The Good, The Bad, and The Ugly of Shared Services Implementations
Shared services is not a new concept. Even the Egyptians centralized stone quarrying activity to facilitate the construction of multiple pyramids. For the modern-day world, the shared services concept started gaining traction on the business scene about twenty years ago as a tactic to reduce cost and streamline organizations. It was also an alternative to outsourcing. Why not in-source certain activities needed by an organization? Keep it under the company’s control-- but also take advantage of economies of scale through centralization. The functions typically targeted for shared services have been Supply Chain Management (SCM), IT, HR and Finance.
The expected outcome has been services offered at a lower cost with an equal or better service level. It didn’t stop there. Companies soon began looking beyond the cost and streamlining savings expecting more value: better service quality, faster turnaround, greater consistency (especially important in a quickly scaling enterprises) and ideally, increased business value. Additional functions being swept into the shared services movement included: communications, security, legal, facilities and other functional activities conducive to this approach, e.g., engineering services and product marketing. All of this sounds great on the surface-- but not so fast.
Lessons Learned from Shared Services Successes & Mistakes
Shared services implementations have a mixed track record of success. Depending on the industry, company and employee population surveyed, you’d likely discover that such an undertaking can also increase costs, add complexity and even slow internal processes and programs. The Hackett Group (consulting organization sampled 250 companies) which produced a report on this topic in 2009 noted that only about a third of all organizations that had completed shared services implementations were able to generate cost savings of 20% or more. In my view, even a cost savings of 10% would be a fantastic number if other targeted objectives were achieved too (meaning that the functional cost savings didn’t somehow negatively impact employee productivity or business viability elsewhere in the organization.)
However, after leading and/or participating in the implementation of shared services models in five global companies, I wave the red flag of caution and say: "Be careful what you do." I believe many implementations rob Peter to pay Paul, so to speak, and sometimes even create lasting disabilities in the fabric of an organization.
Reasons to Implement Shared Services
- Reduce cost. Unless you’re dealing in gold or diamonds, most of your cost is wrapped up in personnel. It’s likely that there is a cost advantage for some degree of centralization if designed and implemented correctly.
- Efficiency & Accuracy. Scaling a regional or global business that doesn’t utilize centralized services invariably winds up building in duplication, inconsistency and conflict of processes, programs and systems. It took SAP (70,000 employees worldwide) four years to unravel and true-up their HRIS system even though SAP makes and sells HRIS software. When I first arrived at this company I couldn’t generate a single global headcount report for key clients without directly calling each of the 23 country liaisons – in order to validate numbers. We were flying blind.
- Domain Expertise. A centralized service can generally offer more access to domain expertise. In a decentralized environment, few divisions or business units can afford the heavy-hitting compensation, IT or SCM process gurus. Having state-of-the-art experts (in central centers) can create a competitive advantage in Total Rewards, HCM processes and just about every area imaginable. Also, in a distributed model each specialization will likely have different levels of expertise, program timing and priorities. This can create huge operational conflict for companies. One part of the organization may emphasize people development and rotational assignments, another might not. The result is that employees will defect from one group (where they may be needed most) and jump to another for a better career trajectory.
- Continuity of Services: Shared services can promote ubiquity of programs, services and processes across a company. As organizations scale, there is nothing worse than conflicting infrastructure. It forces companies to become internally focused as they bog down in complexity – the reverse of what’s needed in a competitive, customer focused business environment. Tasks such as assimilating a new employee, transferring an employee from one group to the next, even administrating a performance review can become nightmarish for managers, employees and the support groups charged with facilitating these activities.
- Accurate Measure of Cost & Value: When certain services, programs and processes are distributed, it is very difficult to understand the true value and associated cost of the service. Centralization promotes common understanding, definition, service level agreements and associated metrics. This increases visibility into the cost/value equation of internal services.
I didn’t capture every upside for implementing shared services. I’m sure that you can add to the list. Bottom line: there are compelling benefits to considering this model. This is why many companies jump on the bandwagon without fully understanding that there are downsides, too. Some of these downsides can create organizational disabilities.
Cautions When Implementing Shared Services
- One Size Doesn’t Fit All: The underpinning of a Shared Services Model is efficiency and consistency. In practice the shared services leaders drive to make the service and program offerings vanilla or "we do it one way". It’s a logical path to economies of scale. Yet, most businesses when growing multi-nationally or globally require more than vanilla from support functions. If the shared services organization can’t accommodate variations on a theme in support of local business needs then a condition occurs that I call an "organizational disability". That is, hard driving business leaders expect support functions to be enablers – not energy sapping, procedurally heavy bureaucracies. And, they will find workarounds if they can’t get what they want from internal groups. Complex, confusing internal processes can undermine innovation, company spirit and productivity on levels that are hard to quantify but the impact on profitability can be substantial.
- Systems, Technology, Tools & Management Capability Must Support Centralization: Many times organizations jump into shared services only to learn that the appropriate infrastructure isn’t available to automate and support a central delivery mechanism. If the technology and infrastructure is lacking these aspects can bog down the shared services implementation for years and at the same time add greater complexity and unwanted drag to the growth and development of an company.
- Shift Administrative Burden and Other Duties to Employees and Managers: There is a clichê that comes with shared services implementations as local finance, IT and HR employees are eliminated when centralization is in effect: "These are tasks that managers and employees should have been doing all along." Wrong thinking. This may be true with regard to making sure managers complete employee performance reviews – and now that you have a central data base (thanks to shared services) the outstanding reviews can be identified and managers reminded to get these done. However, to think that a company is somehow saving money by shifting administrative burden to managers and employees doesn’t make sense. Anything that detracts from designing, building and selling products and services is generally bad for business.
- The Primary Purpose is Maximum Savings: Shared services offers a variety of competitive advantages. If the overriding objective is cost savings (very typical for companies to become myopic this way) then the intended effort will likely backfire. For example, one company implemented a preferred vendor program as the recruiting function was centralized. This restricted business units in specialized disciplines from using niche recruiting firms too expensive to fit the preferred vendor RFP. The result was increased time-to-fill, lower quality candidate slates and delayed project work. This infuriated hiring managers under pressure to deliver product; unable to hire the right expertise. The centralized recruiting function had saved money and was running at a lower cost with this preferred vendor program. Did this benefit the company overall? I would suggest that this approach, cost savings as the primary focus, created a business disability such that the company’s competitive position was compromised. What is really scary is that many companies don’t acknowledge this point. A short-term, bottom-line only focus is easy to justify when money is tight but this short-sighted approach can create dangerous disabilities that undermine productivity.
Shared services offers a cost-effective, efficient and consistent platform for service delivery. It is generally a good strategy to facilitate regional and global scaling. Just acknowledge that this is a paradigm shift. Leadership and those charged with implementing such a model need to invest time upfront to make sure that the intended result is achieved. The risks are too high to informally throw the dice and engage in this activity. In-depth thinking is required to make informed decisions about support function value and service levels.
I have developed and used the model below to provide a contextual framework for how to determine service delivery expectations and the ultimate organization design. As indicated in the graphic, it is possible to under invest as well as over invest in a service. The trick is to think deeply enough on the topic to understand the value-add of each service and the degree of investment needed. That way you’ll prevent a shared services organization from becoming too expensive and providing unnecessary services or the converse, too lean and creating organizational disabilities in the process.
For example, if every dime possible is squeezed out of an IT support function (under invested) you might wind up with all employees needing to become mini IT experts. I recently worked at a Silicon Valley based company that outsourced IT and the balance of the IT organization was moved into a shared services model. The projected savings was about ten million dollars. When the implementation was completed engineers started defecting to other companies because the IT environment was so arcane that it severely impacted work effort. Login times increased, equipment was substandard, servers were slow.
Did the company save money with this centralized IT initiative? I would suggest the reverse to be true. Employee productivity dropped, voluntary employee attrition increased and morale became negative. It is hard to quantify but I guarantee that the cost to the business exceeded the ten million dollars in savings – many times over. The IT disability is alive and well in this company today. Employees have devised an elaborate series of work-arounds to secure better equipment and needed IT support.
My parting suggestion: If you are leading or participating in a shared services implementation, get off to a good start by determining the value proposition of each function to be centralized. Map these programs, processes and services against the value proposition model depicted above. Vet your conclusions with leadership, service providers and end users. That way your project will start off on the right foot, with the proper expectations set and a higher probability that the best result will be achieved.