Improvisation is at the heart of Jazz music, as we already saw in note #1. In any performance, each musician will probably get the opportunity to improvise one or more solos, while the rest of the band provides an appropriate backing. Being able to improvise is key — and even if you follow a structured methodology, there is always place for improvisation… but follow Art Tatum’s advice: you’ll have to be prepared in order to improvise!

Let’s consider an activity that’s common to all methodologies: how do you estimate your pending work? No matter what you do, this will always be an error prone task… and being fixated on a single way of doing things will probably be a point of pain in your process. So, have you considered improvising, and changing the way you estimate? For example, if you use story points, have you thought about dropping them altogether? Let’s consider some possible alternatives, so you can give them a try to see what fits you best.

Probably the best known method for story evaluation implies Story Points, with several variants, such as “power of two” (0, 1, 2, 4, 8, 16…) and “Fibonacci” (0, 1, 2, 3, 5, 8, 13…). However, this level of granularity may become complicated, and there always exists a tendency to “point inflation”, which causes tasks with similar complexity to be rated higher with time. Also, points may become an objective by itself (“We must do 120 points in this sprint…”), a way to attempt comparing teams (“Why does your team do fewer points than that other team?”) or even a contract issue; all negative things.

Some ideas to get you started –and if you google around, you’ll find these and more– may be:

  • Non-numeric scales, such as T-shirt sizes (XS, S, M, L, XL) or more whimsically, as a web site suggested, dog breeds (from “Chihuahua” to “Great Dane”). Some people suggest using only a few categories, plus an “Undefined/Don’t know yet” category for those stories that cannot be estimated at the moment.
  • Work by comparison: take a few stories, rank them along a line (to the left, the simpler ones, and to the right, the more complex ones), and then rank the rest of the stories by comparing them to the ones already on the line. Don’t aim to very fine distinctions; instead of a continuum, go for a few “buckets” of similarly complex stories.
  • Trim to fit”: decide a maximum size (on whatever terms you usually apply) for any item, and then discuss each story to determine if it’s already that size, below it, or above it. In the latter case, break down the story into sub-stories, until all your too-big stories have been pared down to lower levels of complexity. (This is really an inversion of concept: instead of assigning a measurement to each task, make each task fit a single measurement!) Of course, don’t just break stories in haphazard fashion, just to make them fit.

As you develop your own version of whichever methodology you use, take into account the possibility of adding variations to any step, or to change your way of doing things. Of course, this isn’t an excuse to change things without analysis, or to alter your ways at a whim without consistency — but with appropriate care and consideration, you may perfectly well improvise your way to your own methodology!

P.S.: Art Tatum was considered legally blind, but this didn’t impact his virtuosity at the piano: check out “Tea for Two”, a recording made in 1933, and “All the Things You Are”, from twenty years later, to hear his incredible technique.  

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>