The price of speed: Technical debt (II of II)

Dealing with technical debt

Continuing on the previous post on technical debt, we will talk about approaches to tackle tech debt in this post.?

Some companies tackle technical debt like some people tackle credit card debt. They pay a credit card bill with another credit card. You might buy some time, but only for a short time before the debt ultimately must get paid. Fortunately, your enterprise can remediate its technical debt. It’s not an instant fix, however, but requires patience, investment and dexterity, like untangling a knot in a necklace.

A worthwhile journey from red to black

?Any modernization journey needs to measure, assess, prioritize and remediate technical debt. ?

Many of the clients I worked with had this dilemma of how to measure technical debt. CFOs expect this to be a dollar number that can be counted as a negative number on the balance sheet. And CIOs would love to give this number to the CFO to get more investment. In reality, there is often no single numeric value but a composite of multiple factors that companies can choose based on their goals. For e.g., cost of maintaining system inefficiencies, or the opportunity cost for new developments. Some companies express this as a percentage of their IT budget (Costs to resolve tech debt / Total IT budget). Regardless of whatever measure you use, align it with your broader objectives and make it transparent within tech org and to business.

Best in class companies prevent tech debt accumulation by regularly assessing and prioritizing which areas to address rather than waiting for issues to compound and be fixed in ad-hoc manner. Assessment approach can be bottom-up (i.e., engineering teams surfacing issues) or top-down (driven by targeted business initiatives, e.g., new product build in one business unit). Prioritization on which technical debt to remediate is critical and should always be linked to the broader business goals. At one of the clients, the lens was the impact to the end customer. Every tech debt item was prioritized based on how it can improve the experience of end customer.

?Final step of remediation isn’t easy, but doable – and a worthwhile investment. Some companies commit a fixed amount of effort within each sprint to address tech debt stories. 10-20% effort is common but I have seen one client, who was lagging in modernization, commit 40% of effort. More important than the commitment, is the constant dialog between tech engineers and business (Product Owner) to create a holistic picture and make decisions based on trade-offs. It is akin to the proverbial ‘sharpening the saw’ vs ‘cutting more trees’ trade-off.

One of our banking clients took a creative approach, levying a “tax” for any development made on legacy systems. The tax was levied to the business units using those systems and created the pull from business side to address tech debt. Another client reduced the SLAs on applications which had heavy tech debt. Best practice is to deploy tools like CAST or SonarQube to create transparency on the remediation progress and incorporate tech debt resolution in the broader demand and portfolio management.

Three archetypes?

There are roughly three archetypes of modernization journeys to remediate tech debt:

?●?????Tech focused: With this approach, you’ll focus on long-neglected tech landscape and platforms with an eye toward bringing them into the present.

An example: A fortune 500 company was lagging on modernization after years of constraints. The new CIO decided to take a bold approach and aggressively invested in remediating debt based on tech landscape. Started with the most urgent tech issues and went into more proactive tackling of debt. Key criteria was the impact it would have on the IT teams to deliver. Eventually they created a true platform architecture which increased the speed of delivery tremendously.

●?????Business-outcome focused: In this approach, you focus on tech landscape within a business domain with the goal of improving the specific business outcome.

A retailer, for instance, wanted to expand its e-commerce offerings to stay competitive, but was stuck with legacy systems. We helped this client map all its systems associated with online commerce, and identify the technical debt blocks. The remediation was prioritized along with the actual e-commerce changes and made a significant impact to the business outcomes. The speed of releases into the market was increased from quarterly to monthly, increasing its competitiveness.

●?????Maintenance focused: Organizations that continually modernize their tech landscape may address their technical debt in “maintenance” mode, recognizing that debt happens and striving to keep it at a healthy level. To do this, a mature enterprise puts in standards and guardrails to proactively avoid tech debt.

One of our tech clients, already has a modern technological infrastructure, and anticipates a certain amount of technical debt. Its engineers maintain an active technical debt backlog and remediate it in varying ways, dedicating a certain percent of their time to the task or, when the occasion calls for it, finding and correcting errors in daylong hackathons. They choose which debt to address by consulting the risk/efficiency/scaling limits framework devised by leadership as well as the time allotted to them for maintenance.

Get started now

Finding and fixing all the errors and workarounds in your complex maze of systems and software code can seem like an enormous and overwhelming task. But it can be done.

To begin, create a backlog of technical debt, setup tools like CAST to quantify tech debt and find one small business use case where the value of addressing tech debt can be demonstrated. Learn from this experience and determine broader remediation and change management approaches to be deployed.

Lack of?Technical debt can be a huge competitive advantage – when your organization wants to make an acquisition, for instance. Waiting to bring your tech debt down to healthy levels might cause you to be too late to take advantage of a growth opportunity – but it’s never too early to start.?