NDMA Home Page
Index of topics on this NDMA website
Search this NDMA website on Google
© 2024 N. Dean Meyer and Associates Inc.
Excerpt from www.NDMA.COM, © 2024 N. Dean Meyer and Associates Inc.

Shared Services

Use Case: Shared Services Versus Decentralization

overview of the trade-offs, and how to consolidate shared services

by N. Dean Meyer

[excerpt from the book, How Organizations Should Work]

Shared Services

Shared services promises many benefits, including cost savings, improved safety and quality, and synergies.

Nonetheless, many business-unit executives advocate decentralization -- for good reasons.

This executive summary provides an overview of why people want decentralization (the benefits), the (not-insignificant) costs, and what a shared-services organization must do to consolidate its function.

Definition: What Does "Shared Services" Mean?

While working with the IT leadership of a large company, I found everyone enthusiastic about the concept of shared services. Mary, the corporate CIO, was certain it meant complete centralization of the IT function. Her counterparts, business-unit CIOs, were equally certain it meant that Mary's corporate IT organization would treat them more like customers. Meanwhile, Eric, the CFO, understood it to mean chargebacks.

Actually, it's none of the above.

"Shared services" describes an organization that provides services to multiple groups of internal customers.

Technically, every internal service group is a "shared service" to its customers. But decentralized functions are only shared within their business units.

When people speak of shared services, they typically mean enterprise-level organizations. Instead of each business unit owning its own little group, business units buy some (or all) of the support services they need from a central "shared-services" group.

Any function (except product management itself) is amenable to shared services.

Typically, people think of business support functions such as IT, HR, Finance, Supplier Management (procurement), Facilities, Administration, Public Relations, etc.

In more advanced cases, the concept also applies to product-related services such as Engineering, Manufacturing, Logistics, Supply Chain, Marketing, and even Sales. Virtually all functions can be shared across product lines.

"Decentralization" is the opposite of shared services. It means breaking up a function into small groups, each serving a subset of the enterprise (such as a single business unit).

Benefits of Decentralization: What Internal Customers Want

Decentralization happens for a reason. Said plainly, business-unit executives choose not to do business with enterprise shared-services organizations, and instead insist on owning their own support functions.

They may have legitimate reasons, including the following:

  • Control: Business-unit executives are accountable for their bottom lines. So, naturally they demand authority over all their factors of production, including support functions.

    You wouldn't go to a shared-services grocery store unless you can control what you eat. That includes control of how much you spend, and what you get.

    Business leaders are no different. They know they can control the priorities of groups that report to them. But they may feel they have to beg, play politics, wait in line, or justify their requests to get what they want from an enterprise shared-services organization.

  • Strategic alignment: Local staff are better aligned with local strategies.

    What does everybody have in common? Administrative processes!

    While business units certainly share enterprise-level strategies, each of them signs up to deliver a unique component of those broad goals. They're unique in their missions and local strategies, and hence in their requirements for services.

    They may be open to shared services for what they have in common -- routine administrative needs. But if a shared-services organization prefers to work on enterprisewide needs, business units that depend on it may lose lucrative opportunities that are unique to them and their local strategies.

    Business leaders believe that a group reporting to them will be "closer to the business" and better understand their local strategies, and will be more willing to address their unique strategy-driven requirements.

  • Customer focus: When a decentralized function reports to you, they respect you.

    But business-unit leaders may feel that corporate staff treat them as a nuisance rather than a customer, or worse, as unruly children to be controlled.

    Business-unit executives will not do business with a shared-services organization that treats them as supplicants, nuisances, or victims. They know they deserve to be treated as respected customers. And they know they'll get that respect from people who report to them.

No matter how great the benefits of consolidation may be, business-unit executives are not going to willingly agree to shared services until these reasons for decentralization are addressed.

Having recognized the benefits of decentralization, let's examine the costs. Then, we'll look at how a shared-services organization can deliver what business-unit leaders rightfully want, as well as the many benefits of consolidation.

Costs of Decentralization

Decentralization increases costs and reduces quality for two reasons: reduced specialization, and fragmentation.

Reduced specialization: When staff are scattered among the business units, they cannot specialize as much as they could within a larger consolidated function.

Consider this IT example: Each business unit might have a small team of applications engineers to support their entire suite of systems. If those same staff were consolidated, they could specialize in the various types of applications, such as finance, customer, order-to-ship, human resources, etc.

Regardless of the function, a larger shared-services organization permits a greater degree of specialization.

When people focus on a single specialty, they accumulate more experience in their field of study and get good at what they do. They keep up with the pace of change in their discipline, enabling innovation. They improve their methods and tools, develop templates and reusable objects, and ultimately deliver higher quality at a lower cost.

Specialists perform better than generalists. It's that simple. That's why you wouldn't go to your family doctor (a "general practitioner") for surgery!

Also note that a greater degree of specialization provides more interesting career paths for technical professionals. It attracts top talent, and motivates people to succeed.

Fragmentation: Fragmenting staff inevitably fragments their services. For example, where IT is decentralized, it's not uncommon to find different financial, customer, and procurement applications in each business unit.

When solutions are fragmented, costs rise for many reasons:

There's duplication of sustenance (indirect) costs. It requires parallel training, product R&D, policy formulation, and business support functions.

There's also a duplication of work products, instead of sharing common components (even in customized solutions).

Fragmentation also foregoes many opportunities for savings through sharing licenses and assets.

Economies of scale are lost. This certainly applies to assets (infrastructure). But it affects staff too. Imagine three business units, each of which needs a certain specialty for roughly two-thirds time. Instead of each hiring their own person, the three could share just two people.

"Safety stocks" (in inventories) must be larger; consolidation can reduce inventory carrying costs. This affects staffing, too. One business unit's peak load may occur at a time when other business units are slow. Thus, total demand is less than the sum of each's peak.

And bargaining power with vendors is diminished. Almost always, a bigger organization can drive a better deal.

Beyond just costs, decentralization typically impairs the quality of service. Small groups cannot adequately support a broad, diverse product line. They find it difficult to provide support globally, 24 by 7. And small groups can't afford senior executives who bring strategic leadership to the function.

More insidious, synergies across business units are lost. Corporate synergies can be found in every external interface. Business units may share customers, vendors, investors, regulators, media, and the community. Synergies can be attained when they collaborate, and common solutions, services, data, and policies make that easier.

Fragmenting support staff undermines these enterprise synergies. As business units find themselves using disparate solutions and services, they find it harder to collaborate across boundaries. The implications for enterprise strategy are as profound as the reasons why these business units are under the same corporate umbrella in the first place.

Mini case study: decentralizing manufacturing....

Decentralization undermines that collaboration among business units. Here's one case example:

In the spirit of autonomy, each division in a global heavy-equipment company built its own manufacturing plant. This increased costs for a number of reasons:

Capacity sat idle in one plant while another was pressured beyond its capacity. Ultimately, lots of excess capacity ("safety stock") was built into the system.

One division had a plant in a country while another division did not. Instead of using that local plant, the other division shipped its products from its own plant in a neighboring country. As a result, transportation costs and duties rose.

One division pioneered new manufacturing techniques, but other divisions didn't learn about them until much later when the CEO intervened and commissioned a corporatewide task-force to rationalize production capacity.

And while there could have been significant economies if some plants were designed around long, stable manufacturing runs while others were designed for rapid re-tooling, none of the divisions alone could afford to specialize to that degree.

Mini case study: decentralizing design engineering....

In a heavy-equipment manufacturing company, each division also had its own design engineering function.

With all the reinvention, the number of parts the corporation had to make or buy skyrocketed. Some years later, an internal study found over 50,000 nuts and bolts in their parts catalog, many of which had the same functionality!

That same CEO-sponsored task-force found a dozen different electric motors with roughly the same specifications.

Costs of development, manufacturing, inventories, and support all were higher than they could have been.

Mini case study: decentralizing IT....

I personally experienced the effects of decentralization of IT as a customer of an insurance company.

The decentralization of its IT function resulted in multiple customer databases. Since customer-numbers varied, the enterprise was not able to spot customers who bought from multiple business units.

One day, the Specialty Automotive business unit cancelled the policy covering my vintage sports car, saying they were no longer interested in that type of business.

Annoyed, I moved all my insurance to another company -- the policies for my other cars, my home, my personal liability umbrella, and my corporation.

Of course, they never knew this. Because their customer databases were fragmented among their business units, they knew me as a few discrete policies, not as one customer with diverse needs.

Benefits of Shared Services

The benefits of shared services are the mirror image of the costs of decentralization, including all the advantages of specialization, economies of scale, and cross-business-unit synergies:

  • Productivity
  • Quality
  • Speed
  • Innovation
  • Career paths that attract talent
  • Savings by elimination of redundant efforts
  • Economies of scale
  • Reduced safety stocks
  • Bargaining power
  • Quality of service
  • Enterprise synergies

The bottom line is, a shared-services organization should be able to offer better internal products and services at a lower cost than business units can afford to do for themselves. And that applies to business-unit-specific (custom) as well as enterprise solutions and services.

The Struggle

In theory, shared-services should always outperform decentralization. But unfortunately, in practice it doesn't always do so.

Some corporate groups see their role as controlling business units, for example, limiting business units' spending on a function or forcing one-size-fits-all solutions on them inappropriately. These corporate functions may be shared, but they're a far cry from being service oriented.

Sometimes, shared-services organizations don't have people whose job is to deeply understand business units' strategies, operations, and personalities. As a result, they're in a poor position to deliver strategic value.

Sometimes, shared-services organizations set their own priorities.

In these situations, business units rightfully prefer their own internal service groups, even if it costs more. At least they get the job done.

Of course, if corporate staff then demand a monopoly -- that is, force business units to buy from them "for the good of the company" -- resentment builds and the pressures for decentralization increase.

Amidst this tension, some corporations go to great lengths to remain decentralized and "patch" the problems that creates. They talk of "federated" functions; they impose "governance" in the form of committees and working groups; and they try to hold enterprise shared-services executives accountable for things they cannot control.

Some corporate departments retreat to the safe territory of "common solutions" -- e.g., in IT, the network, computer center, and corporatewide applications like ERP -- and leave the rest to business units. This narrow vision certainly will not achieve the promise of shared services.

Ultimately, these patches are ineffective. The right answer is consolidation. But only if business-unit leaders' concerns are addressed.

How to Go About a Shared-services Consolidation: The Hare Versus the Tortoise

For shared-services consolidations, executives have a choice of approach. Consider two alternative strategies:

A true story....

Pierre, the CIO of a Fortune-100 sized pharmaceutical and medical device conglomerate, saw the huge potential cost savings that could be gained by consolidating the currently decentralized IT infrastructure.

Using a reputable consultant, he developed a compelling financial case. He personally invested his time and "credibility chips" in selling it to the corporate CFO, CEO, and ultimately the Board of Directors.

After months of effort, Pierre's plan was approved. IT infrastructure was consolidated -- over the objections of the business units. Two months later, Pierre resigned "for personal reasons." He won the battle and lost the war.

A happier story, equally true and a stark contrast in executive style....

During the same time frame, Preston, another CIO of a Fortune-100 pharmaceutical and medical device conglomerate, worked on a different strategy.

Preston said, "I'd rather have a customer than a hostage!" He focused on building a shared-services IT business-within-a-business that earned clients' business through performance, value, and great customer relationships.

He raised the opportunities of consolidation to executives' attention, but with the caveat, "I've got plenty to do as it is," he said. "But if you'd like to save money and improve the quality of service, we'd be happy to bid your business."

Through voluntary consolidations, the "market share" of corporate IT grew from approximately 40 percent to over 80 percent of total enterprise IT spending. And as promised, Preston delivered better quality services at lower costs with each consolidation.

The Hare

One way to establish shared services is through edict. A corporate mandate may force consolidation over the objections of business-unit executives. This was Pierre's strategy.

This may appear to be the quickest approach (the hare in Aesop's proverb); but in the long run, it rarely works. It sets up an antagonistic relationship between the shared-services organization and the internal customers it's supposed to serve. That, alone, undermines its effectiveness.

Meanwhile, it drives decentralized work underground, for example, engineers doing IT work, and administrative staff doing HR work. Eventually, the pendulum swings back to decentralization because business leaders still don't like doing business with that corporate function.

Perhaps worst, shifting work to an organization that's not ready to deliver great service forces business units to do without services they had counted on. This can undermine the performance of the entire enterprise.

The dangers of a forced consolidation based on the numbers include:

  • Complete failure of the consolidation process due to the overt resistance of business executives.

  • Worse, failure of the resulting organization to deliver on promises due to the covert resistance of business units and its own poor performance.

  • A breakdown in relationships which inhibits the ability of the shared-services organization to partner and deliver strategic value.

These consequences can easily overwhelm any benefits of shared services.

Even if these risks are overcome, ultimately the benefits may not be delivered any faster this way. Savings and synergies are delayed due to the confusion and strife caused by the politics, and the poor performance of the shared-services organization.

The Tortoise

A safer and more effective approach -- Preston's strategy -- is to transform an existing enterprise function into the "supplier of choice" to the business units. Then, as a high-performing business within a business, it earns "market share" over time as it offers business units better services at lower costs.

"If you build it, he will come."
Field of Dreams (1989)

An internal supplier of choice can be characterized by the following philosophies:

  • We'd rather have a customer than a hostage; we want to earn market share through performance and relationships, not command it through monopolies.

  • We're not empire builders; we're not the advocate of consolidation, but we're happy to help if it will save you money and improve quality of service.

  • We're willing to work through, not around, your decentralized staff, as their outsourcing vendor; you won't lose control.

  • We're not talking about business-process harmonization; that's a business decision. We're just suggesting shared services (even for custom solutions), and consortia when clients do agree to share solutions.

  • We promise to deliver the same or better levels of service, at the same or lower costs.

Once a shared-services organization addresses the reasons people prefer decentralization, they'll be positioned to take on more business.

Services will be consolidated when it makes sense to do so -- that is, when the shared-services organization can deliver the benefits, and when business-unit executives are comfortable letting go of their decentralized functions.

This way, consolidation occurs at the right time, to the right degree, and with minimal political stress.

To take this safer and more effective path, a shared-services organization must become a high-performing business within a business before it actively pursues consolidation opportunities. But it's going to have to accomplish that transformation eventually in any case. The "tortoise" approach just gets the sequence right.

What Shared-services Organizations Must Do to Earn the Business

If you consider the reasons people advocate decentralization, all can be addressed by a high-performing, customer-focused shared-services function.

Here's what shared-services organizations have to do to overcome the forces that cause decentralization:

  • Control: You don't need to own your own grocery store to control what you eat because you control your checkbook.

    The same is true of enterprise shared-services functions. Business-unit executives must be able to control what they "buy" from internal service providers.

    This does not require committees and bureaucracy; and it does not require chargebacks. What it does require is resource-governance processes based on market economics.

    Customers want:

    • An understandable service catalog that defines what they get (not what the supplier does).

    • Accurate cost-based rates (whether charged to customers or to the service-provider's budget).

    • Input on the shared-services organization's budget to advocate for the deliverables they need.

    • If the shared-services organization is given direct budget or allocations (as distinct from fee-for-service chargebacks), control over its priorities.

    • Great value. Its rates (unit costs) must be lower than those of decentralized groups (fully burdened, to compare apples to apples). Ultimately, it must be a better deal than outsourcing.

  • Strategic alignment: You're unique.

    Shared services must be willing to deliver solutions tailored to customers' unique requirements (not "one size fits all"), whenever the benefits justify the costs.

    To do this, shared-services organizations need to deeply understand their internal customers' businesses and strategies.

    Customers want:

    • A liason dedicated to them who specialize in knowing their business, their people, and their strategies.

    • Help discovering strategic opportunities for the organization's products and services which contribute to (or even enable) business-unit strategies. (In IT, these are captured in the phrase, "digital enterprise.")

    • Unique products and services that directly address their unique needs, while maximizing synergies through shared components, data, and talent.

    • Help forming consortia with other business units to buy shared processes and solutions (e.g., in IT, a shared ERP application).

    • Innovation in its service offerings to create future strategic opportunities.

  • Customer focus: You're my valued customer.

    It must treat clients with respect (as customers, not nuisances or victims) -- the essence of customer focus.

    Beyond that, it must invest in highly effective relationships with its business-units customers.

    Customers want:

    • Impeccable delivery of every commitment. Operational services must be available, reliable, and safe. And projects must be delivered on time and on budget, all the time.

    • To be treated as a customer. This must go beyond just staff's style of interaction with clients. Shared-services staff must serve clients needs, without an agenda of their own. They must never say, "We know what's best for you."

    • No judgments. A shared-services organization cannot judge and control its customers in any way, and then expect to be considered their partner and vendor of choice. You just can't say, "I'm from the IRS and I'm here to help!"

Essentially, the shared-services organization must be a highly effective business within a business.

Get to Root Causes

Wait! Don't start work on the above list of requirements!

Don't fix symptoms (or you'll find yourself fixing them again and again). Get to the root causes, the organizational systems that make or break an organization's performance.

Here are the five organizational systems, and what's needed in each to succeed at shared services:

  • Structure: both the organization chart and workflows. This provides:

    • Definitions of the lines of business of each group, which is the basis for service catalogs and accountability for results.

    • Cross-boundary workflows that bring together any needed specialists to every project or service. (It does no good to consolidate staff and then leave them substructured by business unit.)

    • Relationship managers of the appropriate stature, dedicated to business units. This is, essentially, a Sales function. Relationship managers are responsible for a healthy relationship between business units and shared-services staff, and for aligning the shared-services organization with the strategic needs of the business units.

  • Internal economy: resource-governance processes including budgeting, demand management, and reporting. This provides:

    "I manage unit costs.
    You manage volume."
    Preston Simons

    • Cost control at the unit-cost level (not control of total spending which is driven by customer purchase decisions).

    • Investment-based budgeting: a business and budget planning process that forecasts the costs of what you plan to "sell," not just what you plan to spend.  

    • Accurate rates for services in the catalog, at full cost (with fair share of indirect costs) but without burdens for any enterprise costs that external vendors wouldn't include in their prices. (This isn't difficult if you implement investment-based budgeting.)

    • Business-driven demand management (priority setting) processes for the entire portfolio (not just major projects).

    • Explicit channels of funding for innovation.

  • Culture: a clear definition of how we work. Beyond values, this includes pragmatic operating principles in themes such as:

    • Customer focus (pleasing customers)

    • Entrepreneurship (remaining competitive)

    • Integrity (the behaviors that inspire trust, including impeccable delivery of every commitment)

    • Interpersonal relations (working well with other people)

    • Teamwork (collaborating on projects and services)

    • Empowerment (authorities and accountabilities match)

    The good news is that best-practices behavioral principles in all these themes are well defined, and changing culture really isn't all that tough or time consuming.

  • Methods: the skills, processes, and tools by which staff deliver work. Key examples relevant to shared services include:

    • Opportunity discovery, to help business units discover ways to leverage their strategies, or enable new strategies, using the products and services of the organization.

    • Benefits estimation, including strategic value, to assist in justifying and prioritizing strategic investments.

    • Consortium facilitation, to help multiple business units collaborate to share common solutions and information.

    • Commitment (contract) tracking, and excellence in project management, to ensure reliable delivery of all projects.

    • Service management to ensure reliable, safe, efficient, and relevant services.

  • Metrics and rewards: the dashboards and incentives that give people feedback and incentivize performance. This reinforces all the above influencers of behaviors.

A Shared-services Strategy

In summary, the place to start work on shared services is not analyzing potential cost savings via consolidation. It's an organizational (transformation) strategy that develops the kind of internal service provider which business units want to do business with -- their internal "outsourcing" supplier of choice.

"If a man can write a better book, preach a better sermon or make a better mouse trap than his neighbors, though he builds his house in the woods, the world will make a beaten path to his door."
Ralph Waldo Emerson (1882)

As the shared-services organization becomes better and better at serving its internal customers, it can "bid" services that replace decentralized functions whenever the opportunities arise. With this approach, business units don't feel threatened about a corporate "take-over." They're simply looking at opportunities for better services at lower costs.

This is exactly what Preston did. "Build it, and they will come," he believed. And as it turned out, they did.

It's All About Leadership

How you get there is a matter of leadership. The choices might be framed as: political hard-ball, settle for the obvious shared services that have to exist at the enterprise level, or invest in a high-performance business within a business.

Clearly, the third choice is the right choice in the long run.

A leader with a vision of a high-performance business-within-a-business can deliver all the benefits of decentralization along with all the benefits of consolidation. The key is a shared-services organization that earns clients' business thanks to its structure, internal economy, culture, methods, metrics, and of course talent. It's the right thing to do.


Free library


Speech abstracts

NDMA coaching/consulting services