When the Shortcuts Come Due — Technical Debt as the Strategic Debt of the Pandemic Sprint
Debt you have consciously chosen to carry is a position. Debt you are carrying because no one looked is an accident waiting for a strategy to expose it.
Executive Summary
Eighteen months ago the profession discovered it could move at a speed it had spent a decade insisting was impossible. Systems that had sat on three-year roadmaps went live in a fortnight. The organisations that survived the first shock of the pandemic did so on the back of that speed, and they were right to. What is less often said is that almost none of that speed was free. It was borrowed — against future engineering time, against architectural coherence, against the patience of the people who now maintain what was built in a hurry.
This essay argues that the technical debt accumulated during the pandemic sprint is best understood not as an engineering nuisance to be tidied away in a quiet quarter, but as strategic debt: a claim on the organisation’s future freedom of movement. Ordinary technical debt slows a team down. Strategic debt narrows what the enterprise is able to become. The distinction matters because the two are paid down in completely different ways, by completely different people, and the reckoning now arriving across the sector is a reckoning of the second kind. The organisations that will handle it well are those that stop treating the debt as a backlog and start treating it as a portfolio decision — one made in daylight, with the cost of the borrowing finally written down where leadership can see it.
The Green Report and the Groaning Platform
Picture a monthly delivery review, sometime this summer. The status report is green. The digital onboarding journey that was stood up in eleven days in the spring of 2020 — branch closed on the Friday, customers self-serving by the following Tuesday — is still live, still taking volume, still, on the dashboard, a success. Everyone in the room knows it was one of the finest pieces of work the organisation has ever done under pressure.
What the dashboard does not show is the team of five who spend the first three working days of every month reconciling that journey against a core system it was never properly integrated with. It does not show the manual export, the spreadsheet that has become load-bearing, the one contractor who understands the mapping and whose day rate has quietly become a strategic risk. It does not show that the journey cannot be extended to a second product without redoing the integration properly, and that “properly” is now a nine-month piece of work rather than the fortnight the original took.
The report is green because the thing works. The platform is groaning because the thing works only just, and only at a cost that never appears on the same page as the success. This is the characteristic shape of the pandemic’s legacy: outcomes that were real and compromises that were also real, sitting in separate ledgers, waiting to be reconciled.
The pandemic did not teach us that we could move faster. It taught us that we had always been able to, and had been choosing not to. The debt is the price of that lesson — and it is a lesson worth having paid for.
What We Actually Bought When We Bought Speed
It is worth being precise about the transaction, because the loose talk of “cutting corners” obscures what really happened. In the spring of 2020, organisations made a series of entirely rational exchanges. They traded architectural generality for immediate fit: a service built to do one thing for one channel, rather than a platform built to do many things for many. They traded documentation for delivery: the knowledge lived in three people’s heads because there was no time to write it down and no expectation that those three people would ever leave. They traded automated testing for manual confidence: someone checked it worked, shipped it, and moved on. They traded integration for interface: a screen that a human copied numbers out of, rather than a contract two systems honoured between themselves.
Every one of those trades was correct at the moment it was made. That is the part the tidy-minded find hard to accept. The debt was not incurred through incompetence or laziness; it was incurred through judgement, under conditions where the alternative to fast-and-flawed was not slow-and-perfect but nothing at all. An organisation that had insisted on doing it properly in March 2020 would in many cases simply have failed to serve its customers, and no amount of architectural elegance would have redeemed that.
But a correct trade is still a trade. Something was given up, and the thing given up has a habit of sending an invoice. The invoice is arriving now.
Why “Debt” Is the Right Word — and Where the Metaphor Misleads
The debt metaphor has been with the profession for thirty years, and it has earned its place because it captures two things the alternatives miss. It captures that the borrowing can be rational — you take on debt to seize an opportunity you could not otherwise afford, and that is often the wise move. And it captures that the borrowing compounds — the interest is the extra time every future change takes because the foundation is not clean, and that interest accrues silently, month after month, whether or not anyone is looking at it.
Where the metaphor misleads is in its implication of a fixed principal. Financial debt is a known number. You can look it up. Technical debt is not a number anyone can look up, and the pretence that it is has done real harm. I have watched organisations commission a “technical debt register” as though the debt were a fixed sum awaiting a repayment schedule, and then treat the register as the problem solved. It is not the problem solved. It is a photograph of a moving object. The point of the exercise was never the list; it was the conversation the list was supposed to start, about which of these compromises actually constrains the organisation and which are merely untidy.
That first-hand pattern is worth paying off concretely, because it is where most debt programmes go wrong. The register that helped listed everything by engineering discomfort — this code is ugly, this library is out of date, this service has no tests. The register that mattered would have listed things by strategic consequence: this compromise means we cannot launch in a second market; this one means our month-end takes three days and a contractor; this one means a single person’s departure would stop a revenue line. The first register produces a backlog no one will ever fund. The second produces a small number of items a board will fund tomorrow, because they are phrased in the language of what the organisation can and cannot do.
The Strategic Layer: Debt That Shapes What You Can Become
Here is the distinction on which the whole argument turns. Most technical debt is tactical: it slows a particular team working on a particular thing. It is real, it is worth managing, and it is, in the end, a productivity tax. You pay a little more for every change in that corner of the estate. Annoying; survivable.
Strategic debt is different in kind, not degree. Strategic debt is compromise that has migrated up the stack until it constrains not the team but the enterprise — what markets it can enter, what products it can offer, how fast it can respond to a competitor, whether it can pursue a given strategy at all. The onboarding journey that cannot be extended to a second product is not a team’s problem. It is a strategy’s problem, because “launch a second product through the same journey” was on the plan, and the debt has quietly removed it from the plan without anyone deciding to remove it.
“Tactical debt makes your teams slower. Strategic debt makes your options disappear — and it does so silently, so that the first anyone hears of it is when a strategy meets a foundation that cannot bear it.”
This is why the pandemic’s debt deserves the adjective strategic in a way that ordinary accumulated cruft does not. It was incurred at the load-bearing points of the business — customer acquisition, order fulfilment, payments, the core flows — because those were the flows that could not wait. The compromises did not settle in some backwater. They settled in the arteries. And debt in the arteries does not merely slow you down; it dictates the shape of what you can do next.
The Case for Leaving It Alone
There is a serious argument against everything I have just said, and it deserves to be put at its strongest rather than waved away, because a great many capable people hold it.
The argument runs like this. Debt that never comes due costs nothing to service. A great deal of what we grandly call technical debt sits in parts of the estate that will never change again — the compromise is real but permanently dormant, and paying it down is pure waste, gold-plating a door no one will ever open. Worse, “paying down debt” is the most reliable way ever devised for engineers to spend a year rewriting something that already worked, in pursuit of an elegance the customer will never see and the market will never reward. The pandemic proved that shipped-and-flawed beats elegant-and-late every single time. Why would we now unlearn the one genuinely valuable lesson of the last eighteen months and retreat into the very perfectionism that the crisis exposed as a luxury? Leave the debt where it lies. Ship the next thing. Let the dormant compromises sleep.
I want to concede how much of this is right, because the concession is the point. Most technical debt should be left alone. A blanket remediation programme is usually waste. The instinct to rewrite for elegance is one of the profession’s most expensive vices, and the pandemic did expose over-engineering as a habit dressed up as a discipline. An argument that ignored all of this would be worthless.
…and Why It Doesn’t Hold
But the argument proves less than it claims, and it fails at one specific joint: it assumes we can tell dormant debt from strategic debt without looking. We cannot. That is the whole difficulty. The compromise in the onboarding journey looked dormant too, right up until the moment the strategy called for a second product and discovered the door was welded shut.
The “leave it alone” case is really an argument against indiscriminate remediation, and on that narrow ground it wins completely. What it cannot license is ignorance — the refusal to do the strategic triage that tells you which of your dormant-looking compromises is in fact sitting on the critical path of next year’s plan. The alternative to a blanket rewrite is not doing nothing. It is doing the diagnosis: mapping the strategy the organisation intends to pursue onto the foundation it actually has, and finding the small number of places where the two collide. That map is cheap. What it prevents is expensive.
So both instincts are right about different things. The perfectionist is wrong to want to pay everything down. The complacent are wrong to think that because most debt is harmless, all of it can be ignored. The discipline the moment requires is neither remediation nor neglect. It is selection — and selection demands that someone first make the debt visible in strategic terms.
Paying It Down Without Stopping the Line
The practical difficulty is that no organisation can halt to renovate. The line has to keep moving; the next quarter’s outcomes still have to land. So the reckoning, done well, is not a programme with a start and an end. It is a change in how the ordinary work is governed. A few disciplines separate the organisations getting this right from the ones merely worrying about it.
- Write the interest down where leadership can see it. The single most useful act is to stop hiding the cost of the compromise on a different page from the success it enabled. When the onboarding journey is reported, report the three days of month-end reconciliation alongside it. Debt that is invisible on the dashboard will never be funded, because it is competing with things that are visible.
- Triage by strategic consequence, not engineering discomfort. The question is never “is this code we dislike?” It is “does this compromise stand between us and something we have decided to do?” A short list built that way is fundable. A long list built the other way is not.
- Attach remediation to the next feature, not to a debt programme. The most durable way to pay down strategic debt is to require that the next change through the constrained area leaves it cleaner than it found it — because that change is already funded, already scheduled, and already touching the code. A standalone “debt sprint” competes with delivery for budget and loses. Remediation folded into delivery does not.
- Name the single points of knowledge as risks, not inconveniences. The contractor who alone understands the mapping is not a staffing detail; they are a strategic exposure with a notice period. That belongs on a risk register, costed, not in a quiet corner of someone’s worry.
- Decide, in daylight, what you are choosing to keep. Some of the debt you will look at squarely and elect to live with — the dormant compromise in the backwater genuinely can sleep. The difference between that and neglect is that you decided. Debt you have consciously chosen to carry is a position. Debt you are carrying because no one looked is an accident waiting for a strategy to expose it.
The through-line of all five is the same: bring the borrowing into the light, price it honestly, and then make a portfolio decision about it — pay this down because it blocks the plan, carry that because it never will. That is a leadership act, not an engineering one, which is precisely why it has been so hard to get onto the right agenda.
Coda
The pandemic sprint was not a failure of discipline that we must now atone for. It was one of the profession’s finest hours, and the debt it created is the honest price of a set of decisions that were, very nearly all of them, correct. To be ashamed of the debt is to misunderstand it. It is the receipt for having done the right thing under pressure.
But a receipt is not the same as a settlement. The organisations that emerge strongest from the next eighteen months will be neither the ones that flagellate themselves into a doomed programme of total remediation, nor the ones that declare victory on the strength of a green dashboard and let the arteries harden unwatched. They will be the ones that do the unglamorous middle thing: look honestly at what they borrowed, work out which of it is quietly deciding their future, and pay down that — deliberately, in daylight, one funded feature at a time. The bill for the sprint was always going to come due. The only real choice was whether we would open it, or wait for it to open us.