READING NOTE · UPDATED 2026
Examine the coordination cost of size
Team and batch size affect more than the amount of work in a plan. Additional interfaces can create waiting, decision overhead and integration risk. The short original article is a prompt for examining those effects. Do not assume that splitting a team or batch will improve delivery without checking the dependencies that remain.
A practical next step
- Map the interfaces involved in one delivered outcome.
- Identify one costly coordination delay.
- Trial a smaller batch or clearer interface and inspect quality as well as time.
This 2026 reading note is written by the Durnall editorial team. Richard Durnall’s original text follows below.
Original publication: .
Original WordPress feed timestamp, UTC; the old theme’s local date may differ. Check the date source ↗
The original words below retain their period and attribution. The reading note above is the current editorial addition.
Source capture · 2009-11-02 ↗I’m spending a lot of time at the moment working with governance groups, steering committee’s and PMO’s to introduce Lean ideas. There’s a common challenge that I keep coming across. First up, size matters. I was going through the 2009 Standish CHAOS report today to gather some data for a presentation. Here’s what they have to say about project size and it’s relationship to success…
- 61% of all successful projects are under US$750,000.
- 80% of all successful projects are under US$3M.
- Projects over US$10M have less than a 2% chance of success.
So, we need to keep things small, breaking projects apart where possible to deliver in small chunks to increase our chances of success.
The tension is that governance groups often prefer to manage one $10M project than lot’s of small ones. This is generally a pattern in large corporates rather than on-line product companies, who have understood this for years, often moving away from project models and towards product models where key products are slowly evolved over time, using the returns they generate to reinvest.
Keep it small. Keep it simple. Software by numbers.