Faced with ever-growing pressures to optimize costs and improve performance, most companies have taken steps to increase the efficiency of their support functions. An estimated 80 percent of Fortune 500 companies report using some form of a centralized shared-services operating model—but most companies have only scratched the surface of the potential value available. Worse, many have wasted significant time debating the right approach. Should they focus on centralizing processes and functions to increase efficiency, or automate processes through digital technology?
The COVID-19 pandemic has forced companies to act fast. It has also created a window for companies to reimagine the way support functions operate. In the next normal, we believe that the answer to long-debated question is ‘yes’ to both. Companies that centralize processes and functions first are still likely to find that they need to automate them—the “ship then fix” approach. Conversely, other companies may make faster progress by automating processes first and then centralizing them—“fix then ship.”
The right sequence depends on the organization’s starting point and its unique combination of circumstances and needs. And, regardless of the path a company chooses, there are a set of foundational measures that will lead to better results. Rather than continuing to debate, companies can take action now to capture value that otherwise may slip away.
Lessons from the ship-then-fix tradition
For companies improving their support functions, ship-then-fix has been the default option for several reasons—starting with the fact that historically it offered the fastest path to value. Centralizing functions through a shared-services model typically requires far less upfront investment than trying to digitize processes first, and therefore offers a more straightforward business case. Because many organizations have used this approach to reduce costs and increase efficiencies, the approach is perceived (with some reason) as relatively low-risk: the changes are often self-funding, with the savings then available for reallocation toward digitizing select processes (exhibit).
In contrast, digitizing existing processes first under the fix-then-ship model requires an initial investment before the organization can begin to capture value. The investment is not only of money: rethinking processes to maximize digital’s impact requires significant management focus as well. It also means working with rapidly evolving—and often immature—technologies, making the whole endeavor seem higher-risk. Enticed by the promise of, say, natural-language processing for interactive voice-response (IVR) systems, companies can spend more time trying to make the latest technology work than on thinking through the full range of potential solutions—some of which might be substantially cheaper and more reliable.
However, as process-automation technology stabilizes, the argument for the ship-then-fix approach becomes less compelling, especially since and it can lead to unplanned consequences of its own. Too often, once the first steps toward centralization allow managers to capture near-term savings through consolidation and cost arbitrage, the organization loses momentum. In essence, the “ship” part of “ship-then-fix” yields results, but the “fix” part falls short, with too little follow-through in implementing new technologies that would further increase support-function efficiency.
Centralizing without effectively automating can dissatisfy stakeholders whose expectations are increasingly informed by the seamless, intuitive interactions now on offer in many consumer experiences. An executive who still must collect paper hotel records to send to an offshore expense-reimbursement team may not be very impressed, especially when her health insurer may allow her to scan and send her medical receipts in seconds using a smartphone app.
This is particularly true for the digital natives of Gen Z—defined as people born from 1995 to 2010, who have been exposed to the internet, social networks, and mobile systems from early youth—who have high expectations for interactions with companies as an employee, customer, and supplier. For example, many banks today benchmark their websites and mobile apps against technology-savvy companies such as ride-hailing services and streaming networks. These organizations, unburdened by legacy operating models, accomplish work through self-directed, agile teams that are better able to meet rapidly-evolving needs.
In comparison, companies that continue to rely on outdated, manual, and often cumbersome processes can appear even further behind on their digital transformation journey. Centralization and reduced costs alone are not likely to be enough for companies that fail to improve the experience for employees, suppliers, and customers.
We strive to provide individuals with disabilities equal access to our website. If you would like information about this content we will be happy to work with you. Please email us at: McKinsey_Website_Accessibility@mckinsey.com
Which comes first: centralize, or digitize?
Technologies such as robotic process automation, artificial intelligence, machine learning, and cognitive solutions are no longer experimental tools limited to early adopters. Instead, they are becoming the industry standard across a growing number of applications, business units, and functions. Together, they make fix-then-ship a more viable approach: companies can capture significant value early on, and dramatically improve the user experience for processes before centralizing them.
That said, many organizations still struggle with implementation. Recent studies by our colleagues suggest that very few legacy organizations have been successful in rapidly scaling up digital transformations.
The challenges to automating processes through digital fall into three key areas:
- Fragmented, nonstandardized underlying processes often lead to multiple technologies handling the same task, increasing costs and eroding the business case
- Highly complex, fragmented core technologies, such as legacy IT platforms and systems
- Inadequate resources and digital talent to drive transformation at scale
While these problems may be virtually universal, there is no one-size-fits-all solution to the issue of centralizing and automating functions. Both are needed, in the right sequence, to capture the full value at scale.
But which sequence is right? We find that companies with more standardized processes and integrated technology can often automate first, while others acutely experiencing the three core challenges to digital transformation can capture more immediate value by centralizing first.
Five essentials for digital transformation
Yet the ship-then-fix versus fix-then-ship decision resolves only part of the problem. In both situations, we find that organizations must tackle five critical prerequisites in planning, designing, and implementing a solution.
Conduct a baseline assessment of the value opportunity for digitizing key processes. For support-function leaders to have a meaningful discussion—grounded in facts—about the right sequence of priorities and investments, they will need a clear-eyed, objective view of the value at stake. Leaders might start by assuming that faster cycle times for billing processes would substantially reduce value leakage, but discover that accuracy matters even more for certain high-value customer segments. Accordingly, this kind of quantitative analysis not only gives the global business services organization a specific improvement metric to target (and a benchmark for measuring progress and holding leaders accountable), but also can lead to differentiated approaches in adopting and deploying technologies.
Treat scale as an enabler for digitization. Effectively capturing value from process digitization is often a matter of scale. Processes that are fragmented, narrow in scope, or operated out of multiple locations can limit the potential benefits that a company can generate through digitization—meaning rationalization and centralization might need to happen first. Conversely, processes that are newer, standardized, and not tied to a specific location are more attractive candidates for digitization. Commodity procurement, for example, may lend itself well to digitization, whereas for strategic procurement of critical supplies, creating a center of excellence where specialists can collaborate might matter more.
Standardize processes according to a common logic. Rather than simply digitizing processes in their current forms, companies can take the opportunity to eliminate needless—or even counterproductive—variation and inefficiency. If different regions or business units follow different policies for travel, for instance, the transformation program is a good time to examine why those variations emerged, and whether to continue them. In similar fashion, applying lean-management practices to a process can uncover value that digital can then multiply when deployed thoughtfully.
Stabilize core IT platforms and master data. Without the right IT infrastructure and data, even the best digital solution will likely prove to be a short-term fix at best. Instead, most organizations will likely want to stabilize and align their core systems—such as for enterprise-resource planning, customer-relationship management, and procurement—and integrate them with master data. This approach helps managers and senior executives generate data-driven insights in the future. Exceptions arise when a legacy IT system is mission-critical and the organization seeks to extend its lifespan: ship-then-fix may be the only feasible option.
Consider the organizational culture. As noted above, a fix-then-ship approach entails uncertainty and usually requires more upfront resources in both talent and capital. It also is more achievable with an organizational culture that is comfortable executing in the face of change, via an agile approach, rather than merely conducting business-as-usual operations. Some companies have this type of organizational culture in place are a more natural fit to apply a fix-then-ship sequence, and can reinforce the approach by aligning incentives to the degree of impact achieved against a bold objective.
Centralization first, via a ship-then-fix model, and automation first, through fix-then-ship, offer significant potential to improve the efficiency and service quality of support functions. Rather than treating them as an either-or decision, companies can expect to use each as needed, as they set their course for recovery and reimagination in the next normal. By understanding their starting point in each specific process and considering the five prerequisites defined above, organizations can be systematic in determining the right sequence, the right model, and the right outcomes for customers and the business.