by Holger Harreis, Harald Kube, Henning Soller, and Asin Tavakoli
Data brings big promises for businesses: advanced analytics for customer-churn prevention in banking, self-learning algorithms for automated insurance claims management, artificial intelligence for autonomous driving, an Internet of Things for more efficient manufacturing, optical pattern recognition for earlier and more reliable disease detection, smart machines to relieve workers of repetitive and dangerous tasks, and augmented reality for next-generation capability building.
In real life, driving value with data isn’t always quite as straightforward as such killer applications suggest. How do you size the economic potential of data transformations? How do you identify the biggest value pools, from efficiency gains to new revenue sources? How you get the organization to rally around these opportunities? What data architecture and governance do you need to reap the benefits, and how do you avoid running into trouble with the authorities as regulators tighten the screws on privacy protection?
To get answers, more than 100 top executives and leading experts from all over Europe and a wide spectrum of sectors gathered in London in June for McKinsey’s European Data Summit.
Here are some of the success factors they mentioned in these conversations:
Be clear about the value
- Put impact before data. “Five years ago, we created a data lake with an off-the-shelf interface, assuming the organization would figure out what to use it for. We failed miserably. Very few people used it at all, and everybody else tried to prove the output wrong. Now, we work with our most experienced people to size the impact potential and build our data regime one use case at a time. To get people to want to work with data, they need to see how it can make their lives easier and their businesses more successful,.” —Insurance executive
- Be prepared to reinvent your business model. “For us, it will depend on data whether our products become a commodity or an integral part of sought-after solutions. If we just keep doing what we do, we will become suppliers to more data-savvy companies that provide bigger value to customers, and our cost structure is such that we can’t win that game. So for us, mastering data is a question of survival. That’s why we’re pursuing multiple disruptive use cases in parallel, from R&D to marketing and after-sales service.” —Automotive executive
Know the power of process and governance
- Invest in end-to-end governance. “Data is a value driver. We need to keep track of where it comes from, who touches it, and how it’s used. We will have to establish clear and consistent processes and protocols around the way we work with data across business units and functions. Because much of our data comes from external sources, we will also bring our partners into the circle of trust. Everybody will benefit from reliable standards, especially in a sector as sensitive as healthcare.” —Healthcare executive
- Beware of the swamp. “Some companies have tried to solve all their data issues simply by filling the data lake with copies of data from the source systems, without any governance. But this just replicates the issues that exist in the sources. We believe that, before importing data into the lake, it is essential to establish at least a basic level of governance, like appointing data stewards, populating business metadata, choosing the golden sources, and implementing data-quality controls. This will result in a clean lake rather than a muddy swamp and will enable successful business applications.” —Banking executive
Develop a data culture
- Assemble a team with a broad set of skills. “Of course you need data engineers and scientists to create the architecture and wrangle the data, but it’s just as important to bring business owners on board to get traction. And it doesn’t stop there. We found that you also need to involve strategy experts, designers, change managers, and privacy lawyers to build robust use cases. It’s all about the right team.” —Advanced-analytics expert
- Lead by example. “To me and my team, pulling data directly from the data lake with my tablet is second nature. But our salesforce is still working with makeshift spreadsheets and rolls of paper churned out by dot-matrix printers. We are years away from having a data culture across the entire organization. My job is to lead by example and make everybody see the value data can bring.”—Basic-materials executive
Many companies are still in the early days of managing their data. By understanding what makes data really work, companies can get a jump on their competitors.
Holger Harreis is a senior partner in McKinsey’s Düsseldorf office, where Asin Tavakoli is a consultant; Harald Kube is a partner and Henning Soller is an associate partner in the Frankfurt office.