Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
SlideShare a Scribd company logo

Effort Estimation - Chinmay VS - Scrum Bangalore 19th Meetup

•Download as PPTX, PDF•
•
Scrum Bangalore
Scrum Bangalore

The document discusses reasons why estimates tend to be wrong and ways to improve estimating. It identifies five common reasons for inaccurate estimates: ambiguous requirements, assumptions instead of facts, lack of breakdown structure, single-point estimates, and poor work processes. It also notes that off-the-cuff estimates in response to vague questions like "can this be done by Friday?" contribute to errors. The document advocates for more accurate estimation techniques like defining requirements, breaking down work, and acknowledging uncertainty through ranges rather than precise dates.

1 of 10
EFFORT ESTIMATION Write down the number of “squares” in the above image…
5 REASONS WHY ESTIMATES TEND TO BE WRONG 1. Ambiguous/Incomplete Requirements. No documented use-cases. 2. Using Assumptions/Gut-feeling. Instead of facts. 3. Large-Scale. No breakdown structure = Omitted activities. 4. Single-Point Estimates. Average ≠ Conditional. 5. Bad work processes Bad design, code  Bugs  Rework/Firefighting.
THE 6TH SECRET REASON WHY ESTIMATES ARE WRONG OFF THE CUFF ESTIMATES ! Q. Can you get this done by Friday? Answer: Looks do-able. I will try. What is the right question? What is the right way to answer it?
CONE OF UNCERTAINTY
¯_(ツ )_/¯ NOT KNOWING THE ANSWER IS OKAY
1. Ambiguous/Incomplete Requirements. No documented use-cases. 2. Using Assumptions/Gut-feeling. Instead of facts. 3. Large-Scale. No breakdown structure = Omitted activities. 4. Single-Point Estimates. Average ≠ Conditional. 5. Bad work processes Bad design, code  Bugs  Rework/Firefighting. 5 REASONS THE 6TH SECRET REASON WHY ESTIMATES ARE WRONGQ. Can you get this done by Friday? Answer: Looks do-able. I will try.
RIGHT QUESTION? / RIGHT WAY TO ANSWER IT? 1. Ambiguous/Incomplete Requirements. No documented use-cases. 2. Using Assumptions/Gut-feeling. Instead of facts. 3. Large-Scale. No breakdown structure = Omitted activities. 4. Single-Point Estimates. Average ≠ Conditional. 5. Bad work processes Bad design, code  Bugs  Rework/Firefighting. Q. Can you get this done by Friday? Answer: Looks do-able. I will try.
ESTIMATES THE USUAL SUSPECTS Reference: Software Estimation Demystifying the Black Art – Steve McConnell
PRECISION = UNINTENDED CONFIDENCE.It is10:30 am. You are on your way to meet a friend… You: “Will see you at the mall by 11 o’clock.” Sounds like will be there sometime before 11:05 VS. You: “Will meet you at the mall at 10:58am.” Sounds like will be there sometime before 10:59 The 2nd estimate was simply based on an ETA of 28mins shown by the navigation app at 10:30.
INTERESTING BITS FROM AROUND THE WORLD... Calculating Functions Points - A practical example…  stackoverflow.com/q/34473698/319204 “Function Points? Valuable, but too much effort to get right.” - Mike Cohn (Scrum Alliance)  stackoverflow.com/a/153458/319204 12 tips to better estimates  stackoverflow.com/a/35571856/319204 PMBOK – 5th Edition  A Guide to the Project Management Body of Knowledge – 5e

More Related Content

Effort Estimation - Chinmay VS - Scrum Bangalore 19th Meetup

  • 1. EFFORT ESTIMATION Write down the number of “squares” in the above image…
  • 2. 5 REASONS WHY ESTIMATES TEND TO BE WRONG 1. Ambiguous/Incomplete Requirements. No documented use-cases. 2. Using Assumptions/Gut-feeling. Instead of facts. 3. Large-Scale. No breakdown structure = Omitted activities. 4. Single-Point Estimates. Average ≠ Conditional. 5. Bad work processes Bad design, code  Bugs  Rework/Firefighting.
  • 3. THE 6TH SECRET REASON WHY ESTIMATES ARE WRONG OFF THE CUFF ESTIMATES ! Q. Can you get this done by Friday? Answer: Looks do-able. I will try. What is the right question? What is the right way to answer it?
  • 6. 1. Ambiguous/Incomplete Requirements. No documented use-cases. 2. Using Assumptions/Gut-feeling. Instead of facts. 3. Large-Scale. No breakdown structure = Omitted activities. 4. Single-Point Estimates. Average ≠ Conditional. 5. Bad work processes Bad design, code  Bugs  Rework/Firefighting. 5 REASONS THE 6TH SECRET REASON WHY ESTIMATES ARE WRONGQ. Can you get this done by Friday? Answer: Looks do-able. I will try.
  • 7. RIGHT QUESTION? / RIGHT WAY TO ANSWER IT? 1. Ambiguous/Incomplete Requirements. No documented use-cases. 2. Using Assumptions/Gut-feeling. Instead of facts. 3. Large-Scale. No breakdown structure = Omitted activities. 4. Single-Point Estimates. Average ≠ Conditional. 5. Bad work processes Bad design, code  Bugs  Rework/Firefighting. Q. Can you get this done by Friday? Answer: Looks do-able. I will try.
  • 9. PRECISION = UNINTENDED CONFIDENCE.It is10:30 am. You are on your way to meet a friend… You: “Will see you at the mall by 11 o’clock.” Sounds like will be there sometime before 11:05 VS. You: “Will meet you at the mall at 10:58am.” Sounds like will be there sometime before 10:59 The 2nd estimate was simply based on an ETA of 28mins shown by the navigation app at 10:30.
  • 10. INTERESTING BITS FROM AROUND THE WORLD... Calculating Functions Points - A practical example…  stackoverflow.com/q/34473698/319204 “Function Points? Valuable, but too much effort to get right.” - Mike Cohn (Scrum Alliance)  stackoverflow.com/a/153458/319204 12 tips to better estimates  stackoverflow.com/a/35571856/319204 PMBOK – 5th Edition  A Guide to the Project Management Body of Knowledge – 5e