The Interest You Pay on Yesterday's Shortcuts: A Process Debt Audit for 2026
Most UK SMEs do not operate on the processes their founders designed. They operate on a sediment of fixes: the spreadsheet built in a hurry one January, the email approval that quietly replaced a form, the person everyone asks because the system never quite worked. Each of these was a sensible response to a real constraint. The difficulty is that none of them arrived with an expiry date.
Process debt is the accumulated cost of those shortcuts. It is a close cousin of technical debt, but it rarely sits with the IT team. It sits in operations, in finance, in the diary of the managing director. It does not appear on the balance sheet, yet it is settled every month in hours, errors, customer friction and, most expensively of all, management attention.
The argument for auditing it is not that old processes are embarrassing. It is that workarounds compound. Left alone, they multiply, each fix layered on the last, until a business is spending a meaningful share of its capacity servicing its own improvisations.
The rational past and the expensive present
Almost no process debt is the product of laziness. A workaround exists because someone needed an answer before the system could provide one. The sales ledger that lives in a spreadsheet exists because the finance package was never configured for the way the business actually invoices. The manual check before every payroll run exists because of an error in 2019 that nobody wants to repeat.
That history deserves respect. It also deserves a review. The question is not whether the workaround was justified, but whether it is still earning its keep. A shortcut that solved a problem four years ago and has not been examined since is not a safeguard. It is a habit with a cost centre attached.
Six forms worth auditing
1. The human integration point
This is the workaround that lives in a person rather than a system: the colleague who knows how to export the report, the one who reconciles the two ledgers, the one who is copied in because otherwise nothing moves. It looks like teamwork. In practice it is a single point of failure wearing a friendly face. The interest is paid in key-person risk and in the queue that forms whenever that person takes a holiday.
2. The spreadsheet that has become a system
A tracker with fourteen tabs, conditional formatting and a macro nobody can edit is not a tool any more. It is an unaudited database. The interest is paid in version conflicts, in the afternoon lost when a formula is overwritten, and in the fact that no one can say with confidence which copy is correct.
3. The re-keying chain
Data entered once, exported, pasted, re-entered and then checked by hand. Each transfer is a chance to introduce an error and a certainty of consuming time. This is often the single largest recoverable cost in an SME back office, and the easiest to overlook because everyone involved is working hard rather than working wrongly.
4. The approval nobody can find
The decision that was agreed on a call, confirmed by text message and never written down. Approval-by-WhatsApp feels fast until a supplier queries a purchase order or an auditor asks who authorised a payment. The interest is paid in disputes, in duplicated requests and in the quiet reluctance of anyone to act without written cover.
5. The reconciliation ritual
The monthly exercise in which two sources of truth are forced to agree. Sometimes it is necessary. Often it is a substitute for fixing the point at which the two records diverged. The interest is paid in close periods that stretch, in late management information and in the fatigue of the person who performs it.
6. The exception that became the standard
Bespoke quotes, special delivery arrangements, customer-specific invoice formats. Exceptions are a legitimate commercial tool. When exceptions become the majority of transactions, the standard process no longer describes the business, and every new team member has to learn a rulebook that exists only in other people's heads.
Pricing the interest
The word debt is useful because it invites arithmetic. For each workaround, estimate four things: the hours consumed each month, the loaded cost of those hours, the frequency and cost of errors it causes, and the customer-facing impact where relevant. Multiply, and you have a monthly interest charge.
Add a fifth line for attention. Senior time absorbed by firefighting is the most expensive item on the list and the one least often counted, because it never appears in a timesheet. A useful discipline is to ask what the managing director would be doing with those hours if the workaround did not exist.
Deciding what to retire first
Not every workaround needs a rebuild, and treating them all as projects is its own form of waste. Sort them into three groups.
Retire: the workaround that can simply be deleted, because the original problem has gone or the system now handles it. These are the cheapest wins and the most frequently overlooked.
Contain: the workaround that must stay for now, but with a named owner, a written purpose and a date for review.
Rebuild: only where the underlying process is genuinely core to how the business creates value. This is where investment belongs, and it should be a small number of items, not a wish list.
A simple score helps: monthly interest paid, multiplied by the blast radius when it fails, divided by the effort to remove it. The arithmetic is less important than the argument it forces.
Every workaround needs an owner and an expiry date
This is the practical heart of the matter. Keep a process debt register, no longer than two pages. For each entry record what it is, why it exists, who owns it, what it costs and when it will be retired or reviewed. Review it quarterly, alongside the rest of the management agenda. The act of writing the expiry date changes behaviour: workarounds with dates get fixed, while workarounds without them become permanent fixtures.
A place to start this quarter
Choose one workaround from each of the three groups and act. Delete the one that no longer needs to exist. Give the one that must stay an owner and a review date. And take the one with the largest interest charge and put it on the agenda for the next quarter, with a named sponsor and a completion date before the end of 2026.
Process debt is not a sign of a badly run business. It is a sign of a business that has been busy. But the interest is real, it is paid monthly, and unlike most borrowing it can be repaid simply by deciding to stop.
