A team’s velocity is their ability to deliver value over time. Managers often use velocity as a key performance indicator (KPI) to assess team performance.
Velocity measurements show whether a team is consistent and if they are improving over time.
The problem with using it as a KPI, however, is that velocity doesn’t tell you what is holding your team back or why.
Your team could be burdened with significant internal overhead, yet still showing a stable velocity. To truly tell how a team is doing, we need a better metric.
Velocity debt is anything that prevents teams from spending their time creating value for customers.
Velocity = Capacity - Velocity Debt Cost
Velocity debt It is a silent killer, sucking up time and energy on activities that may not show up on sprint reports.
Teams accumulate velocity debt as they grow, sometimes starving value-adding work entirely. It is not uncommon for teams to waste over 50% of their time on velocity debt.
As its name indicates, velocity debt slows down your velocity, but it also impacts team morale.
Engineers want to spend their time building things, not wading through bureaucratic stumbling-blocks. Teams laden with velocity debt have lower happiness, higher attrition, and drive away talent.
On the other hand, nothing makes people more productive than job satisfaction. If you remove the impediments and roadblocks created by velocity debt, your team will take care of the rest.
Anyone who works with software has heard the term technical debt, which is a deficiency that makes software more difficult to modify and maintain.
While technical debt is one component of velocity debt that gets a lot of attention, there are others that can have a big impact and are important to consider. Velocity debt also includes things like inefficient meetings, interruptions from external requests, merge conflicts, and more.
People sometimes refer to other persistent impediments as “process debt”, “managerial debt”, etc. Velocity debt encompasses all of these concepts together.
There are a lot of things that can interfere with software development, but they tend to fall in the following high-level categories based on where people spend time:
There are a few different approaches for getting a handle on how much your team suffers from velocity debt.
One method is just to ask each team member to approximate the percentage of time they spent on value-adding development vs. other activities each sprint. The nice thing about this approach is that it is easy to get started.
A downside of surveying is that it requires manual effort. The more serious problem, however, is that people have a strong availability bias, which causes them to remember the things that felt frustrating, while not considering velocity debt that has become a normal part of how things are done (e.g., we always have to spend a day fixing merge conflicts at the end of each sprint).
Another way to measure velocity debt is to ask people to log all of their time with a time tracking system. This eliminates availability bias because you are not asking people to remember anything.
One issue with time tracking is that the manual effort required is more significant than for surveys, which ironically creates more velocity debt.
More importantly though, time tracking is still limited by human judgment and has limitations on its granularity. Are people going to stop the clock on their task and start it on “Slack interruption” to answer a quick question? They probably won’t, which means a lot of velocity debt will continue to fly under the radar.
The most accurate, granular information about velocity debt comes from the artifacts people leave behind in version control, ticketing, and communication systems as they build software.
For large organizations with vast data engineering and analysis resources, it might make sense to gather all this information into a data warehouse to help identify activity that stifles team velocity.
The obvious downside here is that ingesting and analyzing this data is a complex undertaking, and not many organizations actually have vast data engineering and analysis resources.
A velocity debt management system like minware handles all the complexities of pulling data from various APIs, analyzing timing artifacts, and enriching events to provide a comprehensive picture of where people spent their time, along with how much was taken away by non-value-adding activities.
This approach provides the power of in-house data analysis at a fraction of the cost.
Here you can see how minware computes overall time ratios and lets you drill down into individual velocity debt activities. This example shows an 8-person team that spent only 37% of their time on completed tasks in a particular 2-week sprint.
Once you have measured the level of effort that goes toward paying “interest,” you can model the monetary amount of velocity debt – that is, how much you should be willing to pay to eliminate it. The formula is as follows:
Velocity Debt Amount (V) = Annual Team Salary (S) × Opportunity Cost Ratio (O) × % Velocity Debt Overhead (D) / Discount Rate (R)
Here is an explanation of the terms in the formula:
To understand how the formula works, imagine that you took out a $1m loan with an interest rate of 15%, which costs $150,000 annually. You then turn around and pay this $1m to consultants who magically fix some velocity debt that is taking up 10% of your team’s time each sprint.
The team is paid $500,000 annually, so you just freed up time that was costing you $50,000. Given your organization’s estimated opportunity cost of 3x for engineers, you expect this to generate $150,000 annually in value for the company, thus breaking even:
$1,000,000 (V) = $500,000 (S) × 3 (O) × 10% (D) / 15% (R)
As you can see, the value and cost you should be willing to pay to fix modest amounts of velocity debt can be immense. This model also doesn’t include the human cost, so if you factor in the risk of losing your highest performers, the value can be even greater.
While many teams assess performance by tracking velocity, few are able to say how much velocity they lose to non-value adding activities.
By analyzing the cost of velocity debt, teams can identify and prioritize changes that will improve their speed, predictability, and overall happiness.
Velocity debt tracking is easy to get started with surveys, and teams interested in diving deeper can quickly ramp up to comprehensive velocity debt management with minware.
The cost of not managing velocity debt can be quite high and manifest in unexpected ways like team attrition. Product and engineering leaders can get ahead by forming a velocity debt management strategy to help improve their team’s speed, predictability, and overall happiness.