Mythbusting Software Estimation

schedule Mar 27th 03:15 PM - 04:00 PM place Grand Ballroom 1

Estimating software projects has proven to be particularly challenging.  Over-running schedules happens frequently in our industry.  As a result software estimation is often viewed as a black art.  In this session Todd will look into some of the reasons for these challenges by exploring a number of myths of software estimation and then setting out to validate or bust these myths.

Some of the myths that will be explored include:

  • Historical Estimation Accuracy
  • Relative Estimation
  • The Cone of Uncertainty
  • Velocity
  • Scope Creep
  • Estimation Tools
  • Wisdom of Crowds
1 favorite thumb_down thumb_up 2 comments visibility_off  Remove from Watchlist visibility  Add to Watchlist

Outline/structure of the Session

This talk is a minor update of the presentation I gave at Agile India 2014.

I start by having everyone estimate the number of jelly beans in a jar.  I come back to this later.  

I then do a bit of test first by having the audience predict whether a 10 myths about software estimation are true or not.  

I then go through each of the 10 myths and present research results to validate or disprove the myth.  



Learning Outcome

Learn some of the misinformation about estimation that has permeated our world. Also learn a bit about what you can do about it.

Target Audience


schedule Submitted 3 years ago

Comments Subscribe to Comments

comment Comment on this Submission
  • Joel Tosi
    By Joel Tosi  ~  3 years ago
    reply Reply

    Hi Todd,

        Just curious - is there enough delta from last year in the event someone saw that presentation?



    • Todd Little
      By Todd Little  ~  3 years ago
      reply Reply

      It's improved, but not substantially different.  AI2014 was the first time that I ran this as a standalone talk (previously I used it as part of a tutorial).  Since then I've given it quite a few times and I've polished it a bit along the way.  

      I wasn't sure how much repeat attendance there would be.  I had a short email chat with Naresh and I will likely submit a fresh session or two that are different from the 3 I gave last year.