Tuesday, April 3, 2018

Scale down, or "too small to succeed?"

What about scaling down?
Most of the posts I read are about scaling up to ever larger projects, but what about scaling down? What if you are doing bug fixes and small scale projects with just one or two people? Is a methodology of any help, and if you're working with software, can Agile be helpful scaled down?

To the first point, my answer is: Sure! A methodology -- which is just a repeated way of stringing practices together -- can help. Why invent the 'string' (methodology) each time you do something? Just follow the same practices in the same sequences each time. Thus, you can have a methodology for just driving your car... perhaps the ultimate team-of-one.

Small scale Agile
Re agile and small scale: Really, the agile manifesto does not inherently break down at small scale; nor do the agile principles. It's actually easier to go smaller than to go larger.

But, not so fast! What about:...
  • Teams and all the stuff about team work if you are only one or two people?
  • Pair programming?
  • Redundancy and multi-functionalism?
  • Collaboration and communication by osmosis with others?
Admittedly, you give up a few things in a team-of-one situation -- perhaps pair programming and redundancy -- but there's no reason to give up the other ideas of agile:
  • close coordination and collaboration with the customer/user
  • a focus on satisfying expectation over satisfying specification
  • quick turn around of usable product
  • personal commitment and accountability
  • collaboration with peers and SMEs on a frequent and informal basis
  • lean thinking
  • Kanban progression and accountability
Got redundancy?
The hardest thing to deal with in a too-small team is lack of redundancy -- or not being anti-fragile -- and lack of enough skill and experience to work over a large domain. It's pretty obvious that one person team has a single point failure: when you're away, the team is down. Such a failure mode may not tolerable to others; such a failure mode is obviously fragile, unable to absorb large shock without failing.

And, one person's experience only extends so far... thus limiting the domain and extent of possible solutions (leading, sometimes, to: "if all you have is a hammer, then everything is a nail" misuses and abuses of whatever your skill set is. I once had a person who was a loner and also an expert with MSAccess... no matter the problem, he seemed to solve it with Access!)

Management challenges
As managers we are responsible for making it work. So in the situation of really small teams, there can still be, and we should insist upon:
  • an opening narrative
  • a conversation about requirements and approach (to include architecture)
  • peer review by other experts
  • code inspections and commitment to standards
  • rotation of assignments among context to drive broadening
  • reflection and retrospective analysis

Read in the library at Square Peg Consulting about these books I've written
Buy them at any online book retailer!
Read my contribution to the Flashblog