(2016-09-22) Corey Haines Methodology Per Business

Corey Haines gave a talk at NCSA Sports. Some crude notes (very possibly mis-summarizing)...

The goal is Agility not Agile.

Each business you work at needs its own process. (One Methodology Per Project)

At one place (is this always true? always true under high-growth?), he concluded that probable value-per-feature was exponential (Power Law) (meaning the biggest wins would be multiples-bigger than the medium wins, not just a bit different), and this swamped the difference in cost-per-feature. Therefore

  • you didn't need much precision in Business Value, just picking biggest winners (cf BottleNeck)
    • the key challenges with this model I've found:
      • not having any big wins - this is a sign you need to focus on Idea Generation but many are resistant
      • having all your big wins be really too big (in cost) (or, more to-the-point, un-chunkable), so you get 0 value until you spend 20 man-years which you don't have, so you have to go back to Idea Generation
  • you could ignore cost (Estimating) (at this level)
    • cf "too big to do" point above

Edited: |

blog comments powered by Disqus