Continuously improving!

Assessments and continuous process improvement are intertwined. Assessments being both a source of ideas and a tool to validate change and other experiments.  For example, the Shewhart Cycle popularized by W Edwards Deming, breaks the cycle of process Improvement into four segments. The steps in the Shewhart Cycle are “plan, do, check, act.” The use of assessments easily fits into the ‘check’ step, and can also be used to feed data to planning, doing, and acting.  The use of assessment as a tool to support continuous process improvement requires several considerations. (more…)

Develop a plan of attack

There are times when just letting go and going with the flow is a great idea.  I plan to be spontaneous at least twice before I die. Agile assessments are not one of those events that work best without planning.  An even broader rule is that any form of assessment requires a framework and a plan. A framework and a plan are required if the results are to be reliable and repeatable. The type of assessments and the reason for the assessment will go a long way to determining what needs to be looked at in a general way, however, the assessment plan needs to get down to the nitty-gritty.  The assessment plan will need to explicitly determine what will be looked at including behaviors, decision-making capability, ceremonies and deliverables, and then communicate those decisions to the assessment’s stakeholders. The areas covered in an solid assessment plan include: (more…)

You Are Here?

Assessments in agile come under a wide variety of names: appraisals, health checks, audit or even assessments. These terms are commonly conflated.  Assessments are a tool to prove a point. There are many approaches to assessing agile in a team or organization ranging from self-assessment questionnaires to formal observer-led appraisals. What gets assessed and the approach an organization chooses depends on the point of the assessment. All assessments create a baseline, a line in the sand from which to measure change.  At the same time, an assessment is a benchmark. Benchmarks are a comparison against a standard (real or implied). For example, an organization could use the principles in the Agile Manifesto or the framework in the Scrum guide as a standard to compare their behavior against to generate a benchmark. Put very succinctly, baseline defines where an organization (or team) is at a point in time, while a benchmark is a comparison to a standard. Assessments of any type, whether to generate a baseline, a benchmark or both, require time and effort which might be better spent creating a product, unless there is a good reason to do an assessment. There are three macro reasons why an organization might assess the agile journey that can generate value: (more…)

Five Dysfunctions of a Team

Five Dysfunctions of a Team

Happy Holidays!  Today we have a short version of Re-Read Saturday.  This week’s re-read of The Five Dysfunctions of a Team by Patrick Lencioni (Jossey-Bass, Copyright 2002, 33rd printing), we address begin the major section titled, The Model.  The Model culminates Lencioni’s book and includes four chapters. Today we will re-read the sections titled, An Overview of the Model and Team Assessment.  A short and sweet entry and then I am off for potent eggnog and just maybe a seat at the new Star Wars Movie.  Two more weeks on this book. (more…)

Assessment Requires Observing The Process!

Assessment Requires Observing The Process!

Shakespeare wrote “to be or not to be” but when deciding whether a project requires intervention, we might paraphrase the quote a bit to say “to act or not to act”. In each case we need to decide whether a project warrants or needs intervention. Not all projects we viewed as troubled are in bad enough shape to require an intervention and by intervening we might waste resources or a learning opportunity. For example, does a project that is two weeks behind schedule after a year and with an estimated duration of an additional year of duration warrant an intervention? What if the implementation date has been committed to in the marketplace?  All of the indicators of a project under duress need to be interpreted as part of entire organization and at the same time through the lens of known set of tolerances.

Assessments can be done using a variety of methods.  Each method has its own set of strengths and weaknesses.  The overall strength in using a method (any method) is that the results become less about opinions and feelings and more about facts.  There are three basic forms of assessment: model, process and quantitative.

Three Types of Assesments

 

Quantitative assessments are easily the most common of the three approaches used to identify troubled projects.  This method is the most common because a majority of projects have a known budget, timeline and acceptable level of quality (even if not stated) that are reported against on status reports.  The assessment process is fairly simple, has the project spent (or planning to spending) more than the budget, if yes, then it is in trouble.  The same comparison can be made to the promised date, or the number of defects the project has found and logged into the backlog. The bigger the difference the bigger the hole the project will find itself in. What the quantitative assessment does not answer is why the project is in trouble and what to do about it.

Model based assessments use industry standard frameworks such as the CMMItm to look for process gaps. The process gaps can be used to explain why the project is having problems and suggest areas that need to be implemented or tuned to help the project recover.  Model based assessment are generally very formal in nature which can require substantial effort and be quite invasive.

Process assessments look as the qualitative attributes of the targeted process. Targeting generally done based on interviews with project leaders and stakeholders.  These interviews and later qualitative process appraisals require skilled interviewers that are experienced in project rescues.  These methods are generally less invasive than formal model based appraisals and be deployed faster, however, they can be skewed by the biases of the interviewers. I suggest combining qualitative and quantitative assessments to help minimize potential biases.

Once the assessment has been completed, the final step is to determine an overall direction for the course of action. We suggest that if an intervention is required that there are only two possible strategies. The first is to reset the project (build on what has occurred using new techniques and team structures) and the second is to blow it up and start over. Again a set of criteria should be leveraged to lessen the passion around the decision process.  The following table shows an example of a set of decision criteria:

Criteria

Action
Is there a coherent vision of the projects goals? If Yes, then reset
Does an external estimate to complete indicate that humans can actually deliver the project? If Yes, then reset
Can the organization afford what is required to deliver the project?

If Yes then reset

If the answer to any of the criteria is “No” then stop the project. Every organization will have develop specific criteria that meets the organization’s culture. However, recognize that trying to turn around a project that is either not feasible or is not pursuing a coherent goal will generally lead to throwing good money after bad which is why assessing the project using a process will help an organization to make a less emotional decision.