Software Contracting and Agile Development in the Norwegian ICT Industry A Qualitative Survey Anders Ganes Snorre Nævdal Master of Science in Computer Science Submission date: June 2008 Supervisor: Reidar Conradi, IDI Co-supervisor: Hanssen Geir Kjetil, IDI Norwegian University of Science and Technology Department of Computer and Information Science Problem Description As more and more ICT companies migrate to agile methods, the existing contracts and contractual frameworks are mostly tailored for plan-based development. It has been indications from the industry that there are several challenges using agile methods with various contract standards available today. This assignment aims to find out what the potential challenges are with using today’s contract models with agile development. In addition, the assignment will look into what work is done in adapting contract standards to better comply with agile principles. Assignment given: 16. January 2008 Supervisor: Reidar Conradi, IDI Abstract This thesis takes a closer look at how various contract models affect the ability to use agile software development, and the work being done to help improve contractssotheybetterfacilitateagileprinciples. Agile development methods are becoming increasingly popular, while most contractual standards are meant for plan-based development methods. There are done little research regarding the subject, but it is a growing concern in the industry. Work is also currently done on Norwegian contractual frameworks thataimtoupdatetwocommonlyusedcontracts. Theresearchdoneconsistofseveralinterviewswithindustrypractitionersrep- resentingNorwegianconsultancycompaniesexperiencedinagiledevelopment methods. It also includes interviews of contract experts in addition to a search andstudyoftheliteratureonthesubject. Theseinterviewshavebeenanalyzed andcomparedtoliteratureonthesubject. The results of the research revealed a number of challenges with today’s con- tractmodels. Severaldiscoveriesshoweddifficultieswithbureaucracyandcus- tomer involvement. Using traditional contracts could also easily require more upfront work than what is sensible when using agile methods. It is possible to bypassthesechallengesbyeitheradaptingthedevelopmentmethod,orputting thecontractaside. i Preface The assignment for this thesis is given by Department of Computer and Infor- mationScience(IDI)attheTheNorwegianUniversityofScienceandTechnology (NTNU). The context of the work is within the EVidence based Improvment of SOFTwareengineering(EVISOFT)project,partlyfundedbytheResearchCoun- cilofNorway(NorgesForskningsra˚d)underGrant174390/140. Theprojecthas as its goal to improve software engineering processes, and to do this to such a degreethatitbecomeseconomicallybeneficial. Theworkonthisthesishasprogressedthroughthespringof2008andispartof theourMScinComputerScience. Ourworkconsistsofaliteraturestudy,anda surveyofboththeexperiencewithdifferentcontractsforsoftwaredevelopment in combination with agile development methods, and the work done to make twoNorwegiancontractualstandardsmorecomplianttoagileprinciples. The thesis started as a case study at a company called Ciber, who approached the institute with several research topics they would be interested in pursuing along with students at NTNU. After a review of the topics the authors along with the co-supervisor Geir Kjetil Hanssen, agreed to focus the thesis on the potential challenges associated with traditional contracts used in agile projects. TheworkwasplannedasacasestudyusingseveralavailableprojectsatCiber, but after the first round of interviews it became clear that the participants with most experience on the subject were people involved in sale, and negotiation of contracts. We therefore expanded the interviews to include people with this experience in other companies. During the literature study we also discovered the currently ongoing work that Mari Vestre and PROMIS were doing to facil- itate two Norwegian contractual standards, respectively The Norwegian Gov- ernmental Standard Agreement for System Development (agile version) (SSA- S) and Project Management 2000 (PS2000) for easier use with agile methods. Thisdiscoveryleadustofocusonthisaspectaswell. We would like to thank our supervisor Reidar Conradi. We would also like to thank Ciber, represented by Marianne Selle and Sigurd Gimre, for suggesting iii both the cooperation we enjoyed and the basis for the research. Thank you to alltheparticipatingcompaniesandinterviewees,whoseexperiencesandinput on the subject has been invaluable. In particular the people who welcomed us warmly at Ciber (Marianne Selle), ObjectNet (Sverre Tinnen and Lars Ivar Næss), Bekk (Reidar Sande), Promis (Kjetil Strand and Jørgen Petersen) and DIFI(MariVestre). Finally, we would like to thank our co-supervisor Geir Kjetil Hanssen for his guidance and valuable insights on the subject. He has been a driving force throughouttheentireprocessofwritingthisthesis,andhiscommentsandfeed- backhavebeeninvaluabletous. Trondheim,June2008 AndersGanes SnorreNævdal iv Contents 1 Introduction 1 1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Projectcontext . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Problemdefinition . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.4 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.5 DocumentOutline . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 Background 5 2.1 AgileManifesto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.2 Agileterminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.3 Developmentprocess . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.3.1 Start-up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.3.2 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.3.3 Prioritisation . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.3.4 Development/testing . . . . . . . . . . . . . . . . . . . . . 10 2.3.5 Release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.3.6 Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.4 Agileinresearchliterature . . . . . . . . . . . . . . . . . . . . . . . 11 2.5 Agilemotivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.5.1 Customercollaboration,adaptingtochanges . . . . . . . . 13 2.5.2 Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.5.3 ReturnOnInvestement(ROI) . . . . . . . . . . . . . . . . . 15 2.5.4 Improvedcontrol . . . . . . . . . . . . . . . . . . . . . . . . 15 2.5.5 Moraleandproductivity . . . . . . . . . . . . . . . . . . . . 16 2.6 Agilechallenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3 Stateoftheart 19 3.1 Thebuyer-supplierrelationship . . . . . . . . . . . . . . . . . . . . 19 3.2 Thepurposeofcontracts . . . . . . . . . . . . . . . . . . . . . . . . 21 3.3 Agiledevelopmentandcontractmodels . . . . . . . . . . . . . . . 22 v Contents 3.3.1 Whyisagiledifferent? . . . . . . . . . . . . . . . . . . . . . 22 3.3.2 Contractmodels . . . . . . . . . . . . . . . . . . . . . . . . 22 3.4 Norwegiancontractualstandards . . . . . . . . . . . . . . . . . . . 24 3.4.1 PS2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 3.4.2 SSA-S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.4.3 SSA-U . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.5 Challengeswithtoday’scontractmodels . . . . . . . . . . . . . . 29 3.5.1 Agiledevelopmentwithfixed-pricecontracts . . . . . . . 29 3.5.2 Agiledevelopmentwithothercontractmodels . . . . . . . 31 4 Research 33 4.1 Researchgoalandquestion . . . . . . . . . . . . . . . . . . . . . . 33 4.2 Researchapproach . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.3 Researchdesign . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 4.3.1 Interview . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 4.3.2 Interviewsubjects . . . . . . . . . . . . . . . . . . . . . . . 35 4.4 Researchprocess . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4.5 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.6 Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 5 Results 41 5.1 Developmentmanagers . . . . . . . . . . . . . . . . . . . . . . . . 41 5.1.1 SverreTinnen . . . . . . . . . . . . . . . . . . . . . . . . . . 41 5.1.2 ReidarSande . . . . . . . . . . . . . . . . . . . . . . . . . . 43 5.1.3 MarianneSelle . . . . . . . . . . . . . . . . . . . . . . . . . 46 5.1.4 Othersources . . . . . . . . . . . . . . . . . . . . . . . . . . 48 5.2 Contractexperts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 5.2.1 JørgenPetersen,KjetilStrand . . . . . . . . . . . . . . . . . 49 5.2.2 MariVestre . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 6 Analysis 57 6.1 Agiledevelopment . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 6.1.1 Motivationsforworkingagile . . . . . . . . . . . . . . . . . 57 6.1.2 Challengeswithagile . . . . . . . . . . . . . . . . . . . . . 58 6.2 Contractsandagileprojectsingeneral . . . . . . . . . . . . . . . . 58 6.3 Agilemethodsandfixed-pricecontracts(SSA-U) . . . . . . . . . . 58 6.3.1 Motivationsforusingfixedpricecontracts . . . . . . . . . 59 6.3.2 OpinionsonSSA-U . . . . . . . . . . . . . . . . . . . . . . . 59 6.3.3 Challengeswithfixedpricecontracts . . . . . . . . . . . . 59 6.3.4 Positiveaspectswithfixedpriceandagilemethods . . . . 60 vi
Description: