Business Value of Agile Methods: Benefits of Testing Early & Often


Published on

Published in: Technology, Business
  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Business Value of Agile Methods: Benefits of Testing Early & Often

  1. 1. Business Value ofAgile MethodsBenefits of Testing Early & OftenDr. David F. Rico, PMP, ACP, CSMTwitter: @dr_david_f_ricoWebsite: http://www.davidfrico.comLinkedIn:’s Agile Articles:
  2. 2. Author Background DoD contractor with 28+ years of IT experience B.S. Comp. Sci., M.S. Soft. Eng., & D.M. Info. Sys. Large gov’t projects in U.S., Far/Mid-East, & Europe2 Published six books & numerous journal articles Adjunct at George Wash, UMBC, UMUC, Argosy Agile Program Management & Lean Development Specializes in metrics, models, & cost engineering Six Sigma, CMMI, ISO 9001, DoDAF, & DoD 5000 Cloud Computing, SOA, Web Services, FOSS, etc.
  3. 3. Today’s Whirlwind Environment3OverrunsAttritionEscalationRunawaysCancellationGlobalCompetitionDemandingCustomersOrganizationDownsizingSystemComplexityTechnologyChangeVagueRequirementsWork LifeImbalanceInefficiencyHigh O&MLower DoQVulnerableN-M BreachReducedIT Budgets81 MonthCycle TimesRedundantData CentersLack ofInteroperabilityPoorIT SecurityOverburdeningLegacy SystemsObsoleteTechnology & SkillsPine, B. J. (1993). Mass customization: The new frontier in business competition. Boston, MA: Harvard Business School Press.Pontius, R. W. (2012). Acquisition of IT: Improving efficiency and effectiveness in IT acquisition in the DoD. Second AnnualAFEI/NDIA Conference on Agile in DoD, Springfield, VA, USA.
  4. 4. Traditional Projects4 Big projects result in poor quality and scope changes Productivity declines with long queues/wait times Large projects are unsuccessful or canceledJones, C. (1991). Applied software measurement: Assuring productivity and quality. New York, NY: McGraw-Hill.Size vs. QualityDefectDensity0.003.206.409.6012.8016.000 2 6 25 100 400Lines of Code (Thousands)Size vs. ProductivityCodeProductionRate0. 2 6 25 100 400Lines of Code (Thousands)Size vs. Requirements GrowthPercentage0%8%16%24%32%40%0 2 6 25 100 400Lines of Code (Thousands)Size vs. SuccessPercentage0%12%24%36%48%60%0 2 6 25 100 400Lines of Code (Thousands)
  5. 5. Global Project Failures5Standish Group. (2010). Chaos summary 2010. Boston, MA: Author.Sessions, R. (2009). The IT complexity crisis: Danger and opportunity. Houston, TX: Object Watch. Challenged and failed projects hover at 67% Big projects fail more often, which is 5% to 10% Of $1.7T spent on IT projects, over $858B were lost16% 53% 31%27% 33% 40%26% 46% 28%28% 49% 23%34% 51% 15%29% 53% 18%35% 46% 19%32% 44% 24%33% 41% 26%0% 20% 40% 60% 80% 100%199419961998200020022004200620082010YearSuccessful Challenged Failed$0.0$0.4$0.7$1.1$1.4$1.82002 2003 2004 2005 2006 2007 2008 2009 2010Trillions(USDollars)Expenditures Failed Investments
  6. 6. Requirements Defects & Waste6Sheldon, F. T. et al. (1992). Reliability measurement: From theory to practice. IEEE Software, 9(4), 13-20Johnson, J. (2002). ROI: Its your job. Extreme Programming 2002 Conference, Alghero, Sardinia, Italy. Requirements defects are #1 reason projects fail Traditional projects specify too many requirements More than 65% of requirements are never used at allOther 7%Requirements47%Design28%Implementation18%DefectsAlways 7%Often 13%Sometimes16%Rarely19%Never45%Waste
  7. 7. What is Agility? A-gil-i-ty (ə-ji-lə-tē) Property consisting of quickness,lightness, and ease of movement; To be very nimble The ability to create and respond to change in order toprofit in a turbulent global business environment The ability to quickly reprioritize use of resources whenrequirements, technology, and knowledge shift A very fast response to sudden market changes andemerging threats by intensive customer interaction Use of evolutionary, incremental, and iterative deliveryto converge on an optimal customer solution Maximizing BUSINESS VALUE with right sized, just-enough, and just-in-time processes and documentationHighsmith, J. A. (2002). Agile software development ecosystems. Boston, MA: Addison-Wesley.7 
  8. 8. What are Agile Methods?8 People-centric way to create innovative solutions Product-centric alternative to documents/process Market-centric model to maximize business valueAgile Manifesto. (2001). Manifesto for agile software development. Retrieved September 3, 2008, from http://www.agilemanifesto.orgRico, D. F., Sayani, H. H., & Sone, S. (2009). The business value of agile software methods. Ft. Lauderdale, FL: J. Ross Publishing.Rico, D. F. (2012). Agile conceptual model. Retrieved February 6, 2012, from CollaborationWorking SoftwareIndividuals & InteractionsResponding to Changevaluedmore thanvaluedmore thanvaluedmore thanvaluedmore thanContractsDocumentationProcessesProject Plans Frequent comm. Close proximity Regular meetings Multiple comm. channels Frequent feedback Relationship strength Leadership Boundaries Empowerment Competence Structure Manageability/Motivation Clear objectives Small/feasible scope Acceptance criteria Timeboxed iterations Valid operational results Regular cadence/intervals Org. flexibility Mgt. flexibility Process flexibility System flexibility Technology flexibility Infrastructure flexibility Contract compliance Contract deliverables Contract change orders Lifecycle compliance Process Maturity Level Regulatory compliance Document deliveries Document comments Document compliance Cost Compliance Scope Compliance Schedule Compliance
  9. 9. NetworkComputerOperating SystemMiddlewareApplicationsAPIsGUIHow Agile Works Agile requirements implemented in slices vs. layers User needs with higher business value are done first Reduces cost & risk while increasing business success9Shore, J. (2011). Evolutionary design illustrated. Norwegian Developers Conference, Oslo, Norway.Agile Traditional1 2 3 Faster Early ROI Lower Costs Fewer Defects Manageable Risk Better Performance Smaller Attack SurfaceLate No Value Cost Overruns Very Poor Quality Uncontrollable Risk Slowest Performance More Security Incidents Seven Wastes1.Rework2.Motion3.Waiting4.Inventory5.Transportation6.Overprocessing7.OverproductionMINIMIZES MAXIMIZES JIT, Just-enough architecture Early, in-process system V&V Fast continuous improvement Scalable to systems of systems Maximizes successful outcomes Myth of perfect architecture Late big-bang integration tests Year long improvement cycles Breaks down on large projects Undermines business success
  10. 10. What is Agile Testing? Traditional testing is a late, manual process Agile testing is an early and automated process The goal of agile testing is to deliver early and often10Rico, D. F. (2012). Agile testing resources. Retrieved Sep. 9, 2012, from, L., & Gregory, J. (2009). Agile testing: A practical guide for testers and agile teams. Boston, MA: Addison-Wesley.Grant, T. (2005). Continuous integration using cruise control. Northern Virginia Java Users Group (Novajug), Reston, Virginia, USA.Traditional TestingCombining source filesCombining software and environmentCombining software and dataCombining software and testsCombining developersAgile TestingCode is frequently checked inCode is automatically retrievedCompilation is done automaticallyTests are done automaticallyCode reports are generatedDevelopers get instant feedbackCode is automatically deployed orpackaged for delivery
  11. 11. Basic—Test Driven Development Term coined by Kent Beck in 2003 Consists of writing all tests before design Ensures all components are verified and validated11Beck, K. (2003). Test-driven development: By example. Boston, MA: Addison-Wesley.
  12. 12. Thousands of TestsContinuously ExecutedNo More Late BigBang IntegrationAdvanced—Continuous Integration User needs designed & developed one-at-a-time Changes automatically detected, built, and tested System fully tested and deployed as changes occur12Humble, J., & Farley, D. (2011). Continuous delivery. Boston, MA: Pearson Education.Duvall, P., Matyas, S., & Glover, A. (2006). Continuous integration. Boston, MA: Addison-Wesley.BuildIntegrationServerVersionControlServerBuildScriptsUsesWatchesBuildStatusProvidesDeveloper ADeveloper BDeveloper CCommitsChangesCommitsChangesCommitsChangesBuildsDatabaseAnalysisTestingReportingDocumentationDeploymentEarly, Automated, Fast,Efficient, & RepeatableConstant ReadinessState & CM ControlLean, Waste Free, Low WIP,No Deadlocked Test QueuesRapidly & SuccessfullyDev. Complex Systems
  13. 13. Agile Testing Done Early & Often Eliminates big-bang integration in the 11th hour Creates a repeatable and reliable testing process Evaluates system-wide changes throughout project13Maeda, M. K. (2009). Agile testing: Early, often, and Smart. Arlington, MA: Cutter Consortium.
  14. 14. Agile Testing Practices Agile testing consists of seven broad practices Includes automated builds, testing, inspections, etc. Also includes reporting, documentation, deployment, etc.14PracticeBuildingDatabaseInspectionsTestingFeedbackDocumentationDeploymentDescriptionFrequently assembling products and services to ensure delivery readinessFrequently generating/analyzing database schemas, queries, and formsFrequently performing automated static analysis of product/service qualityFrequently performing automated dynamic product and service evaluationFrequently generating automated status reports/messages for all stakeholdersFrequently performing automated technical/customer document generationFrequently performing automated delivery of products/services to end usersDuvall, P., Matyas, S., & Glover, A. (2006). Continuous integration: Improving software quality and reducing risk. Boston, MA: Addison-Wesley.Humble, J., & Farley, D. (2011). Continuous delivery. Boston, MA: Pearson Education.
  15. 15. Agile Testing Workflow15Traditional vs. Agile Cumulative FlowWork(Story,Point,Task)orEffort(Week,Day,Hour)Time Unit (Roadmap, Release, Iteration, Month, Week, Day, Hour, etc.)Work(Story,Point,Task)orEffort(Week,Day,Hour)Time Unit (Roadmap, Release, Iteration, Month, Week, Day, Hour, etc.)Traditional Cumulative Flow Agile Cumulative Flow Late big bang integration increases WIP backlog Agile testing early and often reduces WIP backlog Improves workflow and reduces WIP and lead timesAnderson, D. J. (2004). Agile management for software engineering. Upper Saddle River, NJ: Pearson Education.Anderson, D. J. (2010). Kanban: Successful evolutionary change for your technology business. Sequim, WA: Blue Hole Press.
  16. 16. Agile Testing Costs & BenefitsGrant, T. (2005). Continuous integration using cruise control. Northern Virginia Java Users Group (Novajug), Reston, Virginia, USA.Fredrick, J. (2008). Accelerate software delivery with continuous integration and testing. Japanese Symposium on Software Testing, Tokyo, Japan. Most agile testing tools are “free” open source A build server is no more than a commodity PC 10x more efficient/effective than traditional testing16
  17. 17. Agile Testing Economics Traditional testing finds a defect in about 10 hours Manual code inspections find a defect in 1 hour Agile testing finds a defect every 6 minutes17Rico, D. F. (2012). The Cost of Quality (CoQ) for Agile vs. Traditional Project Management. Fairfax, VA: Gantthead.Com.
  18. 18. Agile Cost of Quality (CoQ) Agile testing is 10x better than code inspections Agile testing is 100x better than traditional testing Agile testing is done earlier “and” 1,000x more often18Rico, D. F. (2012). The Cost of Quality (CoQ) for Agile vs. Traditional Project Management. Fairfax, VA: Gantthead.Com.
  19. 19. Agile Testing Statistics Fewer builds leave in higher bug counts A high number of builds eliminates the defects Goal is to have as many, early builds as possible19Lacoste, F. J. (2009). Killing the gatekeeper: Introducing a continuous integration system. Proceedings of the Agile 2009 Conference, Chicago, Illinois, USA, 387-392. 
  20. 20. How Google Tests Software Google early adopter of agile methods and Scrum Google also uses agile testing at enterprise scale 15,000 developers run 75+ million tests per day20Micco, J. (2013). Continuous integration at google scale. Eclipse Con, Boston, MA.Whittaker, J., Arbon, J., & Carollo, J. (2012). How google tests software. Upper Saddle River, NJ: Pearson Education. 15,000+ developers in 40+ offices 4,000+ projects under active development 5,500+ submissions per day on average Single monolithic code tree with mixed language code Development on one branch - submissions at head All builds from source 20+ sustained code changes per minute with 60+ peaks 50% of code changes monthly 75+ million test cases run per day
  21. 21. Scaling Agile Testing Agile testing slows down with very large systems Slow testing slows integration and increases bugs Agile testing can speed back up with proper attention21Kokko, H. (2009). Increase productivity with large scale continuous integration. Proceedings of the Agile 2009 Conference, Chicago, Illinois, USA.Wide Impact TuningFast builds – less changes – more greenParallelization of test runsClearCase to subversionPre-installing as much as possibleRemoval of randomnessCompilation in memoryInstallation starting parallel with systembuildFocused Impact TuningMore memory and CPUsParallelize buildsReplace 3rd party test librariesReduce/remove timeouts in testsSelect different testsRefactor code & componentsTune the network & softwareTune the database
  22. 22. General Agile Metrics Agile methods are based on traditional measures Velocity, burnup, and burndown are basic measures Top-notch shops use business value, Agile EVM, etc.22Rico, D. F., Sayani, H. H., & Sone, S. (2009). The business value of agile software methods. Ft. Lauderdale, FL: J. Ross Publishing.Basic Metrics Velocity Productivity Cycle time Effort Cost Schedule System quality Customer SatisfactionBusiness ValueBurnup & BurndownAdvanced MeasuresAgile Earned Value Epic burnup & burndown Feature burnup & burndown Release burnup & burndown Iteration burnup & burndown Weekly burnup & burndown Daily burnup & burndown Business value per epic Business value per feature Business value per release Business value per iteration Business value per week Business value per day Business value per story Story points per release Planned sprints per release Planned budget per release Current to planned sprint Current to planned points Planned release points Release points completed Customer collaboration Customer trust & loyalty Teamwork cohesion & quality Purpose, autonomy, & mastery Workflow capacity & throughput Workflow efficiency & reliability Organizational culture & agility Information systems flexibility
  23. 23. Agile Testing Metrics Agile testing also based on traditional metrics Agile testing measures include basic volumetrics Key metrics are RTF, code coverage, complexity, etc.23Rico, D. F., Sayani, H. H., & Sone, S. (2009). The business value of agile software methods. Ft. Lauderdale, FL: J. Ross Publishing.Volumetrics Number of user stories Number of story points Number of lines of code Number of function points Number of check-ins Number of unit tests Number of builds Number of deploymentsCode CoverageRunning Tested CodeDefect DensityCode Complexity Running tested epics Running tested features Running tested releases Running tested iterations Running tested builds Running tested user stories Epic coverage Feature coverage Release coverage Iteration coverage Build coverage Statement coverage Function coverage Other coverage types Epic complexity Feature complexity Release complexity Iteration complexity Build complexity Cyclomatic complexity Actual complexity Other complexity types Defects per epic Defects per feature Defects per release Defects per iteration Defects per build Defects per user story Defects per story point Defects per other types
  24. 24. Agile Cost & Benefit Analysis Costs based on avg. productivity and quality Productivity ranged from 4.7 to 5.9 LOC an hour Costs were $588,202 and benefits were $3,930,63124Rico, D. F., Sayani, H. H., & Sone, S. (2009). The business value of agile software methods: Maximizing ROI with just-in-time processes and documentation.Ft. Lauderdale, FL: J. Ross Publishing.d1 = [ln(Benefits  Costs) + (Rate + 0.5  Risk2)  Years]  Risk   Years, d2 = d1  Risk   Years 51i
  25. 25. Studies of Agile Methods Dozens of surveys of agile methods since 2003 100s of Agile and CMMI case studies documented Agile productivity, quality, and cost better than CMMI25Rico, D. F. (2008). What is the return-on-investment of agile methods? Retrieved February 3, 2009, from, D. F. (2008). What is the ROI of agile vs. traditional methods? TickIT International, 10(4), 9-18.
  26. 26. Benefits of Agile Methods Analysis of 23 agile vs. 7,500 traditional projects Agile projects are 54% better than traditional ones Agile has lower costs (61%) and fewer defects (93%)Mah, M. (2008). Measuring agile in the enterprise: Proceedings of the Agile 2008 Conference, Toronto, Canada.Project Cost in Millions $0.751.502. AgileAfter Agile61%LowerCostTotal Staffing1811Before AgileAfter Agile39%LessStaff5101520Delivery Time in Months51015201813.5Before AgileAfter Agile24%FasterCumulative Defects6251250187525002270381Before AgileAfter Agile93%LessDefects26
  27. 27. Agile vs. Traditional Success Traditional projects succeed at 50% industry avg. Traditional projects are challenged 20% more often Agile projects succeed 3x more and fail 3x less oftenStandish Group. (2012). Chaos manifesto. Boston, MA: Author.27Agile TraditionalSuccess42%Failed9%Challenged49%Success14%Failed29%Challenged57%
  28. 28. Hoque, F., et al. (2007). Business technology convergence. The role of business technology convergence in innovationand adaptability and its effect on financial performance. Stamford, CT: BTM Institute.28 Study of 15 agile vs. non-agile Fortune 500 firms Based on models to measure organizational agility Agile firms out perform non agile firms by up to 36%Benefits of Organizational Agility
  29. 29. Agile Enterprise Delivery ModelBeck, K., & Fowler, M. (2001). Planning extreme programming. Upper Saddle River, NJ: Addison-Wesley.Highsmith, J. A. (2010). Agile project management: Creating innovative products. Boston, MA: Pearson Education.Larman, C., & Vodde, B. (2010). Practices for scaling lean and agile development. Boston, MA: Addison-Wesley.Leffingwell, D. (2011). Agile software requirements: Lean requirements practices for teams, programs, and the enterprise. Boston, MA: Pearson Education. Begins with a high-level product vision/architecture Continues with needs development/release planning Includes agile delivery teams to realize business value29
  30. 30. Agile Adoption30House, D. (2012). Sixth annual state of agile survey: State of agile development. Atlanta, GA: VersionOne. VersionOne found 80% using agile methods today Most are using Scrum with several key XP practices Lean-Kanban is a rising practice with a 24% adoption ContinuousIntegration●●●●●●●●●●●●
  31. 31. Agile ProliferationScrum Alliance. (2012). Scrum certification statistics. Retrieved February 6, 2013, from, D. K. (2012). Agile developers needed: Demand outpaces supply. Foster City, CA: eWeek. 31 Number of CSMs have doubled to 200,000 in 2 years 558,918 agile jobs for only 121,876 qualified people 4.59 jobs available for every agile candidate (5:1)
  32. 32. Agile Industry Case Studies 80% of worldwide IT projects use agile methods Includes regulated industries, i.e., DoD, FDA, etc. Agile now used for safety critical systems, FBI, etc.32IndustryShrinkWrappedElectronicCommerceHealthCareLawEnforcementOrg 20 teams 140 people 5 countriesSize 15 teams 90 people Collocated 4 teams 20 people Collocated 10 teams 50 people Collocated 3 teams 12 people CollocatedU.S.DoDPrimaveraGoogleStratcomFBIFDAProjectPrimaveraAdwordsSKIwebSentinelm2000PurposeProjectManagementAdvertisingKnowledgeManagementCase FileWorkflowBloodAnalysis 1,838 User Stories 6,250 Function Points 500,000 Lines of CodeMetrics 26,809 User Stories 91,146 Function Points 7,291,666 Lines of Code 1,659 User Stories 5,640 Function Points 451,235 Lines of Code 3,947 User Stories 13,419 Function Points 1,073,529 Lines of Code 390 User Stories 1,324 Function Points 105,958 Lines of CodeRico, D. F. (2010). Lean and agile project management: For large programs and projects. Proceedings of the First International Conference on LeanEnterprise Software and Systems, Helsinki, Finland, 37-43.
  33. 33.  Structure, reward, decision, staffing, leadership, etc. Top-down, individualism, regulation, compliance, etc. Focus on reforming acquisition & procurement system33Type/Kind Common DoD Agile PerceptionsDisciplineReality with Respect to Agile MethodsScalabilityDomainManagementRequirementsArchitectureQualityInspectionsSecurityUndisciplined Cowboy CodingOnly Applies Small ProjectsOnly for Protoperational SystemsFlexible Scope/Cant Use EVMDoesnt Use RequirementsSpaghetti Code from IterationsNo Documents/UnmaintainableHigh CoQ from No InspectionsVulnerabilities from Hacking Rigorous process, plans, requirements, QA, CM, testing, documents etc. Used by 100, 500, 1,000, 10,000+ person person projects & organizations Used in DoD, medical devices, avionics, autos, electronics, etc. Lightweight EVM model is used with its release planning methodology Always begins with valuable, well-defined, & prioritized requirements Begins with lean architecture or create waste-free emergent design Electronic plans, requirements, designs, tests, manuals, documents, etc. One or two orders of magnitude more inspections & tests performed Security practices result in smaller attack surface & fewer vulnerabilitiesRico, D. F., Sayani, H. H., & Sone, S. (2009). The business value of agile software methods: Maximizing ROI with just-in-time processes and documentation.Ft. Lauderdale, FL: J. Ross Publishing.Perceptions of Agile Methods
  34. 34. Agile World View “Agility” has many dimensions other than IT It ranges from leadership to technological agility The focus of this brief is program management agility Agile LeadersAgile Organization ChangeAgile Acquisition & ContractingAgile Strategic PlanningAgile Capability AnalysisAgile Program ManagementAgile Tech.Agile Information SystemsAgile ToolsAgile Processes & PracticesAgile Systems DevelopmentAgile Project Management34
  35. 35. Agile Recap Agile methods DON’T mean deliver it now & fix it later Lightweight, yet disciplined approach to development Reduced cost, risk, & waste while improving quality35Rico, D. F. (2012). What’s really happening in agile methods: Its principles revisited? Retrieved June 6, 2012, from, D. F. (2012). The promises and pitfalls of agile methods. Retrieved February 6, 2013 from,, D. F. (2012). How do lean & agile intersect? Retrieved February 6, 2013, from How ResultFlexibility Use lightweight, yet disciplined processes and artifacts Low work-in-processCustomer Involve customers early and often throughout development Early feedbackPrioritize Identify highest-priority, value-adding business needs Focus resourcesDescope Descope complex programs by an order of magnitude Simplify problemDecompose Divide the remaining scope into smaller batches Manageable piecesIterate Implement pieces one at a time over long periods of time Diffuse riskLeanness Architect and design the system one iteration at a time JIT waste-free designSwarm Implement each component in small cross-functional teams Knowledge transferCollaborate Use frequent informal communications as often as possible Efficient data transferTest Early Incrementally test each component as it is developed Early verificationTest Often Perform system-level regression testing every few minutes Early validationAdapt Frequently identify optimal process and product solutions Improve performance
  36. 36. Conclusion36 Agility is the evolution of management thought Confluence of traditional and non-traditional ideas Improve performance by over an order of magnitude“The world of traditional methods belongs to yesterday”“Don’t waste your time using traditional methods on 21st century projects”Agile methods are …Systems development approachesNew product development approachesExpertly designed to be fast and efficientIntentionally lean and free of waste (muda)Systematic highly-disciplined approachesCapable of producing high quality systemsRight-sized, just-enough, and just-in-time tools Scalable to large, complex mission-critical systems Designed to maximize business value for customersWysocki, R.F. (2010). Adaptive project framework: Managing complexity in the face of uncertainty. Boston, MA: Pearson Education. 
  37. 37. Books on ROI of SW Methods Guides to software methods for business leaders Communicates business value of software methods Rosetta stones to unlocking ROI of software methods (Description) (Description)37