By Lynda Bourne
Have you ever experienced technical debt on a project? As the debt builds up, everything looks good from the outside. However, when the crunch comes and that debt has to be repaid, a major reversal in fortune can occur.
Technical debt refers to the costs of having to go back and resolve problems that arise because an earlier decision was made to take the easy route, instead of the best one. By taking the easy option, the team incurs a debt to the project that will have to be repaid later. While the concept comes from software development, this insidious effect can be seen across industries.
The Crossrail project in London offers a current, extreme example. In July 2018, it was reported that on-time and on-budget completion of the £14.8 billon rail project would occur in December 2018. By August 2018, completion had slipped by a year. Currently, the delay extends to the end of 2020, with a cost overrun of 20 percent.
What’s the main driver of this delay and associated costs?
It appears to be decisions made to ignore problems in the signaling system development. According to Construction Manager magazine, while giving evidence to a government inquiry, Crossrail’s new chief executive Simon Wright said, “We were testing on incomplete systems. Productivity was under stress, but we fought hard to maintain the schedule and thought all along that we could find a solution to bring it back, just like we have done on countless other problems that occurred during the construction program.”
This is a classic example of management decisions building up a technical debt.
In 2015, The Independent newspaper reported that rail experts and engineers were having difficulty creating interfaces for the signaling systems. At the same inquiry, Crossrail’s new chairman Terry Morgan said “problems that emerged were mostly due to difficulties with developing software to allow Crossrail trains to travel safely at speed through three separate signalling systems,” according to Construction Manager magazine. The problem was identified in 2015 and hadn’t been resolved by 2019, despite time and money wasted testing incomplete systems. In fact, the irrelevant testing probably added to the delay and costs by distracting people from the real challenge.
Fixing the problem properly the first time would surely have caused a delay and cost blowout between 2016 and 2018. But in all likelihood, the costs would have been lower, the delay would have been shorter, and the current furor surrounding the project would have been minimized.
The problem with technical debt is that often, the people who need to know about a problem aren’t informed. We will never know what the chair and CEO of Crossrail (both sacked) really knew in the 2016 to 2018 period, or what their senior managers knew about the build-up of the technical debt in the Crossrail signalling systems. But the problem could have been avoided, or at least minimized, if the technical debt had been acknowledged. If people are unaware of technical debt, then they’ll be more likely to identify paths that will result in it being created.
To avoid this lack of insight, everyone in the project group, especially team members, must be in a position to offer insight into technical debt.
The project manager can then choose to act, or not. Aware teams bring up the subject of technical debt in planning meetings, and they keep focused on it. Aware managers pose questions such as, “If this proposed shortcut is the right choice, what is there to gain, and what are the challenges and future implications?”
As with financial debt, there are times when going into debt can be beneficial, but only if you can pay back the accrued debt and interest at the right time.
How much technical debt is your project running? Please share your experiences below.