Organizations around the globe are looking to data to drive better outcomes. They want to better serve their customers. They want to develop better products. They want to run more efficiently. They want to expand their businesses and develop competitive advantages. Seemingly everyone knows that data has a role to play in today’s highly digitized environment, but too many organizations think data alone will light their path. In reality, data is just the raw material needed to generate the insights that can change a business. So, how do we harness the data in a way that generates insights that, in turn, create value? What do industry leaders do differently in this space compared to other businesses?
Google Cloud and Harvard Business Review Analytic Services partnered to better understand business leaders whose organizations have been the most effective in harnessing data to create new business value and to see how they compare to their industry peers with similar aspirations. This report analyzes where data leaders are prioritizing technology investments, how they operationalize the development of business value, and the results they’ve seen. From revenue to profitability and from customer retention to employee satisfaction, data-to-value leaders are finding success where others continue to struggle.
At Google Cloud, we’ve helped customers distinguish themselves from their peers and competitors by delivering the right set of data analytics and artificial intelligence (AI) tools. We help customers break down their data silos. We help businesses bring in real-time data and make it available across the organization. We provide users with AI-powered data analysis, and we do it with tools they already know and understand.
Put another way, Google Cloud delivers a data platform on which leaders have designed drastically more responsive businesses. A “responsive” business is connected, real-time, and intelligent. Here’s how we help bring that vision to reality:
• Empower everyone to develop insights without limits. Google Cloud unlocks the potential hidden in your data with a cloud-native, serverless approach that decouples storage from compute and lets you analyze terabytes to petabytes of data in a fraction of the time. Remove traditional constraints of scale, performance, and cost so you can ask any question of data and solve business problems. Operationalize insights across the enterprise with a proven enterprise cloud data platform.
• Optimize business outcomes with real-time intelligence. Automatically process real-time data from billions of streaming events, and serve insights in milliseconds to respond to changing business demands. Generate highly accurate predictive insights with industry-leading AI and machine learning services to optimize decisions and customer experiences. Augment existing skills to scale the impact of AI with automated, built-in intelligence in familiar tools.
• Maximize value from data with a flexible, open, and multi-cloud platform. Analyze data across multi-cloud environments from a single pane and bring analytics to data where it is. Run your analytics workflow in the most impactful way by choosing the best tool for the job and combining it through a seamless, unified experience. Google Cloud supports choice and flexibility, protecting your business from lock-in.
I hope you’ll learn from the perspectives and insights we’ve gathered in this report. At Google Cloud, we know that datadriven innovation doesn’t have a finish line—it’s a continuous journey. Wherever you are on that journey, I’m confident that we can help with the right technologies for your business. If, after reading this report, you’re eager to learn more about Google Cloud, I’d encourage you to visit our Smart Analytics site.
A Strategy for Success “There are really no businesses of any reasonable scale that don’t recognize that data is gold. In some cases, that’s flat out what they trade on; in the majority of cases, it’s how they get better at what they do,” says Barry Brunsman, principal in KPMG’s CIO advisory practice. “They recognize it as a strategic priority. But they’re stuck because that data is sitting in an application landscape of multiple systems and multiple instances that has evolved over time.” For too long, IT was focused on process or greater automation. “They were not focused on data as the outcome of the technology landscape,” says Brunsman.
Companies have been trying to extract value from data using IT systems for some time. What’s particularly challenging right now is the increased level of complexity of the enterprise data environment today, according to Dan Vesset, group vice president of analytics and information management at International Data Corporation (IDC). The volume of data has increased, with more of it hosted in multiple cloud environments, and a greater variety of data types need to be analyzed from internal and external sources. The velocity at which the data is created has also increased. “All of this creates silos,” says Vesset. “And not just data silos. Analysis is being done in silos and decisions are being made in silos. Most companies are dealing with this. We’re in a transition period between mostly on-premises systems and the next generation of technology platforms to unify the environment.”
Without an overarching strategy for managing that data and applying it to business problems—both internal data and that from third parties—extracting any value from it is a labor-intensive and often disappointing exercise in frustration. “If you don’t have business rules in place, you end up having to glue it all together,” says TBR’s Woollacott. “And if you need to apply that kind of ‘human putty’ to your data stream, you’re at a significant data disadvantage.”
The need to manually manipulate data leads to abysmally low data utilization figures in most organizations and results in data scientists spending the vast majority of their time wrangling data. “That’s the most horrible version of what a data scientist wants to do,” says Northwestern University’s Hammond. “Even the best organizations have this problem.”
Having a clear enterprise strategy is an important foundational step for managing and extracting value from all data. Nearly all of the leaders (97%) have established such a strategy for their own internal data, versus 59% of all other respondents. An even greater portion (99%) of leaders have a similar strategy for incorporating external sources of data, versus 53% of other respondents. Clarity around data management and governance, and a clear direction about what an organization wants to accomplish with data-driven insight, are both critical. While data wranglers play an important role, they need to be directed toward specific tasks and goals. “You can collect and clean and distribute all this data, but if you don’t have any idea what you want to do with it, you’re going to fail,” says Hammond. “You have to understand both the architecture and the task at hand.”
That calls for the creation of data teams operating under a well-defined strategy and data governance structure to bring it all together. Companies that want to transform data into value often create cross-functional teams that include data wranglers, enterprise architects, data scientists, AI experts, and data-savvy businesspeople who can identify the problems that data might solve. “You need a group of people who together have crossplatform technology capabilities, the strategic inclination to step into ambiguity to solve these sorts of problems, and the customer relationship bent to mediate between those who don’t know anything about technology and data and those who do, so you can apply data and analytics to solve problems,” says KPMG’s Brunsman.
The State of Advanced Data Competencies
When asked about the importance of three advanced data and analytics capabilities—to connect data points across a variety of assets, devices, and services; to access and analyze data in real time; and to automate data-driven insights—the leaders begin to differentiate themselves from the rest of the survey respondents. Around three-quarters of leaders say their organization’s overall performance and success are strongly linked to each of these capabilities, while other respondents were much less likely to say so.
Likewise, greater numbers of leaders report having mature capabilities in each of these areas. Over half (58%) indicate that their organizations are quite capable of connecting data points across assets, devices, and services, compared with just 10% of other respondents. The same proportion of leaders (58%) also say their organizations have mature, real-time data access and analytics capabilities versus just 6% of the rest of respondents. And nearly two-thirds of leaders (67%) say they have mature automation of data-driven insight using machine learning built into their workflows, while only 5% of the remaining respondents say so.
That the majority of respondents have yet to integrate these abilities is not surprising. Connecting data points across a variety of assets, devices, and services, for example, is no trivial task. “Because of the history of what we’ve done, we’ve got a lot of data silos,” says Hammond. “There are technical issues related to moving data from where it was produced to somewhere else, as well as organizational and political issues.” Data leaders determine how to bring together data not only from their own heterogeneous mix of systems, but from external sources, too.
There are numerous hurdles to realtime analytics as well. There’s the question of computing resources, for one, says Woollacott of TBR. “Should they put the analytics at the edge near the data or consolidate it at the application level in the cloud or the data center?” In many cases, the underlying application architecture may limit real-time opportunities. “If you’re running on a mainframe,” says Brunsman, “there may be no realtime options.”