Agile - User stories, Backlog Grooming & Relative Estimation
- 1. User Stories, Prioritization, Estimation & Enhanced Process Flow Rosario Poulraj
- 2. INVEST Model NDEPENDENT EGOTIABLE ALUABLE STIMABLE MALL ESTABLE
- 3. SMART Tasks MART EASURABLE CHIEVABLE EALISTIC IMEBOUND
- 4. Functional vs Technical Stories
- 5. Spike Story • Used for activities such as research, design, investigation, exploration, and prototyping • Cannot be estimated until a development team runs a time boxed investigation • The output of a spike is an estimate for the original story
- 6. Definition of Done (Technical Stories) Code Review Documentation Unit Testing Demo
- 7. Definition of Done (Spike Story) Proof Of Concept (POC) Review (for the approach) Further Story Split up (if feasible) Estimation of related Stories Documentation
- 8. Prioritization - Factors & Models Kano Model 1. Threshold (Must have) 2. Linear Features 3. Exciters & Delighters MoSCoW Model 1. Must have 2. Should have 3. Could have 4. Won't have Risk-Value Model 1. High Risk – High Value 2. Low Risk – High Value 3. Low Risk – Low Value 4. High Risk – Low Value Core Factors Financial Value Cost of development Amount of Learning while development Amount of risk can be removed by developing
- 9. Patterns for Splitting User Stories Strategy 1: Breaking down by workflow steps Strategy 2: Breaking down by business rules (Special Cases) Strategy 3: Breaking down by happy / unhappy flow (Validations) Strategy 4: Breaking down by input options / platform (mobile, desktop, touchscreen) Strategy 6: Breaking down by operations (CRUD operations - Create, Update or Delete) Strategy 7: Breaking down by test scenarios / test case Strategy 8: Break down items based on identified acceptance criteria Strategy 9: Break Out a Spike
- 10. Daimler AG Software Estimation - Techniques Titel der Präsentation / Abteilung / Datum / Seite 11 Work Breakdown Structure - WBS (Man Days) Function Points (Functional Size Measurement (FSM) of Software) Relative Sizing (Story Points / T-shirt)
- 11. Daimler AG Merits of Relative Estimation (Story Points – High Level Planning) Titel der Präsentation / Abteilung / Datum / Seite 12 Commitment as a “Team” rather than “Individual” (High level plan) Story-points estimation is typically faster and Easy to perform Story Points reduces fear of commitment (individual) Less Pressure on the team, as there is no hourly estimate in the Story level Less stress brings better estimates as a Team Story Points invites collaboration as team behavior becomes prominent over individuals There is credible evidence that humans are good in relative estimation compared to absolute.
- 12. Daimler AG Merits of Task Estimates (in Hours) Titel der Präsentation / Abteilung / Datum / Seite 13 Absolute Estimates at Lower (Resource) level Really feasible (in hours) at the task level rather than the story level (big piece) Hourly estimate is more specific to the team member’s ability, hence it differs from person to person Individuals can refer the previous tasks, learn from the mistakes to achieve the perfection in upcoming Task estimates Based on the available capacity, the velocity can be re-defined
- 13. Daimler AG Backlog Grooming - Process Titel der Präsentation / Abteilung / Datum / Seite 15
- 14. Daimler AG Sprint Planning (Project XXXXXX) Titel der Präsentation / Abteilung / Datum / Seite 16
- 15. Project : XXXXXXX Problem : Too big stories could not be completed in single sprint End-goal :Refined Stories Solution : Altered workflow with enhanced backlog grooming