Home > Point2 - Process and People > Value as an Agile Metric

Value as an Agile Metric

November 5, 2009

At Agile 2009 I attended Brandon Carlson’s talk entitled “Value or Velocity?“.  Velocity is often thought of as a measure of speed or productivity, but if a team can deliver x story points’ worth of the wrong product, and another team can deliver the same number of story points’ worth of an amazing product that everyone wants, but the two teams are said to have been equally productive based solely on their Velocity numbers.

In other words, getting more done is meaningless unless you know you’re getting the right things done.  To make sure you’re doing the right things, Brandon suggested assigning value points to epics.

The simplest method is Value Poker, kind of like Planning Poker.  The Product Owner assigns a value to each epic relative to other epics in the sprint backlog, with the smallest amount of value being a “1”, just like the story that requires the smallest amount of effort is a “1” in Planning Poker.

Another, probably more accurate method is Attribute-Driven value assignment.  In this case, all product owners and stakeholders meet and assign values to epics together.  Each stakeholder will know the most about one or a small handful of attributes.  Some possible attributes include: market differentiation, strategic alignment, usability, system stability, sales commitment.  Draw up a chart with each attribute as a column, and each epic as a row.  The stakeholders for each attribute assign a value of 0-3 in each row for how valuable the epic is for that attribute.  Total up all the values for a row and then you have numbers you can sort on, and that can become your prioritized list,

To a stakeholder it is much more meaningful to hear “this month the team delivered 31 value points” as opposed to “this month the team delivered 31 story points”, because at the end of the day it’s delivered value that customers pay for, not complexity of stories.

By: Tefon Obchansky

Advertisements
  1. November 10, 2009 at 2:39 PM

    Value points is a good idea. I’ve seen some others promoting the same thing.

    This is an orthogonal idea to velocity points. Velocity points are about getting predictable deliveries. Value points aboiut delivering value to the market.

    There could be some very inexpensive, but valuable stories. Its a good idea to do those first. Then there is also a good change that some are very expensive but not very valuable. Good idea to decide not to do them, and maybe look for cheaper ways to solve the business problem.

    A drop off in value points should signal a PO that it is time to find some new valuable market needs. This product might be done, time for a new one.

    • Tefon Obchansky
      November 12, 2009 at 9:51 AM

      That’s an excellent point, and coincidentally the topic of my upcoming post tomorrow. 🙂

  1. November 13, 2009 at 2:54 PM
Comments are closed.
%d bloggers like this: