To Unsubscribe:If you would no longer like to receive this newsletter, please click here
Issue: 002  |  February 2007

NEWS & EVENTS

PAPERS


The business case for metrics

 

The software industry's failure to adopt software metrics is a huge unrecognised cost. Using a metrics program to help achieve targetted improvements in productivity and estimating of software development and maintenance could deliver benefits to the UK software industry in the region of £600M per annum.

Exploiting metrics is also shown to be the key to unlocking many £ Billions in annual benefits if software productivity improvement activities were pursued successfully across the industry. Charles Symons, one of the COSMIC functional size measurement development team, has produced a formula calculating the economic benefits that could be obtained by the judicious use of metrics in relevant parts of the UK software industry. Setting reasonable targets for improvement and using conservative estimates, the figures derived nevertheless give anyone responsible for IT governance food for thought.

 

Outsourcing: Output-based contracts

 

Many output-based contracts incorporate agreement that the unit prices will progressively reduce over time. But low cost is no good if the supplier fails to fulfill the business need of their customer.

Setting contracts which effectively incentivise the behaviours necessary for continuous improvement is a little-understood and vital part of improving software performance.

Grant Rule of SMS draws on recent case history to offer guidance on setting and managing effective output based contracts for the outsourced supply of software services.

 

Changing people

 

Jill Pritchet writes for the BCS@50 on how to engage middle management with process improvement initiatives.

For any change programme to succeed it is absolutely essential that Middle Managers support the changes and are actively engaged in the project. When change is actively managed in a positive environment, then everyone is encouraged and motivated to embrace change and contribute ideas.

 

Delivering Business Value from Information Systems

The role of senior management in the IT department today is less about the technology, and more about the role information systems play in the overall business strategy.

Maximising gains through software process improvement.

Virtually all business systems are now software-intensive, and the IT department is subject to the same management constraints of predictable scheduling, cost control, and value delivery as the rest of the organisation.

CIO Symposium

CAI and the IT Metrics and Productivity Institute present the 2007 CIO Symposium on Software Best Practices. It will be held on Friday 23rd February at the Savoy, London.

...more

 

This was one of the themes raised at the recent Intellect Private Sector presentation dinner looking at the future of the UK software industry. Software developers have tended to see themselves as specialists and innovators, operating outside normal departmental accountability. Those in other roles – such as CFO - have not always had the expertise to know how software development and maintenance can be made accountable.

The result can be a mismatch between software development processes and business need on the one hand, and pressure to reduce costs without regard to the impact on value delivered and the whole life-cost on the other. If there are no comparisons with the performance of other, similar teams, how can the IT department or supplier demonstrate that they are delivering value for money? How can a CIO tell that an outsourced supplier has improved on the service previously delivered by an in-house team? How can the claims of an outsourced supplier be evaluated? How do you set effective output-based contracts?

While it is not appropriate to regard innovation as a repeatable process, much of the work of software development and maintenance does not fall into this category. Models for improving the software process such as the SEI’s CMMI are now widely used, particularly by offshore suppliers, as an assurance of quality. ISO, ITIL, Six Sigma and other industry-specific models provide a similar function.

The map, however, is not the territory. A tick-box approach to software process improvement – regardless of the model used - will more often than not fail to realise the proven benefits which such models have been shown to deliver. To achieve a real improvement in process performance, there is a need for an experienced assessment of the starting point and a clear alignment of improvement activity and business goals. An effective metrics programme is necessary to manage the change programme, and to demonstrate the improvement delivered in terms that relate to the business need.

Best Practice Audit
How do your software development and maintenance processes compare to best practice? Do they deliver what you need? Where should you focus improvement activity?

...more


 

Transformation, Business excellence and Leadership: SMS welcomes a new associate

SMS has consistently taken a holistic approach to software process improvement, building long-term partnerships with our clients to support a programme of continuing improvement. We are pleased to be welcoming new associates, Keith Jackson and his team at TBL Ltd. whose skills and experience enhance our ability to partner large organisations to achieve rapid, measurable, long-term change and improvement to the process of software delivery.

Improvement programmes frequently fail to deliver value over their required timescale. This can be because they take too long; or the value of the improvements is not proven; or people are not sure why the programme is necessary – but the most common cause is a lack of sustained executive commitment. Many improvement programmes lose momentum and management confidence in the first three months. This is often due to an extensive upfront planning phase that suffers from a lack of validated estimates, resulting in broken milestones.

Effective, integrated metrics, and improvements in requirements capture and estimating are intrinsic to the success of a sustainable improvement programme. The combined experience and expertise of SMS/TBL can successfully address all these issues and deliver the results an organisation needs from its software improvement initiative.

SMS/TBL services are delivered over a fixed agreed timescale at an agreed fixed cost.

Introducing TBL

TBL stands for Transformation, Business excellence and Leadership. Founded in 1989, TBL’s expertise focuses on establishing and maintaining executive commitment to performance improvement initiatives. TBL engages with its Client management and staff to deliver and implement solutions that are specifically tailored to the clients business.

TBL’s managing consultants each have at least 25 years knowledge and experience of board level management practices, business strategy, middle management best practices, operations, and IT. They have all held senior management positions as CIO’s and directors of technology, capital markets, marketing and sales and have significant and successful hands-on experience of improving organizational capabilities and business performance in a wide range of industries including Banking and Financial Services, IT, Pharmaceuticals, Telecommunications and Energy. Their client list includes Lloyds TSB, Legal &General, Gartmore Fund Managers, JP Morgan, Dutch DSS, ABN-AMRO, Philips, Johnson and Johnson, Bath & North East Somerset Council.

The most important challenges in the management of information systems today are the ability to respond quickly and flexibly to changes in customer requirements; and the ability to add value by delivering services that improve business performance in areas that achieve corporate goals. TBL and SMS have a range of products and services designed to support CIOs seeking to address these issues.

Can we help you?
To find out how SMS/TBL expertise can support your software management strategy, contact sales@measuresw.com
t. (+44) 01732 863760.

Metrics for accountants

 

The 10th Software Measurement and Improvement Forum presents papers and discussion on delivering business value and informed financial control of software development and maintenance.

Ken Dymond of Process Transition Inc. will look at measuring return from investment in software process improvement. SMS Principal Consultant Graham Lawder will be drawing on experience from recent case history to look at the benefits of an integrated metrics programme, and how to present metrics for accountants.

This forum will be of interest to IT professionals concerned with delivering business value, and for those with responsibility for financial control of IT departments.

This SMIF takes place on 15 February at the Intellect Conference Suite, Russell Square, London. Register for your free place here

COSMIC FFP - finalists for the BCS IT Professional Awards 2006

 

in support of the COSMIC entry :                 (l to r) : David Rogers (EDS), Peter Fagg (SMS) Charles Symons, (COSMIC Team), Ton Dekkers (Shell UK)

 


Primitive counting systems can endanger the health of your improvement programme...

A century ago, explorers in Southern Africa discovered that the Khoisan people, or ‘Hottentots’ as they became known, have one of the most primitive counting systems. They only have words for ‘one’, ‘two’, and ‘many’. A Hottentot who has more than two cows, can only tell you that he has ‘many’ cows, no matter how many.

The IFPUG method of sizing software is comparable to the Hottentot system. For example, an Input elementary process can be counted as 3, 4 or 6 function points, but no more. Does this matter in practice? Well, yes it does, a lot.

..more


BCS SPIN : Promoting best practice  

With a rapidly changing software industry and new challenges facing British software organisations the British SPIN (Software Process Improvement Network) is looking to extend its interest to those areas with the potential to give competitive advantage. It is looking for new members interested in exploring and developing new and innovative ideas for improving software management, development and testing practice, as well as promoting existing SPI good practice.

SPIN encourages practitioners to learn and contribute to the development of software process improvement ideas through participative workshops which encourage members to learn from each other. Clifford Shelley of OSEL Ltd. would like to hear from you if you are interested joining SPIN, and particularly if you would like to help SPIN re-establish itself as a wide ranging and independent process improvement forum.

If you are interested in joining the British SPIN click here

..more

Introduction to the CMMI v1.2

 

Ken Dymond's "A Guide to the CMMI" is being updated for v1.2 and will be available soon from SMS.

Ken is in the UK to deliver an SEI-authorised Introduction to the CMMIv1.2 workshop on 12-14 February 2007.

Featuring the "Dymond Diagrams" as an invaluable learning tool, Ken's presentation of the SEI's material gives participants a thorough and practical grasp of the model and how to apply it to best effect.

Places on the course cost £1000 per head. Register here

..more


BCS @ 50

This year's BCS yearbook celebrates the organisation's 50th Anniversary with a range of articles on IT past, present and future. 

..more

Sizing software by counting ‘one, two, many’ – does it matter? (Yes)

A century ago, explorers in Southern Africa discovered that the Khoisan people, or ‘Hottentots’ as they became known, have the most primitive counting system known to man. They only have words for ‘one’, ‘two’, and ‘many’. A Hottentot who has more than two cows, can only tell you that he has ‘many’ cows, no matter how many.

Other primitive peoples have slightly more sophisticated counting systems. They may also only have words for ‘1’ and ‘2’, but can construct higher numbers by using combinations of the words for ‘1’ and ‘2’, for example

3 = 1 + 2;      4 = 2 + 2;      5 = 2 + 2 + 1, etc.

Interestingly, the most widely used method of sizing software, known as the IFPUG method, is comparable to the Hottentot system. For example, one of the software functions measured by the IFPUG method, an ‘Input elementary process’, can be counted as 3, 4 or 6 ‘function points’, but no more.

Does this matter in practice? Well, yes, recent evidence suggests it does, a lot. Consider the following two cases.

A major European pension fund had been using the IFPUG method for sizing its software and as a basis for estimating. One project turned out to be seriously under-estimated. To investigate how this happened, the software was measured again using the COSMIC-FFP method of sizing software. This method has a size scale for all software functions that starts at 2 Cfsu (COSMIC-FFP size units), rises with increments of 1 Cfsu, and has no upper size limit for any one function.

The comparison found that some elementary processes that scored the maximum 6 or 7 IFPUG function points turned out to be over 60 Cfsu when measured on the COSMIC-FFP method. Furthermore, the processes with size over 40 Cfsu, accounted for almost 80% of the project budget overrun.

(And, by the way, the staff who measured the COSMIC-FFP size for this pension fund knew nothing about the budget over-run, but independently recommended splitting those processes because of their complexity.)

Studies have shown that the sizes of software that has relatively simple elementary processes measured by both the IFPUG and the COSMIC-FFP methods correlate quite well. But as the average size of processes increases, the fact that the IFPUG measurement scale suddenly stops, causes serious distortions. We conclude that IFPUG sizes will be seriously misleading when used for performance measurement or estimating for software with larger, complex transactions.

The major European pension fund has now switched over to the COSMIC-FFP method.

A similar review of sizing methods is going on with a large, global bank. Although they have invested significantly in Software Process Improvement, their project delivery rate (PDR) showed almost no improvement when using IFPUG sizes to measure project work-output. In this case, re-sizing the newer applications using the COSMIC-FFP method has given relatively larger sizes.

The problem has been that the average complexity of the bank’s software functions measured using the IFPUG method has moved slowly towards the maximum possible sizes on that scale. In contrast, COSMIC-FFP measurements properly reflect the increasing sizes of their processes. After substituting COSMIC-FFP sizes, the bank’s PDR shows real improvement.

Apparently, the bank is still hesitating about switching to the COSMIC-FFP sizing method, due to its large set of historical data based on IFPUG sizes. This should not be a barrier. Methods of tracking sizes are frequently re-calibrated in all branches of economics and finance – for example stock market indices are typically re-calibrated every quarter. Re-calibration need not invalidate or make previous data useless.

These findings are pretty devastating. They go a long way towards explaining why so many major projects, which often use the IFPUG method to size the software to be developed, are seriously under-estimated. Now we may also have an explanation why some organizations invest in software process improvement but are unable to measure much benefit from their investment in terms of improved performance. All due to a defective software size measurement method.

In our view, it’s time to move on from the Hottentot era.

COSMIC – the Common Software Measurement International Consortium

For more on the COSMIC-FFP method, click here


BCS SPIN

Introduction

The British SPIN is a specialist group of the British Computer Society, the professional body for IT practitioners and managers.
The objectives of SPIN are: 
  • To be a leadership forum for the free and open exchange of software process improvement experiences and practical ideas
  • To promote the practical and beneficial use of software process improvement in achieving higher levels of business performance
  • To sustain commitment and enhance skills through an active programme of networking, publications, recognition of excellence and mutual support.

Background

Using Software Process Improvement can improve the way an organisation develops, maintains and delivers software systems to achieve real business benefits.

SPIN recognises that its members need support with capability evaluation, understanding and applying Lean and Agile approaches, and the application of models such as ISO 9001/2000, CMMI, ITIL, COBIT, Six Sigma etc. to achieve process improvement.

Organisations faced with the consequences of capability evaluation and management’s demand for improvement are faced with such questions as:

What could I improve?
How do I identify the most cost effective changes?
How can I make these changes succeed?



Activities

Workshops
The group’s main business is carried out in quarterly workshops. These focus on a particular aspect of process improvement.

Birds of a Feather
To maintain the network between workshops, local informal gatherings of members is encouraged. Maintain Contacts Contacts are actively pursued in organisations and groups with similar interest particularly in the USA and Europe.

Newsletter
A quarterly newsletter updates members about meetings and people.

If you are interested in joining the British SPIN click here

Engaging middle managers in improving process performance.

By Jill Pritchet.

Jill is an SMS Principal Consultant specialising in the implementation and upgrade of quality systems to ISO 9001:2000, TickIT and tailored SPI improvement projects.

This article is extracted from a presentation given at the 2006 ESEPG Conference in Amsterdam and appears in the BCS@50 publication.

INTRODUCTION

When an organisation embarks on a change project it is normally for a reason. It may be a reaction to a problem, new management initiative, technical changes, organisational changes, competition etc. Whatever the reason one thing is certain; change is never easy.

Middle managers may see a number of barriers to change: 

  • Resource limitations
  • Too many changesImpact of history
  • Style of the organisation
  • Staff rotation
  • Threat to status
  • Underdeveloped management skills


If one of the aims of a change programme is to improve process performance within the organisation, then we will need to encourage everyone to take a fresh look at what they do, how they do it and how it might be improved.

At the start of a new initiative, it is often assumed that the Middle Managers will somehow automatically jump on board and support the proposals of Senior Management. However, experience shows us that this cannot be relied upon. In reality Senior Management have to win over the Management Team and Key Staff. Failure to do this at the early stages can make the changes more problematic.

This paper considers why Middle Managers may not be enthusiastic in their support for process improvement projects, what some of the barriers to change are, and how to achieve the 'engagement of Middle Managers' in a change project.

WHY ARE MIDDLE MANAGEMENT SO IMPORTANT?

Middle Managers play a crucial role in implementing the policies and strategies set down by Senior Management. They ensure that all the required activities to complete the process, satisfy the customer, comply with regulations or legislation and meet the organisations objectives are carried out.

Middle Management performs a pivotal role in the integration of knowledge and experience, within a given strategic framework to generate practical applications. They are responsible for finding ways of dealing with and alleviating the impacts of problems as they occur, whilst managing employees to ensure that they are well motivated and act responsibly.
They also form the central point of communication, positioned between the Senior Management and the general staff. As part of planning and implementing corporate strategy, they need to communicate with employees to ensure these strategies are interpreted and related to the individual so that they can carry out the appropriate tasks. Equally they will be expected to report back to Senior Management on how business is progressing, the impact of any new initiatives, and any issues raised, that may impact the corporate vision.

They are already key players in the day-to-day smooth running of the operation. It makes sense that they are also an essential element in any change that will affect how the organisation operates.

CATALYSTS FOR CHANGE OR MECHANISM OF MAYHEM?

Middle Managers are the key influencers within any organisation. Not only do they have a great deal of experience within the organisation, they are also the ones making things happen. They are the planners, directing the staff to do the right things at the right time.

For any change programme to succeed it is absolutely essential that Middle Managers support the changes and are actively engaged in the project. But remember, this influence can be either very positive or negative depending on experiences from the past and also the type of individuals they are. So when it comes to implementing process improvement programmes, their performance will determine whether the change initiative is a success or a failure.

In organisations where the Middle Management Team is performing well, it is normally an indicator that they are not only monitoring performance and generating the right change initiatives but also they are able to constantly adapt and improve their strategy during implementation. When change is actively managed in a positive environment, then everyone is encouraged and motivated to embrace change and contribute ideas.

On the other hand, where the Middle Management Team is performing poorly, change becomes increasingly difficult to achieve. Any issues are seen as show stoppers and reasons to turn back to the comfort of the old ways. In this environment, the pace of change itself can slow down or even stop.

Potential Barriers to change

Whenever a new initiative is proposed, one of the most common objections raised is that there are never enough resources available. The day to day activities still need to be supported, while staff need to be released to work on the Change Programme.

As part of best business practice, one of the objectives of Middle Management should be continuous process improvement, so there should be an element of this work in the day to day activities.

If process improvement is important to the organisation, then this should be reflected in how successful improvements and progress is recognised and rewarded. The nature of the reward should reflect the culture of the organisation and what motivates the individuals concerned.

Many Middle Managers believe that their in-depth knowledge and expertise in operating the current practices forms part of their authority. Change however represents uncertainty and will require them to step outside their comfort zone.

Most people prefer to stay with what they know and continue to work as they always have as it is familiar to them and part of their routine. They may think that staying with what they know takes less effort and makes the day-to-day activities run more smoothly. Individuals are not always aware of how inefficient their current way of working might be.

In order to engage Middle Managers in improving process performance, they need to embrace the idea of change. The more that change becomes part of the everyday activity, the more the culture of the organisation becomes associated and accepting of change.

For Middle Managers, this not only means guiding and supporting their staff during change, but also adapting to change themselves.

MEASURE THE EFFECTS OF CHANGE

We have mentioned above that in order to commit to any process improvement or change programme, Middle Managers need to see that it is of benefit to them. Change is only welcomed if it makes life easier, reduces costs or adds value that is in line with the goals and objectives against which management will be assessed.

In order to assess the benefit of change or how much you have improved the process, you must be measuring the activity. You need to establish a benchmark of where things are currently, so that you have a reference point to evaluate the correlation between cause and effect.

The measurements to be used need to be representative of the activity being monitored, unambiguous, quantifiable and repeatable. They must also be acceptable to the managers.

Measurements need to be taken at regular intervals during the project to assess the impact of changes made. You do not want to go too far along the path before realising that the result has been detrimental! Also if you allow too many changes to be made without monitoring the effect, you will not have a clear picture of which changes worked and which didn't.

Regular measurement of the benefits also means that you have evidence to communicate to everyone involved in the project to encourage continued improvements.

THE IMPORTANCE OF COMMUNICATION

Communication is a key factor in all change projects. In order to maintain the momentum of improving process performance it is important to provide regular feedback on what progress is made. If everyone sees the results that are being achieved, they are more likely to continue to move in the direction you want.

Many teams today are literally multi-cultural, but even if everyone is from the same country, they still have their own background, experiences, style and personal preferences. There is no "one size fits all". Different people think differently, communicate differently and respond to different stimuli. When communicating to a group you need to appeal to as many senses as possible to catch the preferences of your audience. Use a variety of communication techniques e.g. visual presentations, reports, meetings, etc. Don't assume that what works for you works for others too!

CONCLUSIONS

Middle Managers are the potential problem in a change project but they are also the potential solution. They hold valuable information about current practices and potential for improving the process performance within the organisation. With the right approach they can become true ‘Ambassadors’ for change and encourage junior staff to contribute to the programme.
Improvement Champion
is published by Software Measurement Services Ltd.
124 High Street Edenbridge, Kent, TN8 5AY


Contact: Sue Rule,Editor
s.rule@measuresw.com
Tel: +44(0)1732 863 760 
Fax: +44(0)1732 864996
For details of all SMS services, together with reference materials, links and news of forthcoming events visit
www.measuresw.com
© Copyright Software Measurement Services Ltd 2006. All rights reserved.