AUGUS T 2004 V1.4 I B T E C H N O L O G Y E X A M P L E S O F C M M I B E N E F I T S James Tower Contents Page JPMorgan Investment Bank Technology Overview 1. Reducing Project Slippage 2. Defect Reduction 3. Cost Savings & ROI 4. Compliance with Sarbanes Oxley 5. CMMI Level 3 - Further Defect Reduction 6. CMMI Level 3 - Increased Productivity 7. Views from Project Managers 8 -14
Overview CMMI IMPACT ON PROJECT PERFORMANCE CMMI IMPACT ON PROJECT PERFORMANCE CMMI LEVEL 2 CMMI LEVEL 3 Improves projects’ “Project Management “Defect Prevention ability to deliver on Best Practices” Best Practices” schedule � Requirements Management � Validat ion Reduces the number � Proj ect Planning, � Testing of defects in the � Peer reviews Monitoring and Control delivered software � Vendor Management � Quality Assurance O F C M M B E N E F I T S � Quality Assurance � Continuous Improvement � Configuration Management � Best Practice S haring � Metrics and Analysis � Organizational Learning REDUCED CUSTOMER COSTS SATISFACTION Addresses 4 of the 6 technology control SARBANES- OXLEY objectives of Sarbanes-Oxley sect. 404 COMPLIANCE IB TECHNOLOGY: 53 out of 59 groups (c.1658 technologists) at CMMI Level 2 and two groups (of c.130 staff) at Level 3. I B T E C H N O L O G Y - E X A M P L E S IB TECHNOLOGY: 53 out of 59 groups (c.1658 technologists) at CMMI Level 2 and two groups (of c.130 staff) at Level 3. • JUPITER – SWAPS BACK • CM NA - LOAN • CONFIRMATIONS • TOKYO - TRADING TECHNOLOGY PROGRAMME, FMPS ORIGINATION/SYNDICATION • REFERENCE DATA AND TRANSACTION PROCESSING • NA INTEREST RATE • ISIS • MOSS • NAPOLI (GLOBALLY) TECHNOLOGY - SETE • CORE • SBL • CLIENT CONNECTIVITY - NORTH • FORT - CONCORDE & OPEN • GEM-EDT • CLIENT SERVER AMERICA SERVICES • MARKETS CREDIT RISK • SCI • DELAWARE EQUITIES • MARKET RISK TECHNOLOGY • C-CLEAR • CSW BUILD • CSW - NORTH AMERICA • FORT – IRD TECH • TCP & DAC PROGRAM • EXCHANGE CONNECTIVITY • AUTO TRADING (+ OTHERS OF • FAMIS • SYDNEY EQUITIES AD • AUTO TRADING EDG) • DOCUMENTUM • EQUITIES RESEARCH ASIA • RISK APPLICATIONS • ECDM + NECTAR • IB PORTAL PACIFIC AD • REFERENCE DATA • RESEARCH - NORTH AMERICA • SALES MIS ( 3 APPS) • HONG KONG EQUITIES AD • CONVERTS • PTP, PYRAMID SERVICES • FLOWBIZ • TOKYO EQUITIES AD • ALADDIN • SCALA • JETS • ASIA TREASURY AND C&R • SATURN/CLIENT VALUATION • RELEASE MANAGEMENT • B2B HONG KONG • EMEA RESEARCH • ASIA REGIONAL CREDIT & JPMSA • ASIA REGIONAL RATES • ASIA IBC M&A • ASIA F&O AD SYDNEY RATES PAGE 1 OF 14
Reducing Project Slippage: average slippage of delivery date by 70-80% Before using CMMI the E-Trading team were delivering with an average slippage of 6-8 weeks. When they achieved CMMI Level 2 the average slippage reduced to one week. E-Trading Projects Milestone Performance O F C M M B E N E F I T S Average Slippage of Delivery Date (In Weeks) 8 Achieved CMMI Level 2 7 6 5 Weeks 4 I B T E C H N O L O G Y - E X A M P L E S 3 Average Slippage 2 (In Weeks) 1 0 Q2 2002 Q3 2002 Q4 2002 Q1 2003 Q2 2003 Q3 2003 Q4 2003 Q1 2004 Q2 2004 Quarter PAGE 2 OF 14
Defect Reduction: defects reduced by more than 80% Asia Treasury and Credit Rates achieved CMMI level 2 at the end of 2003. In the subsequent 6 months* their average number of UAT & production defects dropped by more than 80% (18 proj ects) O F C M M B E N E F I T S 10.00 10.00 9.00 9.00 Achieved CMMI Level 2 8.00 8.00 7.00 7.00 6.00 Average # of UAT defects 6.00 Average # of UAT defects per project per project 5.00 5.00 Average # of Production Average # of Production 4.00 4.00 I B T E C H N O L O G Y - E X A M P L E S defects per project defects per project 3.00 3.00 2.00 2.00 1.00 1.00 0.00 0.00 Q4 2003 Q1 2004 Q2 2004 Q4 2003 Q1 2004 Q2 2004 PAGE 3 OF 14
Cost Savings & ROI: $1m per year cost savings and an ROI of 5:1 For MOS S , achieving CMM Level 2 resulted in fewer defects introduced in development - as a consequence, production support effort dropped 14.1% Taking into account the expense of process improvement (3% ) and quality assurance (2.6% ), the net effort improvement was 8.5% . O F C M M B E N E F I T S This translated into savings of $1,005,618 per year for a 100 person team. An ROI of 5:1. Overnight Errors I B T E C H N O L O G Y - E X A M P L E S Example from a back-office settlement and accounting system PAGE 4 OF 14
CMMI Level 3 - Further defect reduction of 50% CMMI Level 2 reduced Concorde’ s post release defects by more than 40% . On achieving Level 3 a further reduction of more than 50% was achieved. Post Development Defects/Week O F C M M B E N E F I T S 70 60 Achieved CMM ML2 2001 August 50 Achieved CMMI ML3 2003 January 40 30 I B T E C H N O L O G Y - E X A M P L E S 20 10 0 2001 2002 2003 2004 Year PAGE 5 OF 14
CMMI Level 3 - Increased productivity The effort normally expended in fixing problems was instead translated into new business functionality with fewer defects per unit of output. Three releases per year in 2001 increased to 6 releases in 2003. 2001 - 2003 Support vs Development Effort O F C M M B E N E F I T S 3.5 3 2.5 Relative Effort 2 Level 2 Support Level 3 Support 1.5 I B T E C H N O L O G Y - E X A M P L E S Development 1 0.5 0 2001 2003 Year PAGE 6 OF 14
Compliance with Sarbanes-Oxley CMMI Level 2 processes and deliverables to satisfy four out of the six IT control obj ectives required by S arbanes-Oxley. CONTROL CATEGORY CMM LEVEL 2 TEAMS’ PROCESSES & DELIVERABLES CONTROL CATEGORY CMM LEVEL 2 TEAMS’ PROCESSES & DELIVERABLES Each application team will be using either a CTO group-wide lifecycle methodology or application team-specific methodology CHANGE CONTROL O F C M M B E N E F I T S (sometimes called a "Code of Practice"). These will describe the change control procedures. Quality assurance audits/reviews will PROCESS: be done regularly on the teams to check that they are following the procedures. CMM mini-assessments will be conducted every 6-8 months - these will also check that the procedures exist and there is evidence of them being followed. Each application team will be using either a CTO group-wide lifecycle methodology or application team-specific methodology EMERGENCY CHANGES: (sometimes called a "Code of Practice"). These will describe the emergency change control procedures. Evidence to use to illustrate compliance - 1) Lifecycle methodology document (CTO-groups standard or application team code of practice; 2) CAN (Change Activity Notice) procedures, Release Management procedures, actual CAN records. Each application team will be using either a CTO group-wide lifecycle methodology or application team-specific methodology PROJECT LIFE CYCLE: I B T E C H N O L O G Y - E X A M P L E S (sometimes called a "Code of Practice"). Quality assurance audits/reviews will be done regularly on the teams to check that they are following the lifecycle. CMM mini-assessments will be conducted every 6-8 months - these will also check that the procedures exist and there is evidence of them being followed. The application team's lifecycle methodology will describe or reference their testing procedures. evidence to use to illustrate TESTING: � Possible evidence to use to illustrate compliance - 1) Lifecycle methodology document (CTO-groups standard or application team compliance - 1) Lifecycle methodology document/ testing procedures; 2) Records of test sign-off 3) Actual test results 4) "Permit code of practice; 2) SQA review reports will illustrate compliance of actual practice against the lifecycle methodology. to Operate" (a gateway review before release to production) records (note some groups have not introduced permit-to-operate) N.B. The two remaining Control Categories “Application Logical Access Controls” & “Access Administration Process” are not covered by the CMM PAGE 7 OF 14
Views from Project Managers - Cross Business Ecommerce Gerald O’Leary – CBT (E-Trading) “Before we started there was no predictability; the business was crying out for us to commit to milestones, and the motivation of the team was low. O F C M M B E N E F I T S Now the improvement of performance is recognized by the business and the motivation is back as their work hours become more predictable and business partnership has grown. I B T E C H N O L O G Y - E X A M P L E S Another benefit of having well know roles and process was that we were able to double the size of the team rapidly and rapidly integrate and reap benefit from these additional resources.” PAGE 8 OF 14
Recommend
More recommend
