous frameworks on how data is managed and used.
Good data governance starts with being clear about what an organization is trying to accomplish. It should be results-oriented and should be communicated transparently. If what you’re trying to do is for the good of the company or the customers, that’s fine and that should be communicated as well.
A good relationship with customers is essential if the data your business relies on comes from them. The effectiveness of AI and ML solutions depends on the quality, accessibility and management of data. Rigorous and fair governance frameworks for data can help a company ensure that data engineering is rigorous and that biased and unfair models are avoided. They also reassure customers, who are often the sources of this data, that their information is in good hands.
Assemble the stool
Having data is only part of the puzzle. Optimizing its use in a way that produces valuable insights for a business also means investing in technology applications and analytics technologies, such as AI and ML. Removing barriers such as poor data quality, data skills shortages, and customer distrust is key to helping these investments take off. Only then will businesses have a solid stool to stand on that can be used to support growth and reap the competitive advantages that the use of data offers.