This post continues my series of blog posts exploring the acronym METRICS to idenitfy what metrics could, or probably should be to ensure that you are getting maximum value from them. I have previously taken a look at Meaningful Metrics, Evolving Metrics and Trusted Metrics. In this post I consider ‘Repeatable Metrics’.
It is very rare that a ‘one-off’ metric is required. In the majority of cases, metrics will be produced on a regular basis allowing them to be compared over time to allow patterns, trends or changes from the norm to be identified and discussed. In order to ensure that metrics can be compared, repeatable processes need to be put in place in order to ensure that the metrics produced are consistent and to allow for comparisons to be made.
There are many electronic tools that allow for user stories and their associated acceptance criteria to be created. These tools also allow the progress of these user stories to be tracked through an Agile workflow – this will often be through the use of a Kanban board. Electronic tools will automatically capture the date and time that a user story transitions from one state to another. It is this meta-data that can then be used in order to generate metrics.
The most popular electronic tools will have a number of metrics built into them, with the option for more through additional plug-ins. Some or these metrics will be generated through the normal use of the tool, while others may require the additional of specific labels/versions and/or the creation of bespoke queries. The generated tables/charts will then be available on individual pages within the tool and/or they can be added to a dashboard where multiple metrics could be viewed, compared, combined and analysed.
Although the producers of tools will continue to update them in order to include the latest metrics, there will be times where it will be necessary to extract the data held in the tool for more manual manipulation. This extraction of raw data allows for more bespoke metrics to be created, with the ability to drill into the data further as interesting peaks, troughs or anomalies are identified.
Following the manual creation of a metric that cannot be produced within an electronic tool, there can be benefits from automating its production. Automating the production of a metric will ensure that it is consistently produced, reducing the risk or manual error, while at the same time reducing the time that it takes for the metric to be generated. It may also be possible to add conditional formatting and/or alerts to a metric that highlights where there is a variance that requires further investigation.
HEALTH WARNING!
Just because metrics can be repeatably produced, doesn’t mean that they should be. There is only value in the production of metrics when there is human intervention. Unless someone is looking at the metrics and interpreting what they are telling you, why bother? Furthermore, once the metrics are interpreted and areas of interest are found, there may need to be some form of action taken. If no action is taken, why was time spent in creating the metric?
My challenge to you is to identify if you have processes in place in order to consistently and repeatedly produce valuable metrics.