Ian Herbert and Andrew Rothwell explain how shared services are more than just about headline cost savings, based on 10 years of research.
The shared services model has been a success story in both the private
sector and many parts of the public sector and is being actively
considered in higher education.
It is tempting for hard-pressed senior managers in higher education to
grab something from the management toolkit that seems to have been
proven to make worthwhile savings in private and public sector
organisations. The shared service model is one such tool. However,
there are number of more nuanced issues that need careful
consideration if benefits are to be realised.
Let’s kick off with five things you might not expect to read.
1. Think beyond immediate cost savings
While no financial director can be expected to sign off on a business
case in which benefits do not exceed cost, many of the real advantages
of shared services are both qualitative and serendipitous. It is always
very difficult to quantify savings because the services that are being
rationalised had previously been embedded in discrete business units
where the cost base was often opaque and subject to political posturing.
When we’ve asked the question “How much have you saved with the
shared service centre (SSC)?”, we’ve heard a variety of answers. While
there have been some big claims, we suspect the most honest answers
are something like, “We can’t be sure but after two years we have
around half the previous headcount doing more than they did before but
there was also a big implementation cost, some of which might arguably
have had to be spent anyway”.
2. Think big, but don’t try and do the “big plan”
This is counter-intuitive in an era of methodical corporate planning
systems. However, many successful SSCs start in small, sometimes
even happenchance ways, which establish the basis of mutual
understanding and longer term working relationships. Go for the lowhanging
fruit such as payroll (critical but relatively standard) and then
add further activities as appropriate.
3. Don’t assume a new computer will solve all the other problems
It is tempting to start shared services with a big investment that will
galvanise action and enforce common working. After all, technology is
the future and a new ERP system comes with all sorts of guaranteed
benefits. However, there are some notoriously bad examples where
social change has failed because the technological platform has failed or
there have not been the appropriate working relationships and
procedures established on the ground to make things happen in a
4. Contracts are bad, partnerships are good
Again, at the sign-off stage, senior managers want to see a nice
watertight contract that absolves them from accusations of sloppy
governance, even if they cannot sue for satisfactory legal redress. But
such thinking misses the big picture about shared services, which is
about continual mutual adjustment between the SSC and its customers
in the operating units.
All levels of management need to talk about partnerships and not
One of the most insightful quotes in our study came from a divisional
manager: “We spend a lot of time upfront negotiating service level
agreements. This helps to guide a discussion about what we need, what
the SSC can reasonably deliver and how much it will cost. But when the
new financial year starts the SLAs go into my drawer. If I have to take
them out and refer to the letter of the agreement then it means the
service has failed and we cannot allow that to happen. If we have got
things wrong we have to quickly adapt, not start threatening to ‘sue’ each
That’s the big difference compared with outsourcing. One large company
with 5,000 plus SSC workers has abandoned even SLAs, seeing them
as “anti-trust” systems.
5. Shared service centres are not just a “quick fix” – they need to be
Many of the big prizes in adopting the SSC model can only be realised
over time. For example, standardisation of processes and protocols
across a range of operating units is likely to save money. But, the real
rewards do not arise until the last 20% of activity actually becomes
standardised. A study of an SSC implementation in the Swedish justicesystem found that many of the planned benefits did not arise but other
largely unforeseen benefits did emerge, such as dissemination of best
practice, better information security, capturing learning from experiences,
predictability of costs and a reduction of overcapacity.
For more information, see the shared services section on the Efficiency
Ian Herbert and Andrew Rothwell are based at The School of Business
and Economics, Loughborough University. Their longitudinal research on
shared services has been supported by the Charitable and General Trust
of the Chartered Institute of Management Accountants.
You can learn more about this area in a shared services forum at the
University of Northampton on 14 May.