The rich get richer; the rest play catchup. That describes the innovation achievement gap between companies that have massive resources in AI, data, automation, and state-of-the-art enterprise systems and those that don’t. But for companies struggling to keep up, the real problem may not be the disparities between them and their competitors. Rather, it’s the hidden technological disparities within their own organizations that increase inequalities in talent’s access to automation tools and AI reskilling initiatives that ultimately hinder competitiveness.
Here are three of the most common internal disparities and what you can do to right them:
Unequal Access to Data
In our research and work, we see many situations where humans and systems have uneven access to data. For instance, two-thirds of companies rely on a suboptimal mix of cloud-based and on-premise enterprise systems. These sprawling, patchwork systems cannot support business goals because their data is spread across siloes. For example, one government regulatory agency had more than 10,000 separate on-premise data centers, and employees in areas such as administration, operations, policy, and scientific analysis had uneven access to the data and limited ability to connect data across these sites to coordinate their work.
Other companies find that uneven access to data prevents them from creating a pervasive culture of data-driven decision making. This can lead to numerous pockets of “data illiteracy” in the organization. For example, when a large European energy firm wanted to create an enterprise-wide, AI-driven data capability, they discovered that they would first have to engage in extensive company-wide education to bring all departments up to a common level of understanding of the opportunity.
Leading companies create an ecosystem of services and systems mapped to business capabilities. Data, instead of being locked away and available only via batch processing, moves in real time and at scale throughout the enterprise to enable data-driven business decisions. Cloud technology, instead of being used here and there in the enterprise, replaces costly legacy infrastructure and provides an elastic, scalable infrastructure built for speed, productivity, and innovation.
Consider a multinational toy company operating in an industry that changes faster than most, with competitors offering not just toys, but also digitized experiences. To keep pace, the company wanted to enable every department from design to production to distribution to act swiftly and in concert while controlling costs. Instead of using disparate data management products, the company democratized access to data through a system that integrated machine learning in the cloud to inform business decisions. By providing “self-service” analytics to business users who no longer needed data specialists to help them, the company cut in half the time previously required to perform key analyses. And automating data-quality processes reduced data-quality support requirements by 80%. Now, employees across the company can confidently use the system to make better financial, design, and development decisions.
Uneven Power to Invest in Technology
In many organizations, some teams feel empowered to invest in technology to execute their business goals, while others don’t, leaving them feeling disenfranchised. For instance, teams working in the IT function are too often focused on “keeping the lights on” rather than finding innovative solutions to business problems. Meanwhile, more than 60% of investment in information technology comes from outside the IT department, from such functions as marketing or operations. These “shadow systems” don’t show up on the IT function’s radar, leading to mismatched priorities between business capabilities and siloed technology.
Compounding the problem, IT teams aren’t trained on these shadow systems, hampering their ability to support or upgrade them. As one company reported, this disconnect left the IT team feeling disenfranchised. Similarly, a data analytics team that reported to a company’s strategy function felt increasingly unable to deliver strategic insights because other parts of the business repeatedly asked them to perform piecemeal analytical tasks. As a result, the team felt hamstrung and suffered high turnover.
In an aligned organization, systems clearly map to business capabilities, and investments in information technology bring together all stakeholders, instead of creating shadow systems and leaving some groups out in the cold. For example, a U.S.-based hospital system found that various administrative teams — without physician input — were creating their own customized analytics tools for the same healthcare KPIs. For instance, in analyzing the cost of knee replacement surgery for patients, each of the customized tools drew from different datasets. As a result, physicians had no comprehensive view of best practices across the hospital system, yet they were expected to help improve the quality and lower the cost of care. This led physicians to become skeptical of the comprehensiveness and objectivity of any analytical model they saw.
To remedy this, the hospital system integrated all the data into a single, cloud-based tool and appointed a physician to oversee the effort. The system offered physicians, business analysts, and administrators a single objective view of individual patients across regions and hospitals. With all stakeholders aligned on KPIs and best practices, the hospital system reduced costs for patients and the hospital. The cost to patients of knee replacement was lowered by almost $300, for example. Overall, in the first year alone, the system cut the cost of care by $20 million.
Uneven Access to Automation & AI
In many companies, we find a growing divide between teams with ready access to automation and AI tools and teams without. The latter find themselves behind the eight ball both in terms of productivity and AI skill development. At a transportation company, two teams were developing similar products. Team A was able to automate only one project task, while Team B was able to automate more than 30. This resulted in dramatically different delivery cycles and outcomes.
Consider the disparities among software programmers. Some might spend 60% of their day performing automatable tasks. Programmers who leverage AI tools to handle those activities code faster. They also become expert at collaborating with AI systems and less prone to errors. This divide becomes critical as customer expectations rise and the pace of change accelerates. The market today doesn’t tolerate slow engineering delivery cycles. It demands modern engineering practices with quick build-measure-learn cycles that the automation have-nots cannot produce.
Simply spreading automation more widely doesn’t fully solve the problem. What’s required is a broad systematic approach. For example, in 2016, a software company reached a crossroads. Its IT function had become a patchwork of organizations. Business owners had little transparency or shared expectations with IT. With systems burdened by decades of legacy code, the company was spending 80% of its IT budget on fixing its past and only 20% on innovating for the future.
Instead of simply automating more activities, the company embarked on a wholesale reinvention aimed at outpacing the speed and innovation of its digital-native competitors. They adopted a vertical, modern engineering model, where business experts and full-stack engineers work in integrated teams with agile development practices. They migrated 100% of their IT estate to the cloud, which enabled them to make automation tools available to virtually anyone and any team that could benefit from them.
Crucially, too, all of this was achieved through self-funding — not additional investment. Like most organizations, the company spent most of its IT budget keeping the lights on. Its multi-year reinvention changed all that, flipping the IT investment curve and steadily freeing up capital to enable everyone — not just a fortunate few — to innovate at scale. Today, the company spends just 40% of its IT budget on fixed IT costs and devotes 60% to innovation. And the time required to bring new features to market has been cut by 83%. What’s more, when Covid-19 hit, the company was well-positioned to navigate the crisis and emerge even stronger.
Uneven access to data, tech investment, and automation tools can have a corrosive effect on morale and business performance. But companies can begin to address these technological disparities by prioritizing growth areas that require a rapid return — for instance, R&D in Life Sciences or CRM in Consumer Goods — and then span out. One initial step is to look at the depth and breadth of the current data systems inside the organization. Companies must also work to ensure that their systems map to business capabilities and that investments bring together stakeholders from across the organization. Beyond these formal organizational changes, firms can create cross-functional teams of engineers and business experts, adopting agile development practices to prevent IT professionals from being isolated from critical business activities, and vice versa. The sooner organizations address these internal disparities, the faster they will overcome external disparities with competitors.