If the past few decades have proven anything, it is the power of data, broadly defined to include activities that include small data, big data, statistics, analytics, and artificial intelligence by employees of every level of sophistication, to help companies improve their performance, in many ways. At one end of the spectrum, artificial intelligence (AI), especially generative AI, promises to transform business and therefore garners the most excitement. On the other end, basic analysis using small amounts of data It is surprisingly effective in helping companies make better decisions, control and improve business processes, better understand customers, and improve products and services. Read the popular press and you might conclude that data, analytics, and AI are taking over the world.
However, digital natives aside, a deeper look reveals a much darker reality. Despite huge investments, great tools, and many qualified data analysts and scientists, progress AI is slow, expensive, and uncertain. Most data science models are unrealistic put into production, does not provide actual economic value. the tenure most Chief Data and Analytics Officers are so short-sighted that it’s unrealistic to expect them to do much. “Regular people,” those who don’t have data in their titles, don’t know what’s expected of them and, despite claims to the contrary, fear that analytics and AI will change, even eliminate, their jobs. Most companies shy away from small data, robbing themselves of easy-to-get business benefits and the chance to build the organizational muscle needed to attack tougher problems.
It’s all too easy to call for “data-driven cultures,” but actually defining and building such cultures is beyond the reach of most data professionals. Survey data suggests that very few companies have data-driven cultures. Finally, the current level of data quality does not really support data analysis or science at scale.
It’s Time for New Thinking
Digging deeper, it is easy to understand the reasons for these disappointing realities: Companies actually “bold data” in their organizational charts, naming Chief Data Officers and Analytics, established Centers of Excellence, and hired highly educated data scientists, then released them with few guidelines and little supervision. That some succeed is a testament to the strength of some, who have overcome organizational obstacles against high odds. Data leaders are not blind to the issues. They hire data wranglers and engineers to help overcome bad data and product managers to help better connect with the business. However, the whole thing feels like an elaborate (and largely pointless) game of “whack-a-mole.” Continuing on the current path is not good.
Instead, we suggest that a new “management paradigm” for data is needed. As used here, a “management paradigm” consists of a common language, a general vision of the ways that data should contribute, a clearly defined structure of organization that shows how data integrates across the organization, with clear roles and responsibilities for everyone involved. Finally, it should include corporate culture, relationships with universities and vendors, policy, and anything else that enhances, or hinders the effective use of data. The new paradigm uses a more widespread and integrated approach to the use of data, analytics, and AI in business.
But what steps should companies take today to work toward a new paradigm? How should they think about fully integrating data into their business strategies? What targets should they aim for, and what should they do in the beginning? This article builds on our research of hundreds of companies and government agencies grappling with these questions, working with a dozen of them, and our experience with other relevant paradigms to propose three interrelated steps that companies should take today.
Use “Digital Native” as a Guiding Light
First of all, we suggest Google as a candidate “North Star.” After all, data is so mainstream at Google that it doesn’t even have a Chief Data or Analytics Officer, although it does have one. Chief Decision Officer, whose primary mission is to spread the good word about data-driven decisions. Google pioneered People analytics and one of the few whose legal operations to make extensive use of analytics. Analytics and AI are embedded in many of the company’s products and services, and the company’s CEO has stated that it will be “first in AI” in 2016. It pioneered many methods used in generative AI systems today.
Other digital native firms, such as Meta/Facebook and Amazon, have similarly extensive use of analytics and AI, and our research and consulting with these firms suggests that, unlike more traditional firms, the data leader does not need to spend a lot of time. evangelize about the importance of these tools; it is taken for granted and viewed as an essential element of culture.
A few legacy businesses are adopting similar practices, but not enough. At DBS Bank, for example, the largest bank in Southeast Asia, CEO Piyush Gupta argued that the bank’s future competitors will not be traditional banks, but digital native companies such as Alibaba, Tencent, and Ant Financial. To compete with companies, he believes, DBS must collect, manage, and analyze data as well as or better than them. The bank has hired or trained over a thousand data scientists and engineers, developed its own AI applications, adopted an extensive AI literacy program, and encouraged the development of AI use cases across all business functions and units. .
Learn from Other Mainstream Business Activities
Second, for those who find the idea of following the examples of Google or DBS too high a hill to climb today, a closer term model found in almost every organization is the financial organization. Even non-profits and government agencies have finance teams. In our collective experience, we feel that finance, perhaps more than any other function, has achieved the goal of being first.
Going mainstream is the antithesis of being siloed; it means being fully integrated into all aspects of the organization. Consider the characteristics of most financial organizations:
- The financial organization is clearly seen as strategic, with most Boards of Directors having their own finance committee, and expecting financial reports at every meeting.
- The CFO is usually in the “inner circle” of the CEO, and is involved in almost all major decisions.
- The tenure of CFO’s tends to be longer than that of Chief Data Officers, which as mentioned above, tend to be shorter. Additionally, the rapid turnover of CFOs is sending a warning signal to Wall Street.
- Finance employees are integrated into almost every facility, operating plant, division, etc., and have at least one dotted line to corporate finance.
- Almost every manager knows how to perform basic financial tasks, including preparing a budget, calculating ROI, etc. Most managers spend a lot of time on financial work.
- Finance explains what is expected, provides tools and processes that people are expected to follow, and monitors compliance. Compliance with financial procedures is not optional.
- Finance goes to great lengths to ensure that the data it uses, and provides, is of the highest quality.
The list above begins to look at finance as strategic. This will be a significant step forward in getting data mainstream. If learning from and monetizing data are viewed as strategic opportunities, then the organization must consider how to implement each of these bullet points for data.
Finally, Include All
Finally, we have great sympathy for the senior managers. While data analysis has been around for a long time, the technologies themselves are scary and the hype shows that implementing them is easier than it really is. To make matters worse, the urgency to do so seems to be growing rapidly. It’s easy for non-technical employees to get lost in the shuffle. However like reliable data and tight technology implementation, it is important! Consider that even elite data scientists could never define a business problem, understand data, nor deploy a business process model without their help. You also cannot perform necessary data enhancements.
It’s true, we’ve been strong in data for a long time. We see the potential to reduce labor in billions of jobs, increase others, and create new, high-paying jobs (we are equally aware of the risks). At the same time, we think people are wise to be afraid – after all, drudge jobs put food on the table for billions of people around the world. Obviously, it would be best if senior leaders could allay such fears, but we think that’s not possible for most companies. However, we advise senior leaders to acknowledge the fear, just as senior leaders who are not serious about data soon should be doubly afraid.
Instead, we advise senior leaders to recruit as many regular people involved in their data effort as possible. Almost everyone can carry small data and basic analytics to improve their team’s performance. Such efforts build expertise and confidence and together make a real difference. And for many companies, it appears necessary for obtaining larger data and more advanced techniques. Finally, many regular people find a little data empowering and are therefore more than happy to join the fun.
Urgency is in Order
Companies can, of course, manage themselves however they like; although of course they should try to do it in ways that facilitate the performance of its most important task. After all, even the smallest company is a complicated place. One implication is that some items get better attention than others. Until 10 years ago, the relative lack of attention around data made sense – it didn’t in the main, nor should it. It doesn’t make much sense now. Simply hiring chief data and analytics officers is no longer enough. Companies must make some significant changes in management paradigms for data and their top leaders must lead the charge.