it costs in banks: revisit your beliefs! · tion. with our it benchmarking report we intend to...

M a i2 0 0 4

BCG The Boston Consulting Group

BCG's 2003 European IT Benchmark ingin Banking Study

IT Costs in Banks:Revisit Your Beliefs!

The Bos ton Consu l t ing Group i s a genera l management consu l t ing f i rm tha t i s a g loba ll eader in bus iness s t ra tegy. BCG has he lped compan ies in eve r y majo r indus t r y and mar -ke t ach ieve a compet i t i ve advantage by deve lop ing and implement ing winn ing s t ra teg ies .Founded in 1963, the f i rm now opera tes ove r 60 o f f i ces in 37 count r i es . For fu r the rin fo rmat ion, p lease v i s i t our Web s i te a t www.bcg .com.

© 2004 The Bos ton Consu l t ing Group GmbH. A l l r i gh ts rese r ved .

For in fo rmat ion and repr in t au thor i za t ion p lease contac t BCG a t the fo l l owing address :

The Bos ton Consu l t ing GroupMarke t ing & Communica t ions /Lega lLudwigs t raße 2180539 MunichGermany

Fax : +49 (0)89 2317-4718E-Mai l : marke t ing [email protected]

1

P R E F A C E 5

E X E C U T I V E S U M M A R Y 7

A B O U T T H E S T U D Y — M E T H O D O L O G Y 1 1

A b r i e f d e s c r i p t i o n o f o u r a p p r o a c h 1 1D i s c u s s i n g t h e r a t i o s 1 2

R a t i o s b y v o l u m e 1 2R a t i o s b y r e v e n u e s 1 3R a t i o s b y c o s t b a s e 1 4

C o m p a r a b i l i t y o f f i g u r e s 1 4I n f l u e n c e o f r e g i o n a l d i f f e r e n c e s 1 5I n f l u e n c e o f b u s i n e s s m i x 1 6

N I N E F I N D I N G S F R O M T H E B E N C H M A R K I N G 1 9

F i n d i n g 1 : I T c o s t s c o n t i n u e t o r i s e 1 9F i n d i n g 2 : U s e r s a t i s f a c t i o n d o e s n o t r i s e w i t h I T s p e n d i n g 2 1F i n d i n g 3 : L a r g e d i f f e r e n c e s i n l e v e l s o f I T s p e n d i n g e x i s t 2 2F i n d i n g 4 : T h r e e d i f f e r e n t p a t t e r n s o f I T s p e n d i n g i d e n t i f i e d 2 3F i n d i n g 5 : H i g h e r I T s p e n d i n g d o e s n o t t r a n s l a t e i n t o b e t t e r o p e r a t i o n a l e f f i c i e n c y 2 4F i n d i n g 6 : H i g h e r I T s p e n d i n g d o e s n o t l e a d t o m o r e e f f e c t i v e n e s s 2 7F i n d i n g 7 : A d o p t i n g s t a n d a r d a p p l i c a t i o n s d o e s n o t ( y e t ) g e n e r a t e c o s t a d v a n t a g e s 2 8F i n d i n g 8 : M a i n t e n a n c e c o s t s t e n d t o i n c r e a s e w i t h h i g h e r I T s p e n d i n g 3 1F i n d i n g 9 : S c a l e i s n o t b e i n g t r a n s l a t e d i n t o c o s t a d v a n t a g e s 3 2

I T C O S T S I N B A N K S : R E V I S I T Y O U R B E L I E F S !

T H R E E L E V E R S T O I M P R O V E I T P E R F O R M A N C E 3 5

L e v e r 1 : I m p r o v e b u s i n e s s a n d I T s t r a t e g y a l i g n m e n t ! 3 6L e v e r 2 : T i g h t e n I T p e r f o r m a n c e a n d v a l u e m a n a g e m e n t ! 3 6L e v e r 3 : I n c r e a s e l e v e l o f s t a n d a r d i z a t i o n ! 3 8

I n c r e a s e p o w e r o f c e n t r a l C I O o r g a n i z a t i o n ! 3 9C o n s i d e r s e l e c t i v e o f f s h o r i n g a n d o u t s o u r c i n g ! 3 9I n c r e a s e l e v e l o f I T a r c h i t e c t u r e s t a n d a r d i z a t i o n ! 4 1I m p l e m e n t s t a n d a r d d e s k t o p s e r v i c e s f o r a l l e m p l o y e e s ! 4 1

G L O S S A R Y 4 3

T H E A U T H O R S 4 7

2

5

P R E F A C E

IT and IT cost structure play an increasingly decisive role in the performance of every bank. During the lastfew years, banks spent more and more money on IT: today IT consumes some 10% of total bank revenues.This fact clearly shows that efficient and effective IT management constitutes one of the most significantprofitability drivers a bank can address. However, for many banks the cost structure and performance of ITlie well hidden within a "black box." Very few know the exact size and contents of this box, thus many poten-tial levers to increase the level of IT efficiency and effectiveness go unused. This has immediate, damagingconsequences to the overall performance of the whole business.

To demystify this mysterious black box, CIOs must first develop a reasonable estimate for a fair IT cost posi-tion. With our IT benchmarking report we intend to provide figures to help CIOs determine this position.Eleven leading European banks provided data, which was collected in the second half of 2003. The reportcovers actual business and IT data for 2001 and 2002 as well as provides 2003 estimates. These numbers sure-ly reflect the difficult business environment banks experienced during the last three years, and the figuresshould always be viewed within their specific business context. We plan to conduct a study on IT costs inbanks every year henceforth, so we will soon be able to see more clearly how business environment affects ITcosts in the long run.

The results of the study were surprising. The data proved wrong some common beliefs, including the ideathat above-average IT cost ratios mean outstanding value creation: banks with relatively high IT costs com-pared to their revenues did not outperform their competitors in terms of effectiveness and efficiency.

The insights gained from this benchmarking study, our own research, and experiences from our work withclients lead us to the conclusion that it is a well-defined IT strategy and managerial competence that leads tovalue creation, not the level of IT investment itself. Strategic management of IT is the key to controlling ITcosts. We have identified three critical levers for improving IT efficiency and effectiveness in banking—but,as always, they will only have a positive impact upon implementation when the unique situation of the indi-vidual bank is taken into consideration.

DDrr.. RRaaiinneerr MMiinnzz

Senior vice presidentMMrr.. HHeeiinnzz MMöölllleennkkaammpp

Vice presidentMMrr.. RRaallff DDrreeiisscchhmmeeiieerr

Vice presidentMMrr.. FFrraannkk FFeellddeenn

Manager

7

A b o u t t h e S t u d y — M e t h o d o l o g y

■ In the second half of 2003 we asked 11 top-tier European banks from four countries to answer a ques-tionnaire on business and IT data.

■ Additionally, we asked selected participants to fill out IT management scorecards (primarily qualita-tive scorecards concerning the IT value chain) and conducted surveys with internal IT customers onuser satisfaction.

N i n e F i n d i n g s f r o m t h e B e n c h m a r k i n g

■ Finding 1: IT costs continue to rise. IT costs have reached an all-time high: from 2001 to 2002, shareof IT costs1 increased by a percentage point—from 15% to 16%. Banks anticipate an even more dra-matic increase in 2003: share of IT costs is expected to increase to 20%. A typical bank pays some 24basis points of total assets and spends €40 annually per customer on IT. About 10% of a typical bank'stotal revenues—that is, roughly €17,000 per employee—are consumed by IT.

■ Finding 2: User satisfaction does not rise with IT spending. Although IT spending in banks is signif-icant, internal customers are still not satisfied with the services provided. The overall quality of ITservices is consistently ranked "average." In particular, customers named three areas in which ITshould improve: business and IT strategy alignment, first-level user-helpdesk responsiveness, andmission-critical application availability.

■ Finding 3: Large differences in levels of IT spending exist. We found that there is a huge gap betweenbanks with higher IT costs per revenues (high spenders) and those with lower IT costs per revenues(low spenders): for high spenders IT costs per revenues is 60% higher, and IT costs per operativeexpenses is 54% higher.

■ Finding 4: Three different patterns of IT spending identified. Our analysis reveals three differentinvestment patterns in the European banking industry. We've used these patterns to categorize banksinto three groups: The first group, aggressive investors, heavily increased spending on IT, although

E X E C U T I V E S U M M A R Y

1 Within this report, the term share of IT costs is used for IT costs as a percentage of operating expenses.

revenues could not match that increase in the same measure. A second group, moderate investors,showed a fair increase in IT costs combined with stable revenues. Most banks from the sample are inthis group. The third group, the active cost savers, is the only one that decreased IT costs relative torevenues.

■ Finding 5: Higher IT spending does not translate into better operational efficiency. High spendershave a higher overall cost-income ratio (61% in contrast to 59% for the whole sample) and fraction-ally less revenue per employee (€181,000 compared to €182,000 for the whole sample). Obviously,higher IT budgets do not help reduce the overall operating expenses for these banks, nor do theyhelp increase revenues. There is little evidence that high spenders currently create value by increas-ing efficiency.

■ Finding 6: Higher IT spending does not lead to more effectiveness. High spenders in fact generatelower revenues per assets. They could produce revenues of only 190 basis points of assets comparedto 260 basis points for low spenders. There is no evidence that high spenders can transform ITexpenses into competitive advantage and outperform their peer group.

■ Finding 7: Adopting standard applications does not (yet) generate cost advantages. High spendersusually use more standard software. This structural difference can be seen in the significantly lowershare of IT costs for participants using mostly custom-developed software in contrast to banks with amixed structure. In discussions with the benchmarking participants we saw that standard softwaresystems were often installed quite recently. In the short term, then, old systems—already written offin the accounting ledgers—are often more cost efficient for the bank. But the multitude of supportand auxiliary systems added to monolithic core systems over time has created complex interfaces, adifficult-to-maintain job network, and manual-update processes. The resulting "application jungle"is getting more and more difficult to maintain every year and increases the bank's operational risk.Therefore, CIOs expect investments in standard software to pay off in the long run.

■ Finding 8: Maintenance costs tend to increase with higher IT spending. High spenders have main-tenance costs of about one-third of their total application development budget, compared to onlyabout one-quarter for low spenders. This can be attributed at least in part to the fact that during thetransition phase, banks often have to run several operational systems in parallel. The vast majority ofprojects for high spenders fall into the categories, "production, internal processes" and "infrastruc-ture"—project categories with the highest share of maintenance projects. In contrast, low spendersuse 43% of their project budgets in the area of "customer, sales, and marketing." This is 60% morethan the 27% for high spenders.

■ Finding 9: Scale is not being translated into cost advantages. There is no correlation between assetsand IT costs per assets. None of the larger benchmark participants manages to realize economies ofscale.

8

T h r e e L e v e r s t o I m p r o v e I T P e r f o r m a n c e

■ Lever 1: Improve business and IT strategy alignment! There appears to be a mismatch between busi-ness strategy and IT strategy. Many banks still do not maintain a central IT project portfolio, arequirement for prioritizing IT projects based on their total business value. There is often no groupcap on development projects—that is, business units do not have to compete for funding. The busi-ness case for an IT project should be based on exactly the same measures as business projects, and ITmust be explicit about the benefits to the business. Therefore, a key lever for improving business andIT strategy alignment at the business-unit and corporate levels is to introduce well-defined alignmentprocesses and to set up joint committees with members from corporate and decentralized IT as wellas the business side.

■ Lever 2: Tighten IT performance and value management! Most banks have only a low degree of ITcost transparency. There are four basic steps to improving IT performance and value management.After business and IT goals have been aligned, an information model must be developed for keydecision makers. Performance metrics must then be developed and linked to staff compensation lev-els. The key is to focus on simple, meaningful, tangible, and measurable metrics that can be easilyderived across the whole organization.

■ Lever 3: Increase level of standardization! The complexity of banks' IT landscapes has grown signif-icantly over recent years, resulting from heavy investment in application development as well asmerger and acquisition activity in the European banking sector. Maximizing standardization is anabsolute "must" for banks to undertake if they are to reduce IT costs. IT functions should seek toincrease the level of standardization across their organizations. BCG has identified a four-point "stan-dardization map" to assist CIOs in approaching this challenge:

- Increase the power of the central CIO organization

- Consider selective offshoring and outsourcing

- Increase level of IT architecture standardization

- Implement standard desktop services for all employees

9

1 1

A b r i e f d e s c r i p t i o n o f o u r a p p r o a c h

One of the key questions in dealing with IT costs in banks is how to determine a fair cost position. The com-mon scientific approach of calculating what the increase in operating costs would be if a bank had no IT atall does not work: today IT does not primarily constitute a means to running the business more efficient-ly—that is, decreasing operating costs. On the contrary, for a long time now IT has been an enabler, offer-ing products and processes not possible without IT. Therefore, the common approach of focusing exclu-sively on operating costs is doomed to fail.

A more useful and pragmatic approach is to compare banks within a consistent peer group and see howother banks succeed in managing their IT costs. Naturally this approach must address several concernsabout comparability within the sample. We address these issues later in this section (see the subsection"comparability of figures").

A B O U T T H E S T U D Y — M E T H O D O L O G Y

DATA COLLECTION APPROACHIn the second half of 2003 we asked 11 top-tier European banks from four countries to answer a question-naire on business and IT data. Additionally, we asked selected participants to fill out IT managementscorecards (primarily qualitative scorecards concerning the IT value chain) and conducted surveys withinternal IT customers on user satisfaction.

Participants received detailed feedback going well beyond the scope of this report, including both a fulldata appendix showing the individual cost position of all participants and specific feedback from us (seeExhibit 1). In addition, we provided analyses on the business-segment level to all participants deliveringthe relevant data.

D i s c u s s i n g t h e r a t i o s

Obviously, comparing absolute IT costs within a sample of different banks does not make sense, thus ourpragmatic approach works with IT cost ratios. Selecting meaningful reference values for these ratios is con-sequently a decisive issue. There are generally three areas from which to choose these values, each facili-tating a different view of the IT cost position:

■ Ratios by volume reflects the common notion that more customers, more assets, more accounts, etc.will drive IT costs. Since volume does not change quickly (outside of major changes in the businessportfolio such as mergers and acquisitions), it is a pretty good base for long-term IT cost figures suchas IT costs per customer. Banks use this volume to generate their annual revenue stream.

■ Revenue streams are much more volatile, changing mid-term with economic patterns, interest rates,and individual success in the market. Nevertheless, examining ratios by revenue reveals what part ofa bank's revenues is being consumed by IT activities.

■ Finally, one can use ratios by cost indicators, such as general administrative costs or number ofemployees, to determine the importance of IT costs compared to other cost categories.

R a t i o s b y v o l u m e

When it comes to determining the right level of IT costs, size does matter. The "bigger" the bank, onewould obviously think, the higher the absolute IT costs. One would also expect some economies of scale,that is, reduced IT unit costs (we will take up this topic in finding 9).

1 2

Source: BCG's 2003 European IT Benchmarking in Banking Study

Kickoff

Data collection

Workshop

Survey

Data validation

Individual feedback

G E N E R I C D A T A C O L L E C T I O N A P P R O A C H

E X H I B I T 1

But how can a study measure the size of a bank? Today the banking business consists of numerous, differ-ent business lines. In some European countries it is even possible to combine all these business lines with-in a single legal entity—for example, the German "Universalbank." However, no common volume measurefor all business lines exists. Therefore, for this study we worked with six business segments covering theentire banking business and used them to map specific banks' businesses within an independent frame-work. For each segment, we identified meaningful volume indicators and set the segment-specific IT costsin relation to one of the volume drivers. For example, in retail banking we used IT costs per customer andin private banking IT costs per euro assets under management (see Exhibit 2).

The variation of these ratios shows how much an average bank spends on IT (for example, per customer)and what is best practice within the sample. If the ratio exceeds the average on a long-term basis, IT man-agers ought to prove that additional value is being created for the business. Furthermore, these ratios com-prise a good indicator for executives looking for a reasonable, long-term target for their IT costs.

R a t i o s b y r e v e n u e s

This group of indicators shows what portion of revenues IT costs consume. In this study, the term revenuesis used identically to net income before operating expenses and equals the sum of net interest income, netfees and commission income, dealing profits before expenses, and other operating income. It is importantto use only net income figures since changes to, say, the market interest level, should not influence the ITratios by revenues.

For the entire bank, the term IT cost per total revenues is, in this text, referred to as IT cost-income ratio,or ITCIR. This concept shows what share of its revenues a bank has spent overall during the last year on IT.In contrast to ratios by volume, ITCIR always reflects the mid-term business success of a bank. In order to

1 3

Segment

Total assets

Retail banking

Definition of segment Volume drivers

Mass retail and affluent customers with up to 250,000 private assets,and enterprises with less than 5 million to 10 million revenues per year

€

€ €

Corporate banking

Private banking

Assetmanagement

Capitalmarkets

Transactionbanking

Enterprises with more than revenues per year,multinationals, institutions, and governments

€ €5 million to 10 million

Private wealth management for wealthy individuals and familieswith more than 250,000 private assets€

Management of mutual funds and institutional mandates(in Germany: special funds)

Trading and sales, mergers and acquisitions, and new issues orprimary markets

Payments processing, securities transactions and/or securities depositsas well as custody business for external customers

Customers

Assets undermanagement

Revenues Value at risk

Weightedtransactions


V O L U M E D R I V E R S S H O U L D B E D E F I N E D O N B U S I N E S S - S E G M E N T L E V E L

E X H I B I T 2

compare a bank's IT costs within its peer group, we need to consider the actual performance of a bank inits given market environment. Or stated differently, ratios by volume indicate what IT budget a bank oughtto spend, whereas ratios by revenues dictate how much a bank is able to spend. ITCIR can be broken downby the same business segments as discussed above, thereby demonstrating the very important difference IThas per euro of revenues for, say, retail banking and capital markets.

R a t i o s b y c o s t b a s e

This third group of indicators can be used to understand the importance of IT costs compared to othercost categories. One important measure is to relate IT costs to operating expenses. Note that this cost cat-egory already includes IT costs, therefore it is called, in this text, share of IT costs. However, one shouldkeep in mind that operating expenses do not equal total costs since operating expenses do not contain costcategories directly deducted from net income, for example, interest expenses. (See the glossary for moredetails on IT costs and operating expenses.)

Since the cost base is usually more stable than revenues, ratios by cost base is a better choice for long-termcomparison than ratios by revenues. On the other hand, ratios by cost base always contain some efficiencyaspect: a high share of IT costs could mean that inefficient IT is adding significantly to operating expens-es. But it could also mean that an especially powerful IT is enabling the business to run very efficiently,thereby significantly reducing the non-IT part of operating expenses and increasing the share of IT cost.Because of this, ratios by cost base must always be analyzed in combination with absolute figures or otherIT ratios.

Some other ratios relate IT costs to parts of the cost base. One key indicator is IT costs per personnel costs,which assumes higher IT costs should be reflected in higher back-office productivity and thus lower person-nel costs. It usually uses the number of employees as a substitute for personnel costs. Another key measurecompares the cost of IT operations to ongoing operating expenses, trying to blank out investment patterns.

C o m p a r a b i l i t y o f f i g u r e s

Attempts to analyze and describe a fair IT cost position often meet with skepticism when comparing ITcosts by ratios at all. The key point is that only a homogenous participant sample allows a valid comparison.Therefore, the banks within our benchmarking sample had to qualify for participation by matching at leastone of the following three criteria:

■ European player not just focused on the home market

■ Bank with leading position in its home market

■ Bank with total assets over €250 billion

1 4

1 5

Although this study offers a valid comparison of IT costs across banks, readers should bear in mind that thesmall sample size (leading European banks only) does not provide statistically accurate data in a scientificsense. Nevertheless, we believe one can still see major trends and draw very interesting conclusions in apragmatic way. All banks meeting the above-mentioned criteria have the same complex business require-ments for products, processes, market approach, processing volume, covered business segments, and thelike. Since business requirements drive IT requirements, we expect that all the banks in our study have sim-ilar challenges in dealing with IT complexity.

To avoid gathering the "wrong" data through participants having different notions about key terms such asIT costs, application development, and so forth, we provided them with an extensive, consistent set of defi-nitions. We also supported them interactively during the data collection phase to ensure a common under-standing of the figures. The most important terms used in the benchmarking are defined in the glossary.

Even in a sample of leading European banks there are still differences within the group of participants thathave to be taken into account. One is size. We had expected to see some economies of scale within the sam-ple. We discuss our conclusions regarding influence of bank size in finding 9. We discuss two other influ-encing factors in the remainder of this section.

I n f l u e n c e o f r e g i o n a l d i f f e r e n c e s

Regional differences have a strong impact on ratios by revenues. In the 2003 sample, average ITCIR of themost expensive country was twice as high as average ITCIR of the least expensive. The main reasons forthose differences are local variations in the following:

■ Purchase power and salaries

■ Value-chain depth—in particular, use of outsourcing

■ Regulatory requirements—in particular, reporting requirements

■ Tax laws—in particular, V.A.T. and options for deduction

■ Market structure and level of competition—in particular, degree of concentration and the difficulty abank has retaining its business

Consideration of these country-specific differences allows us to calculate correction factors that can be usedto adjust for differences by country. In next year's feedback sessions, benchmarking participants will receivean estimate of what their cost position would be if country differences were taken out of the equation.

I n f l u e n c e o f b u s i n e s s m i x

Some banks within the sample differ significantly in terms of business mix and strategic focus. It is wellknown, for example, that the capital market segment in banks has much more complex requirements thanretail banking. Exhibit 3 illustrates this statement for the benchmarking sample: the higher the share ofretail and private banking, the lower the ITCIR.

The difference is pretty significant: 17 basis points ITCIR per additional percentage point retail/privatebanking business. This proves that it is important to compare banks on a business-segment level. However,please note that the correlation is not very strong (R2 is only 44%). Obviously, other factors come firmlyinto play, such as quality of IT management.

1 6

ITCIR(in %)

0 30 90

Percentage of net income from retail and private banking

2

4

6

8

10

12

14

16

18

20

R = 0.442

40 50 60 70 80


I M P A C T O F B U S I N E S S M I X O N I T C I R

E X H I B I T 3

1 9

In this section we present the main findings of our study. We then use the next section to build on thesefindings to show how banks can use them to improve their IT functions.

F i n d i n g 1 : I T c o s t s c o n t i n u e t o r i s e

Since the early 1970s IT has played an important role for all banks. Today no bank could perform a singlecustomer transaction without massive support from all kinds of IT systems: sales systems to enter transac-tions; product engines designed by product development departments to work through business processes;transaction engines to execute thousands of highly standardized transactions within seconds; bookingengines to post activities to myriad ledgers; controlling systems to monitor banks' businesses; and report-ing systems to meet regulatory requirements. This list is far from complete. From a technical point of view,a bank could be seen as a huge database system with millions of transactions performed with the data.

It is no wonder that banks spend more and more money on IT. From 2001 to 2002, share of IT costsincreased by a full percentage point (see Exhibit 4). Banks anticipate an even stronger increase in 2003:share of IT costs is expected to rise to 20%—that is, every fifth euro would be spent on IT. Next year's reportwill reveal whether the economic environment will have really allowed banks to expand IT expenses to thisextent. Please note that we used a ratio by cost base for showing banks' prognoses for future IT cost struc-tures. This is because cost forecasts are typically more reliable than revenue or business-volume forecasts,as banks spend a lot of time and effort on their budgeting processes and tend to stick to the results.

Increasing IT costs, per se, is not a problem—indeed this is expected as banks automate a growing share of theirprocesses, transforming personnel costs into IT costs. The problem with escalating IT costs arises when the high-er cost is not reflected in a higher return. As we will discuss later, our study shows this is frequently the case: high-er revenues and lower non-IT costs often do not cover higher IT costs (see finding 5).

N I N E F I N D I N G S F R O M T H E B E N C H M A R K I N G

A main reason for this could be that IT is still a big "black box": IT is responsible for the most expensiveprojects in the bank. These projects are almost never completed on time and within budget; and some-times they do not even deliver what they promised at the outset. Even those rare projects that work out asplanned usually significantly increase the ongoing cost of running IT. Trying to understand the reasonsbehind this phenomenon often results in detailed discussions of technical issues that senior managerstend to ignore.

2 0

15%16%

Share of IT costs (IT costs as percentage of operating costs)

20%

2001 2002 2003estimates

21%

6%


S H A R E O F I T C O S T S G R O W I N G

E X H I B I T 4

KEY FIGURES FROM 2003 BENCHMARKING SAMPLEIn our benchmarking study we gathered participants' current cost positions, ratios by volume, revenues,and cost base (see Exhibit 5). Note that all ratios are shown at the total bank level only. This is to helpbanks get a first impression of their individual, overall cost position. In order to identify areas for improve-ment, a more detailed understanding of cost positions within specific business segments is required. Banksfeeling this would help improve their cost positions should consider participating in next year's bench-marking study to receive the more detailed individual feedback from the study.

A typical bank pays some 24 basis points of total assets and spends €40 annually per customer on IT. About10% of a typical bank's total revenues, which means that IT costs account for 16% of all operating expenses—that is, roughly €17,000 per employee—are consumed by IT.

F i n d i n g 2 : U s e r s a t i s f a c t i o n d o e s n o t r i s e w i t h I T s p e n d i n g

Although banks spend a lot on IT, internal customers still are not really satisfied with the services they pro-vide. As part of our research we surveyed internal customer (that is, business) satisfaction with respect tosix different dimensions (see Exhibit 6). None of the banks' customers is "highly satisfied" with its IT serv-ices.

As shown in Exhibit 6, the overall quality of IT services is consistently ranked "average." In particular, cus-tomers named three areas in which IT should improve:

■ Business and IT strategy alignment

■ Responsiveness of first-level user helpdesk

■ Availability of mission-critical applications

2 1

Volume

IT costs

Revenues Cost base

Total assets

24 bps

88

IT costs

Customers

€40/customer

87

IT costs

Revenues

10%

81

IT costs

Operating expenses

16%

84

IT costs

Employees

€17,000/employee

81

-

€17,000/employee

81

Average in sample

R2


K E Y F I G U R E S F R O M 2 0 0 3 B E N C H M A R K I N G S A M P L E ( B A S E D O N 2 0 0 2 A C T U A L S )

E X H I B I T 5

F i n d i n g 3 : L a r g e d i f f e r e n c e s i n l e v e l s o f I T s p e n d i n g e x i s t

To conduct further analysis, we split the banks into two groups: high spenders have higher-than-averageITCIR and low spenders have lower-than-average ITCIR. Highly complex banks should not show large dif-ferences in IT costs per revenue or share of IT costs. What we actually found is that there is a huge gap:high spenders' IT costs per revenues are 60% higher than low spenders', and high spenders' IT costs peroperative expenses are 54% higher.

2 2

1

2

3

4

5

6

Alignment of business and IT strategy

Notsatisfied

HighlysatisfiedAverage

Internal customer satisfaction

Quality of application development

Quality of user helpdesk

Availability of applications

Transparency of cost allocation

Overall quality of IT

Average of participating banks


I N T E R N A L C U S T O M E R S A T I S F A C T I O N

L A R G E D I F F E R E N C E S I N I T C O S T S P E R R E V E N U E S A N D I T C O S T S P E R O P E R A T I N G E X P E N S E S

E X H I B I T 6

ITCIR Share of IT costs

4 6 8 10 12 14 16

(in %)

12 13 14 15 16 17 18 19 20

(in %)

7.5% 13.0%

12.0% 20.0%

Average of full sample High spenders Low spenders

Difference between both groups in %

+60 % +54%

xxSource: BCG's 2003 European IT Benchmarking in Banking Study

E X H I B I T 7

F i n d i n g 4 : T h r e e d i f f e r e n t p a t t e r n s o f I T s p e n d i n g i d e n t i f i e d

Now we examine different patterns of IT investments in relation to business growth. If revenues rise at leastin equal dimension to IT costs, the additional business should generate enough profit to "pay" for the addi-tional IT expenses. This, however, is not the case for most participants.

From a bird's-eye view, banks need to manage their investment cycle effectively. This means a bank needsto operate with stable IT costs for at least three to five years after two to three years of focused IT invest-ments.

Our analysis shows that there are actually three main investment patterns in the European banking indus-try (see Exhibit 8).

All banks within the sample fall into one of the following categories:

■ The first group, aggressive investors, heavily increased spending in IT although revenues could notmatch that increase in the same measure. Such an increase in absolute IT costs by more than 5% usu-ally means an investment in new, major IT systems or a fundamental reengineering of the IT archi-tecture. Surprisingly, these projects are being tackled during difficult economic times. The banks inthis group will have to make sure to keep the remaining investment cycle time short and economicalwith the target of achieving stable IT costs in the mid-term.

■ A second group, moderate investors, shows a fair increase in IT costs combined with stable revenues.This is the biggest group in the sample. This probably reflects the higher inflation rate in the IT indus-

2 3

IT cost growth(in %)

14

12

10

8

6

4

2

0

-2

-4

-6

-8

-10

-12

-14 -14 -10 -6 -2 2 6 10 14-12 -8 -4 0 4 8 12

Aggressive investors(~ 30%)

IT cost increase fasterthan revenues

Revenue increasefaster than IT costs

Moderateinvestors(~ 40%)

Active cost savers(~ 30%)

Revenue growth(in %)


T H R E E I T I N V E S T M E N T P A T T E R N S

E X H I B I T 8

try combined with increasing regulatory requirements such as Basel II and IFRS. Moderate investors donot show a clear investment pattern. They are either in transition between phases in the investmentcycle, or incremental IT cost increases are a long-term pattern. Therefore, these banks must watch therisk of slightly increasing their IT costs every year, slowly building a competitive disadvantage.

■ The third group, active cost savers, managed to decrease IT costs relative to revenues over the lastthree years. At least two banks in this group publicly announced IT cost-cutting initiatives recently.Active cost savers should try to expand the time before their next investment cycles begin withoutneglecting the need to quickly adapt to changing business environments.

F i n d i n g 5 : H i g h e r I T s p e n d i n g d o e s n o t t r a n s l a t e i n t ob e t t e r o p e r a t i o n a l e f f i c i e n c y

Generally, there are two ways IT can create value for business:

■ Increase a bank's efficiency. This means a bank will have below-average operational expenses withoutcutting back on product or process quality.

■ Increase a bank's effectiveness. This means IT functionality will enable unique products or process-es thus providing a competitive advantage in the market.

In finding 3, we divided the benchmarked banks into two categories: high spenders and low spenders(banks with above-average and banks with below-average ITCIR). For this finding as well as the followingfindings we again use this distinction to consider effectiveness and efficiency.

2 4

METHODOLOGYOne of the most difficult topics in IT management is measuring IT's value creation. Many banks justifytheir above-average IT costs with leading-edge functionality, which supposedly provides inimitable com-petitive advantage for the bank; therefore, clear, comprehensible identification of the actual created valueis crucial.

In order to measure and compare effectiveness and efficiency, we require a view on business performance.As was the case for IT costs, we used ratios derived from absolute figures to compare banks. Ratios com-bining volume and revenue figures can work as a first indicator for effectiveness:

■ Total assets per customer, an average of €25,000 in the sample, shows a bank's potential to attractcustomer assets with its product portfolio.

■ Revenues per total assets, an average of 220 basis points in the sample, proves a bank's potential togenerate revenues from the assets it has.

■ Revenues per customer, an average of €580 in the sample, measures a bank's ability to leverage itscustomer base.

As Exhibit 9 shows, high spenders have a higher overall cost-income ratio (61% in contrast to 59% for thewhole sample) and fractionally less revenue per employee (€181,000 compared to €182,000 for the wholesample). Obviously, the increased IT budget does not help decrease the overall operating expenses forthese banks nor does it help to increase revenues.

This result is surprising. One is inclined to expect some back-office productivity improvement throughhigher automation and better IT support. This would enable the bank to run with fewer personnel and,therefore, with lower non-IT operating expenses. The results from the study indicate that this is not thecase. Please note that the general cost-income ratio already includes the IT cost-income ratio because ITcosts are part of overall operating expenses. This strengthens our conclusion that higher IT spending hasno significant impact on non-IT costs.

2 5

These ratios build on the results discussed in the first section of this study (see Exhibit 5). Since they oper-ate on the total bank level, they provide a first impression of a bank's position relative to the study's sam-ple. A more thorough analysis must work on the business-segment level and substitute the volume meas-ures accordingly—for example, assets under management for the private banking segment.

Ratios combining revenues and cost-based figures help estimate a bank's efficiency. Of course, these ratiosare not independent from one another:

■ Operating expenses per revenues—or cost-income ratio, 60% in the sample, is probably the mostgenerally used indicator for efficiency.

■ Operating expenses per total assets, 130 basis points for the sample, is a second indicator of cost-based adequacy, and it is especially well suited for banks with large retail and corporate businesses.

■ Number of customers per employee, 320 customers in the sample, shows the productivity of a bank'sstaff—this ratio can be significantly improved with leading-edge IT support.

■ Revenues per employee, €180,000 in the sample, gives an impression of the automation level a bankhas achieved.

As was noted above, all numbers given are averages of the bank sample and not all data was available fromall participants; therefore, we obtained different sample sizes for these ratios. While it always holds foreach participant that, for example, IT costs per employee divided by IT costs per revenues equals revenuesper employee, this dependency does not hold for the average of the sample.

Looking more closely at back-office employees (see Exhibit 10), one might expect that higher IT investmentslead to fewer back-office personnel (due to higher back-office productivity) and—if back-office personnelcosts are proportional to back-office FTEs—save on back-office personnel costs. But there is no clear patternvisible: only two of the six banks providing the necessary data actually managed to lower costs (total of backoffice and IT), while two are in the quadrant where the total actually increases. At least from a short-term per-spective, we see no proof for improved back-office productivity. To validate this assertion we will have to con-duct this analysis for at least three years, as some delay in effects may come into play. All in all, there is little evi-dence that high spenders currently create value for the bank by increasing efficiency.

2 6

N O E F F I C I E N C Y I M P R O V E M E N T B Y I T V I S I B L E

CIR Revenues per employee (in K )€

56 57 58 59 60 61 62 180 181 182 183 184

57%

18161%

183


(in %) (in K )€


E X H I B I T 9

G R O W T H O F I T C O S T S A N D B A C K - O F F I C E P E R S O N N E L

-30 30 25 20 15 10 5 0 -5 -10 -15 -20 -25 -30

-20

-10

0

10

20

30Growth of ITcosts per totalassets (in %)

Growth of IT costs per total assets vs. growth of back-office personnel (2001–2003e)

Total of IT andback-office personnel

costs per unit decreases

Growth of back-officepersonnel costs (in %)(1)

(2)

(2)

(1) Growth of back-office personnel costs seen as proportional to growth in number of back-office FTEs(2) Growth 2001–2002Source: BCG's 2003 European IT Benchmarking in Banking Study

Growing IT costs leadsto shrinking back-office

personnel

Total of IT andback-office personnelcost increases

Shrinking IT costs paidfor with growing back-office personnel

E X H I B I T 1 0

F i n d i n g 6 : H i g h e r I T s p e n d i n g d o e s n o t l e a d t o m o r e e f f e c t i v e n e s s

Keeping this surprising result in mind, we now consider the other side of the coin: do higher IT costsincrease business effectiveness? First of all, the exact meaning of effectiveness has to be clarified. It is muchmore difficult to measure effectiveness—that is, a bank's market success through superior products orprocesses— than efficiency. Strong competitive advantages should help banks generate higher revenuesthan a similar bank of equal size. Furthermore, those banks should also be able to attract more assets fromtheir customers than their peer group. Therefore, the three business ratios defined in the methodologyinsert in finding 5 (total assets per customer, revenues per total assets, and revenues per customer) couldwork as a first indicator for comparing banks' effectiveness.

It is important to note that IT support is not the only factor related to effectiveness. The benchmarkingsample for this study is not big enough to ensure that all business-driven factors are leveled out. A muchbetter correlation can be achieved when analyzing effectiveness on the business segment level—retailbanking, for example—than using total assets as the volume driver for the overall business.

Taking these issues into account, the following conclusions are still well-grounded enough to allow forsome interesting insights: Exhibit 11 shows that high spenders generate in fact lower revenues per assets.They could actually produce revenues of only 190 basis points of assets compared to 260 basis points in theother group. This is certainly not a result likely to legitimize the request for more "intensive" IT support.Quite the contrary, there is no evidence here that high spenders can transform those expenses into com-petitive advantage that would allow them to outperform their peer group.

2 7

N O E V I D E N C E O F H I G H E R E F F E C T I V E N E S S

bps

bps

Revenues/asset

Costs/asset

Profit/asset

High spenders

CIR:61%

Revenues/asset

Costs/asset

Profit/asset

Low spenders

CIR:57%

190 -110

80

260 -150

110


E X H I B I T 1 1

One can also see that high spenders keep their operating costs per asset lower than low spenders. But dueto lower revenues, the resulting profit is still only 80 basis points of assets in contrast to 110 basis points inthe peer group. This harkens back to the result already discussed in the previous section: high spendersalso have higher cost-income ratios. Please note that our grouping does not reflect business mix or region-al patterns.

F i n d i n g 7 : A d o p t i n g s t a n d a r d a p p l i c a t i o n s d o e s n o t ( y e t )g e n e r a t e c o s t a d v a n t a g e s

To get a better understanding of why some banks' IT functions are much more expensive without provid-ing additional business value, one needs to dig deeper into the details of IT delivery. IT can generally bedivided into two areas:

■ Production: setting up and maintaining the technical environment to run an IT platform

■ Application development: developing, customizing, and maintaining an IT platform to supportbanking processes

Touching on the distinction between high spenders and low spenders used before, Exhibit 12 breaks downIT costs into two categories: application development costs per revenues and IT production costs per rev-enues. If we look at high spenders, we discover that application development increases to 5% of revenues(up 25%) and production to 8% (up 43%) of revenues. What could lie behind this phenomenon?

2 8

I T C O S T B R O K E N D O W N T O A P P L I C A T I O N D E V E L O P M E N T A N D P R O D U C T I O N

Application development costsper revenues (in %)

IT production costs perrevenues (in %)

2.4 2.6 3.6 4.6 5.6 6.6 7.6 8.62.8 3.2

3% 4%

5% 8%

3.6 4.0 4.4


4.8 5.2 5.6

(in %) (in %)


E X H I B I T 1 2

Looking at IT architecture, one might expect banks using legacy systems to have higher costs. But, in fact,as the benchmarking data indicates, in the near term it is mostly cheaper to use old monolithic softwarebecause there are no migration costs and the systems are completely written off in the accounting ledgers.We will later discuss in finding 8 what other considerations should be taken into account here.

If we compare the two groups of banks as introduced in finding 3, we see that high spenders usually usemore standard software (see Exhibit 13).

2 9

B A N K S W I T H L O W E R I T C O S T R A T I O S T E N D T O U S E L E S S S T A N D A R D S O F T W A R E

Mostly standard

High spenders Low spenders

Sales and distribution systems

Mixed

Software type used

Mostly custom developments

Product-processing systems

Basic and support systems


E X H I B I T 1 3

C U S T O M - D E V E L O P E D S O F T W A R E T E N D S T O B E L E S S E X P E N S I V E

Mostly packagedsoftware

Mixed (packaged, project solutions,custom developments)

Mostly customdevelopments

No participant inthis category

Share of IT costs2002 (in %)

Deviations

Average

25

20

15

10

5

0


E X H I B I T 1 4

Exhibit 14 shows that this structural difference can be seen in the significantly lower share of IT costs forparticipants using mostly custom-developed software in contrast to banks with a mixed structure. Pleasenote that in all banks most systems are still custom developed.

However, just taking an accounting view of current cost levels is not enough. In discussion with the bench-marking participants, we saw that standard software systems were often installed quite recently. But the mul-titude of support and auxiliary systems added to monolithic core systems over time had created complexinterfaces, a difficult-to-maintain job network, and manual-update processes. The resulting "applicationjungle" becomes more and more difficult to maintain every year and increases the bank's operational risk.A typical bank's architecture—significantly simplified to ensure confidentiality—is shown in Exhibit 15.

3 0

S A M P L E I T A R C H I T E C T U R E

Wall Street Front office Domestic custodian Other custodiansinternational Reconciliation

Bloomberg

Legacy system C

Stocks

Ordermanagement

Depotreconciliation Reconciliation

Futures & options

Securities lending

Legacy system B Legacy system D

Currency Smartstreamreconciliation

Other custodiansinternational CARS

Xetra

RTS

Brokers

Derivatives

Legacy system A

Securities office

Offerten

Swift Auto Manual Ledger postingsSource: BCG's 2003 European IT Benchmarking in Banking Study

E X H I B I T 1 5

The need for standardization, the countless interfaces, an increasing deficiency in relevant programmingskills and technical know-how, as well as a lack of (comprehensive) documentation all hamper quick reac-tion to new business requirements or regulations.

Banking CIOs believe that the change to new, modular—often packaged—software is expensive but stilloffers sizable advantages in the long run. It helps lower maintenance costs significantly, provide higherflexibility, and—once in a stable state—lower risks for making changes with respect to feasibility, time, andcost, as well as data security and vulnerability.

Furthermore, the vast majority of high spenders' projects fall into the categories, "production, internalprocesses" and "infrastructure" (see Exhibit 17). These project categories have the highest share of main-tenance projects. In contrast, low spenders use 43% of their project budget in the area of "customer, sales,and marketing." This is 60% more than the 27% used by high spenders. This is a real competitive advan-tage if one considers that projects in this category have a fair chance of having a direct impact on the busi-ness by improving efficiency and effectiveness.

F i n d i n g 8 : M a i n t e n a n c e c o s t s t e n d t o i n c r e a s e w i t hh i g h e r I T s p e n d i n g

If we assume that high spenders run relatively new systems, we would expect this group of banks to have alower share of maintenance costs. In reality, maintenance for high spenders uses up about one-third of thetotal application development budget compared to only about one-quarter for low spenders (see Exhibit16). We also know that the total application development costs of high spenders are higher on average.This could mean that banks that have updated their systems have not subsequently managed to lower theirmaintenance costs. This can partly be attributed to the fact that, during the transition phase, banks oftenhave to run several operational systems in parallel. We should see significant improvements here in thefuture.

3 1

A P P L I C A T I O N D E V E L O P M E N T C O S T S B Y P R O J E C T T Y P E

Must Should/could Maintenance

17%

50%

33%

27%

49%

24%

60%

50

40

30

20

10

0

High spenders

Low spenders


E X H I B I T 1 6

F i n d i n g 9 : S c a l e i s n o t b e i n g t r a n s l a t e d i n t o c o s t a d v a n t a g e s

The biggest bank in the benchmarking sample has about three times the assets of the smallest bank. Butdo bigger banks benefit from their size? As Exhibit 18 shows, this is not the case. Actually, there is no cor-relation at all between assets and IT costs per assets (R2 = 1%). Apparently the bigger players in the sampledid not realize any benefit from their size in terms of lower IT costs. This is quite surprising since our expe-rience shows that there should be scale benefits at least in IT operations and desktop services. We also seeindications for this in the group of participating banks, although the sample is too small to derive a gener-alized scale factor. It seems then that larger banks have a relative cost advantage in some cost categories butspend more in others, especially in application development.

3 2

N O E C O N O M I E S O F S C A L E V I S I B L E

ITcosts/totalassets2002(in %)

0

0.1

0.2

0.3

0.4

Total assets (indexed)

0 10 20 30 40 50 60 70 80 90 100


E X H I B I T 1 8

A P P L I C A T I O N D E V E L O P M E N T C O S T S B Y I N V E S T M E N T A R E A

Production,internal processes

51%

38%

Infra-structure

18%

7%

Customer,sales, marketing

27%

43%

Control

5%

12%

60%

50

40

30

20

10

0

High spenders

Low spenders


E X H I B I T 1 7

3 5

T H R E E L E V E R S T O I M P R O V E I T P E R F O R M A N C E

Taking all nine findings into account, it becomes clear that most of the alleged drivers for high IT costsseem to be rooted in structural difficulties that cannot be changed on short notice: IT architecture, sourc-ing models, project portfolios, and staffing mix. Only a solution to these structural challenges allows a bet-ter IT cost base and thus provides competitive advantage to the bank. Since these issues can only beaddressed on a long-term basis, the advantage is sustainable and apparently difficult to imitate.

Strategic internal management of IT is the key issue for controlling IT costs, not day-to-day managementrequired to run the IT department. Strategic IT management needs to focus on redesigning IT so that thestructural environment best supports an efficient IT service delivery. IT has an opportunity to provide realcompetitive advantage to the bank, and BCG has identified three critical levers for improving the efficiencyand effectiveness of IT in banking (see Exhibit 19). These levers provide the opportunity for positive, short-term impact on the IT cost base as well as sustainable, mid-term increases in value creation through IT.

T H R E E L E V E R S F O R I T I M P R O V E M E N T

High Tighten ITperformance and

value management

Improve business andIT strategy alignment

Low

< 1 year > 3 years

“Must do”

“Should do”

“Can do”

Impact on ITefficiency andeffectiveness

Time forimplementation

1

2

3

Increase level ofstandardization


E X H I B I T 1 9

L e v e r 1 : I m p r o v e b u s i n e s s a n d I T s t r a t e g y a l i g n m e n t !

As discussed before, high spenders are not necessarily more efficient (finding 5) nor are they more effec-tive (finding 6): there appears to be a mismatch between business strategy and IT strategy. Indeed, in thecourse of our research and discussions with participants, we frequently encountered the perception thatIT and business speak different languages—IT often claims that business does not respond to its needs ina precise, practical way; conversely, business is not satisfied with the service delivered by its IT function.

A key lever for improving business and IT strategy alignment at the business-unit and corporate levels is theintroduction of well-defined alignment processes and joint committees with members from corporate anddecentralized IT as well as the business side.

IT project-portfolio management is a key factor in aligning business and IT strategy. Many banks have business-unit-specific IT project portfolios that are more or less aligned with the business portfolio of that particularbusiness line. But even those portfolios are often not managed using value-based metrics; the business case foran IT project should be based on exactly the same measures as business projects. IT must be explicit about thebenefits for the business lines; for example, reduced recharges achieved through a consolidation project.

Banks should enforce business and IT strategy alignment by maintaining a central, fully integrated projectportfolio that monitors progress and budget as well as business objectives, functionalities, and prioritiza-tion. The portfolio should be updated regularly and all projects reviewed after each phase. IT projectsshould always be linked to a specific business objective and based on business cases that apply the samemetrics as business projects. This ensures that IT projects focus on economic profit rather than on imple-menting technology for its own sake.

In addition, individual business units should have to compete for their share of the IT project budget on acorporate level, thus ensuring that overall corporate objectives and strategies are met and helping enforcestandards across the organization.

In our experience, significant improvements to business and IT strategy alignment can be realized withinone year. The impact on IT efficiency and effectiveness can be significant.

L e v e r 2 : T i g h t e n I T p e r f o r m a n c e a n d v a l u e m a n a g e m e n t !

During the course of the study and in many client projects, we have found that IT costs are far from trans-parent. Banks in general have struggled to decipher a clear view of their own IT cost baseline, particularlyin relation to value creation (see findings 5 and 6).

3 6

Business cases for IT projects contain cost-benefit analyses, but our research has shown that the delivery ofbenefits and the value they create are not tracked (for example, cost savings are often not built into thebudget once the project has been delivered). But without transparency, IT functions cannot measure theirperformance. They cannot sensibly prioritize or suspend projects, negotiate with vendors, or apply mean-ingful recharges to their business partners.

Value and performance management, however, is more than just cost transparency and benefit tracking.Banks should develop an IT dashboard that not only measures how efficiently IT provides its services, butalso how well IT helps fulfill strategic business objectives and support organizational effectiveness.

In our view, there are four basic steps to improving IT performance and value management (see Exhibit 20).After business and IT goals have been aligned (lever 1), an information model must be developed for keydecision makers. Performance metrics must then be developed and linked to staff compensation levels.

Rigorous but pragmatic performance and value management are vital to ensuring that the IT function isrun efficiently and the bank is in control of its IT investment cycle (see finding 4). Best-in-class banks usea well-defined set of simple metrics that allow them to measure the performance of each individual IT serv-ice provided. These metrics do not just focus on ratios discussed in this report (for example, IT costs perrevenues) but also address unit costs (for example, IT costs per current account or per transaction). Othermetrics are based on an IT-balanced scorecard approach that measures more qualitative aspects such ascustomer satisfaction (see finding 2) and the ability of the IT organization to "learn" and improve.

3 7

F O U R S T E P S T O I T V A L U E A N D P E R F O R M A N C E M A N A G E M E N T

To implement a meaningful system of metrics, adetailed business strategy road map for IT needsto be defined

On the highest level, IT performance should bemeasured against achievement of business-defined goals

Different management levels act on differentdecision levels—from top strategic decisions todaily operational ones

A specific information model is needed for eachparticular decision-making tier

Implementing a “change culture” within the ITorganization is essential to improve performance

This is best achieved by linking individualcompensation to the metrics used for therespective tiers

Business strategy and IT goals need to betranslated into a comprehensive view ofoperational and strategic measures

Measures: strategic needs of enterprise, needs ofindividual customer, internal IT businessperformance, ongoing IT innovation and learning

1Alignbusiness

and ITgoals

Linkmetrics to

compensation

Developperformance

metrics

Developinformation

model fordecision

making

2

3

4

Buildbaseline with

data-collection andanalysis

capability


E X H I B I T 2 0

The move towards shared IT services models means that business units do not directly control their ITbudgets; costs are instead recharged from the center. In general, banks currently do this via a combinationof allocation percentages and defined unit costs. Most banks told us that they are planning to move to a"utility" or "on-demand" type of charging mechanism (for example, per transaction costs) for the longterm. We believe this would vastly increase business transparency of IT costs and provide a better under-standing of the real IT cost drivers. These measures are, however, far from being implemented very soon.

L e v e r 3 : I n c r e a s e l e v e l o f s t a n d a r d i z a t i o n !

A common theme among our discussions with banks is the fact that increasing complexity drives IT costs.The complexity of banks' IT landscapes has grown significantly over recent years, the result of heavyinvestment in application development as well as merger and acquisition activity in the European bankingsector.

Maximizing standardization is an absolute "must" for banks to reduce IT costs. As a result of increased costpressures, many banks have already identified opportunities to create value by standardizing their IT archi-tecture on business-unit and corporate levels. However, our research reveals that fewer than half of theseopportunities have been implemented successfully. Despite the fact that many standardization and con-solidation projects began two or more years ago, very few banks have thus far managed to reduce their ITcosts.

3 8

A F O U R P O I N T S T A N D A R D I Z A T I O N M A P T O R E D U C E C O M P L E X I T Y

High

Increase level of ITarchitecture

standardization

Increase powerof central CIOorganization

Low

< 1 year > 3 years

“Must do”

“Should do”

“Can do”

Impact on ITefficiency andeffectiveness

Time forimplementation

B

A C

Implementstandard desktop

services for allemployees

D

Consider selectiveoffshoring and

outsourcing


E X H I B I T 2 1

We believe that banks have not exploited the opportunities afforded them from economies of scale (asfinding 9 supports) or from standardization. New applications and functional developments have over-shadowed efficiency gained from reduction in complexity. Thus we conclude that there are still significantopportunities for creating value by standardizing IT infrastructure, desktop applications, product pro-cessing, and support systems.

IT functions should therefore seek to increase the level of standardization across their organizations. BCGhas identified a four-point "standardization map" to assist CIOs in approaching this challenge:

I n c r e a s e p o w e r o f c e n t r a l C I O o r g a n i z a t i o n !

In most industries, there is already a strong push toward centralization and standardization of the IT func-tion. This trend is weaker in the banking industry. Moreover, even though some banks have started mov-ing in this direction, the role and empowerment of the CIO tends to vary widely.

A key learning from BCG's Global IT Management Benchmarking2, a wide study of IT functions acrossmany industries, was that in order to drive standardization, companies must first increase the power andmandate of the CIO. In practice, this means the CIO governance model should be top-down rather than"democratic." Business-unit IT functions must agree to accept standards as defined on a corporate level. Itis also crucial that controlling processes are in place. "Degree of standardization" is a key metric for IT per-formance and value management (see lever 2).

C o n s i d e r s e l e c t i v e o f f s h o r i n g a n d o u t s o u r c i n g !

Our research shows no clear correlation between the overall sourcing strategy (in-house vs. outsourcing)and IT efficiency:

■ Many banks in the sample just finalized outsourcing deals or are assessing the option to outsource.

■ Some banks have set up pilots in noncritical areas. The same is true for offshoring: shifting capacityto low-cost countries, either by founding subsidiaries or working with a third-party provider.

■ Other banks rely solely on in-house services yet run their IT function very efficiently.

Most of the banks that recently introduced outsourcing seem to have slightly increased their IT costs. How-ever, the data is not extensive enough to draw definitive conclusions on this issue. In 2004 BCG will dedi-cate a special section of the benchmarking study to outsourcing and offshoring so as to clearly highlighttheir economic value as well as typical problems, and pitfalls.

Drawing on our experience from client projects in various industries, we believe there is already ampleopportunity to reduce costs by outsourcing specific functions. If banks are running subscale IT operations,

3 9

2 See glossary.

outsourcing or joint ventures may be the only possibility for reaching a competitive cost position. Out-sourcing, however, is not a silver bullet: banks must understand their current cost base in detail as it com-pares to the market in order to maximize the value created by the deal.

Whether IT operations are in-house or outsourced, offshoring will remain a major area of focus for IT.Most UK banks and some European banks are already planning to extend their offshoring activities. Whileit may provide a chance to reduce IT costs (typically in the order of 30% to 50%), offshoring also presentssizeable risks. Cultural differences and resulting inefficiencies may outweigh potential cost savings if thetransition is not managed properly.

Banks therefore should have a clear vision of their business and IT strategies before introducing out-sourcing or offshoring. A well-developed and tested sourcing strategy is key to generating value.

This also holds true for determining to what extent external staff should be hired for application develop-ment. External employees are clearly more expensive on an hourly basis. On the other hand, they are high-ly focused on a specific project and are not pulled away by nonproductive tasks. Use of external employeesprovides additional flexibility in that the bank only needs to pay for special knowledge when required, andit can ramp down the resources after a project is complete.

Surprisingly, the benchmarking data indicates that low spenders use almost twice as much externalresource capacity—67% of full-time equivalents (FTEs) in contrast to 38% for their peer group (see Exhib-it 22). It appears that using external resources pays off, although they come with a higher price per hour.This is contrary to our consulting experience in most projects. One explanation may be that there is aproblem managing internal resources efficiently. To validate this hypothesis, it will be necessary to com-pare banks with a high level of external resources to banks with a high degree of internal resources andapplication-development efficiency.

4 0

B A N K S W I T H H I G H E R I T C O S T R A T I O S T E N D T O U S E M O R E I N T E R N A L S T A F F

High spenders Low spenders

62%Percentage ofall staff

33%

Internal

38%

67%

External


E X H I B I T 2 2

I n c r e a s e l e v e l o f I T a r c h i t e c t u r e s t a n d a r d i z a t i o n !

As discussed in finding 7 (advantages of standard software not yet exploited), IT architecture complexityis vital to improving IT efficiency. Compared to other industries, banks still have very complex applicationlandscapes and platforms. Most banks have invested heavily in business and support applications overrecent years. IT functions are now faced with an increasingly difficult task in managing IT architecture.

It appears that complexity drives application maintenance costs. Banks that manage to reduce complexitywill have lower IT cost ratios. Consolidation and standardization should therefore be a key strategic objec-tive for banks in the coming years.

I m p l e m e n t s t a n d a r d d e s k t o p s e r v i c e s f o r a l l e m p l o y e e s !

Over the last two years, several banks have outsourced all of their IT desktop services, including first-level user support. However, we observed significant variations in IT desktop costs per employee amongparticipants.

The key drivers for IT desktop costs include the number of platforms and supported applications. Best-in-class organizations have standardized their desktops group wide and have reduced the number of sup-ported applications. Banks with a high cost base often have many different standards across their businessunits, independent of whether they have outsourced desktop services or not. The key issue when consid-ering outsourcing is to understand the current cost base compared to the market benchmarks; negotiat-ing a deal based on the current cost base may result in reduced overall cost, but compared to competition,it may still be way above market average.

4 1

4 3

G L O S S A R Y

A S S E T M A N A G E M E N TManagement of third-party assets, including management of mutual funds and institutional mandates (inGermany: special funds).

A S S E T S U N D E R M A N A G E M E N TAssets administered by a financial institution that are beneficially owned by clients and are therefore notreported on the balance sheet of the financial institution. Services provided are at least of an administra-tive nature, such as safekeeping securities, collecting investment income, and settling purchase and saletransactions, but they may also comprise portfolio management on behalf of the client.

A P P L I C AT I O N D E V E L O P M E N TAll costs for software development (generation of preliminary studies and specialized concepts, as well asrealization and application testing), advancement, maintenance, and development tools.

B U S I N E S S V O L U M EMeasure of the size of a bank's business. Volume measures are segment specific and reflect key businessdrivers (for example, customers in retail banking).

C A P I TA L M A R K E T SBusiness segment responsible for trading and sales, mergers and acquisitions, and new issues/primary markets.

C O S T- I N C O M E R AT I O ( C I R )Operating expenses per total revenues (see "key financial terms" below).

C O R P O R AT E B A N K I N GBusiness segment responsible for enterprises with more than €5 million to €10 million revenue per year,multinationals, institutions, and governments. Note: the range is used to allow as many participants as posi-ble to use their individual definitions.

C U S T O M D E V E L O P M E N TProprietary application; bank is responsible for maintenance.

D E S K T O P S E R V I C E SShort for desktop services/local support: includes PC hardware/software, LAN, desktop management,first-level user support, self-service terminal operations, support functions comprising specific overheadand support functions such as management, controlling, purchasing, security services, and ATM.

G L O B A L I T M A N A G E M E N T B E N C H M A R K I N G ( G I T M A B )BCG benchmark study for globally operating companies with multiple business units and a mix of centraland decentralized IT organizations. The goal of this study was to generate transparency on current andfuture directions in IT management.

H I G H S P E N D E R SGroup of participating banks with above-average ITCIR.

I N T E R N A L C U S T O M E RBusiness-lines and business-line employees within the bank using the services of the IT function.

4 4

I T A R C H I T E C T U R EAn enterprise-wide blueprint for IT (time horizon up to five years). It reflects the business model and isdriven by business requirements. The IT architecture serves as a framework for making technology deci-sions and typically consists of four components: business architecture (or functional architecture), appli-cation architecture, information architecture, and technical architecture.

I T C O S T SDirect and indirect IT cost accruing in central IT, business segments, and support functions, including per-sonnel costs and costs for external support; material costs comprising hardware, software, network, andother costs (voice communication and postage are not included); other expenses include allocations andmarkup.

I T C O S T- I N C O M E R AT I O ( I T C I R )IT cost per total revenues (see "key financial terms" below).

I T E F F E C T I V E N E S SAbility of the IT function to enable the bank to provide unique products or processes resulting in a com-petitive advantage in the market.

I T E F F I C I E N C YAbility of the IT function to enable the bank to have below-average operational expenses without cuttingback on product or process quality.

I T I N F R A S T R U C T U R ETechnical platforms and components of IT (for example, WAN, servers, mail systems, operating systems).

I T S T R AT E G YA prospective, business-driven view of information needs and systems/technology requirements, together witha coordinated, integrated, company-wide strategy and plan for meeting them in a way that maximizes value.

K E Y F I N A N C I A L T E R M SWithin this study, we do not differentiate between costs and expenses or between income and revenues.Please refer to Exhibit 23 for a detailed definition of all financial terms used.

L O W S P E N D E R SGroup of participating banks with below-average ITCIR.

K E Y F I N A N C I A L T E R M S U S E D I N T H I S S T U D Y ( E X A M P L E )

Interest receivableInterest payableNet interest income

Fees and commission receivableFees and commission payableNet fees and commission income

Dealing profits before expensesOther operating incomeTotal revenues

Administrative expensesDepreciation and amortizationOther expensesProvisions for loan lossesOperating expenses

Profit

100-70

50-10

30

40

4010

-40-30

-5-20

120

-95

25


E X H I B I T 2 3

4 5

M A I N T E N A N C EAll efforts ensuring ongoing operations.

" M U S T " P R O J E C T SProjects prescribed by law or regulations or necessary for ensuring continuous business operations.

O F F S H O R I N GIT services provided by personnel abroad, typically in a low-cost country.

O P E R AT I N G E X P E N S E SAdministrative and other expenses (see "key financial terms" above).

O U T S O U R C I N GIT services provided by an external company; can refer to projects or services (for example, maintenanceof a given application, operation of first-level user support in one country) or complete outsourcing offully functional areas (for example, whole IT production and/or operations).

P R I V AT E B A N K I N GBusiness segment responsible for private wealth management for wealthy individuals and families withmore than €250,000 in private assets.

P R O D U C T I O N / O P E R AT I O N SIT production costs including cost of data center operations and WAN, excluding voice communicationand postage.

R E TA I L B A N K I N GBusiness segment responsible for mass retail and mass-affluent customers with up to €250,000 of privateassets, and small-to-medium enterprises (SME) with less than €5 million to €10 million annual revenue.

R E V E N U E SNet income before operating expenses; equals sum of net interest income, net fees, commission income,dealing profits before expenses, and other operating income (see "key financial terms" above).

S H A R E O F I T C O S T SIT cost as a percentage of operating expenses.

" S H O U L D / C O U L D " P R O J E C T SProjects essential for strategy, efficiency, and controlling or for other related factors.

S TA N D A R D S O F T W A R EAlso: off-the-shelf software; application program developed by software provider for sale to third parties,although the application may be customized to a user's requirements, the lion's share of maintenance isstill done by the software provider.

T O TA L A S S E T SSum of all current and long-term assets (for example, tangible and intangible assets, receivables, debt secu-rities and other fixed-interest securities, shares and other variable-yield securities, investments, and sharesin affiliated companies).

T R A N S A C T I O N B A N K I N GBusiness segment responsible for payments, custody business, securities transactions, and securities-deposit processing. Only considered a separate business segment if the bank offers those services to thirdparties.

U S E R H E L P D E S K ( U H D )Support unit for internal customers of IT, typically structured in first-level helpdesk (common user prob-lems, solution requiring no specialist knowledge) and second-level helpdesk (problems and questionswhich require deeper knowledge). UHDs are often organized into call centers (see also: desktop servic-es/local support).

4 7

T H E A U T H O R S

Dr. Rainer Minz is a senior vice president in BCG's Cologne office and leader of the worldwide IT practice area.

Mr. Heinz Möllenkamp is a vice president in BCG's Cologne office.

Mr. Ralf Dreischmeier is a vice president in BCG's London office.

Mr. Frank Felden is a manager in BCG's Cologne office.

All authors are experts on IT in financial services. They would like to thank the participating banks forsharing their data with BCG and for providing valuable insights.

We owe a tremendous thank you to our internal team—Mr. Juan Jose Alonso, Mr. Emmanuel Baviere, Dr.Astrid Blumstengel, Mr. Christophe Duthoit, Dr. Henning Krüger, Mr. Pedro Soria, Ms. Annette Wolter,and Mr. Tosja Zywietz—who made this study possible. For their feedback on this report and providing addi-tional information, we thank Mr. Martin Tschopp (BCG Zurich) and Dr. Michael Grebe (BCG Munich).

BBCCGG GGeerrmmaannyy

The Bos ton Consu l t ing GroupDi rcksens t raße 4110178 Ber l inGermany

Te l : +49 (0)30 2887-10

The Bos ton Consu l t ing GroupKö lnTurmIm Mediapark 850670 Co logneGermany

Te l : +49 (0)221 5500-50

The Bos ton Consu l t ing Group Ludwigs t raße 2180539 MunichGermany

Te l : +49 (0)89 2317-40

The Bos ton Consu l t ing GroupAn der Wel le 360322 Frank fu r tGermany

Te l : +49 (0)69 9150-20

BBCCGG AAuuss tt rr ii aa

The Bos ton Consu l t ing GroupAm Hof 81010 ViennaAust r ia

Te l : +43 (0)1 53756-80

The Bos ton Consu l t ing GroupStadt to r 140219 Düsse ldor fGermany

Te l : +49 (0)211 3011-30

The Bos ton Consu l t ing GroupChi lehaus AFischer tw ie te 220095 HamburgGermany

Te l : +49 (0)40 3099-60

The Bos ton Consu l t ing GroupKronpr inzs t raße 2870173 Stu t tga r tGermany

Te l : +49 (0)711 2020-70

CCoonnttaaccttss

For fu r the r in fo rmat ion on th i s s tudy,p lease contac t :

DDrr.. RRaaiinneerr MMiinnzz: [email protected] MMöölllleennkkaammpp: [email protected] DDrreeiisscchhmmeeiieerr: [email protected] FFeellddeenn: [email protected]

AmsterdamAthensAtlantaAucklandBangkokBarcelonaBeijingBerlinBostonBrusselsBudapestBuenos AiresChicagoCologneCopenhagenDallasDüsseldorfFrankfurtHamburgHelsinki

Hong KongHoustonIstanbulJakartaKuala LumpurLisbonLondonLos AngelesMadridMelbourneMexico CityMiamiMilanMonterreyMoscowMumbaiMunichNagoyaNew DelhiNew York

OsloParisPragueRomeSan FranciscoSantiagoSão PauloSeoulShanghaiSingaporeStockholmStuttgartSydneyTaipeiTokyoTorontoViennaWarsawWashingtonZurich

BCG The Boston Consulting Group

www.bcg.com

it costs in banks: revisit your beliefs! · tion. with our it benchmarking report we intend to...

Documents