The Hidden Cost of Running Legacy Systems

Every decorated goods business I've worked with has at least one system they've been "meaning to replace." Sometimes it's the ERP. Sometimes it's the website. Sometimes it's a production management tool that was built when the business was a quarter of its current size. And the reason it hasn't been replaced is always the same: the known cost of replacement feels higher than the pain of staying put.

The problem with that calculation is that most of the cost of staying put is invisible. You see the licence fee and the support contract. You don't see the hours your team spends compensating for what the system can't do.

The Direct Costs

The direct costs of a legacy system are the ones that appear on invoices: annual licence fee, support contract, hardware maintenance, and IT time spent keeping the system running. These are visible and usually easy to quantify.

What's less visible is that legacy system support costs tend to increase over time, not decrease. As fewer providers support older systems, the cost of specialist support rises. Hardware that runs an old on-premise system eventually needs replacing — and may require significant work to get the legacy application running on newer infrastructure.

The Hidden Workaround Costs

This is where most of the real cost sits. A legacy system doesn't do something the business needs, so the team builds a workaround. The workaround becomes the process. Nobody questions it because it's "how we do things here." The cost never appears on an invoice.

Common workarounds I find in decorated goods businesses:

• Manual data rekeying. Orders taken on one system, manually entered into another. Purchase orders raised in the ERP, manually entered into the supplier portal. Invoices received electronically, manually typed into the accounts system. Each rekey takes time and introduces error risk.
• Spreadsheet bridges. A spreadsheet that sits between two systems that don't talk to each other. Someone updates it daily. When that person is on holiday, it doesn't get updated, and both systems diverge.
• Reporting workarounds. The ERP can't produce the report that management needs, so someone exports the data monthly and builds the report in Excel. It takes 4 hours. It's wrong sometimes. Nobody knows when.
• Dual approval processes. The system doesn't enforce the approval process, so it's enforced via email. The email chain is the audit trail. Tracking down a historical approval takes 30 minutes of inbox searching.

To quantify the workaround cost, ask this: how many hours per week does your team spend doing something that should be automatic? Put a value on that time. Multiply by 52. That number is often larger than the cost of replacing the system.

The Opportunity Costs

Opportunity costs are the hardest to quantify but often the largest. They're the things you can't do because the system won't support them.

Examples:

• You can't offer a B2B portal because the ERP doesn't support the integration required
• You can't take on a corporate account that requires EDI connectivity
• You can't onboard a wholesale supplier because the integration isn't possible
• You can't offer real-time order tracking to customers because the data isn't accessible
• You can't make quick pricing decisions because the margin analysis takes a day to produce

None of those appear on a balance sheet as a legacy system cost. But each one represents revenue that's going somewhere else.

The Decision Framework: Replace, Defer, or Tolerate

Not every legacy system needs replacing immediately. The decision should be made against a clear framework:

Replace now if: the workaround cost over 3 years exceeds the replacement cost; the system is preventing a commercially significant capability; the security or compliance risk is unacceptable; or the vendor has stopped developing the product.

Defer if: the workaround cost is quantified and manageable; a replacement project is planned and scoped for a defined future date; and there is a clear owner of the deferral decision who will review it at the agreed point.

Tolerate if: the system functions adequately for its specific scope; workarounds are minimal; and the cost of replacement significantly exceeds the cost of continuing for the foreseeable planning horizon.

The mistake is making these decisions without doing the cost calculation. Most businesses tolerate legacy systems by default rather than by decision. The difference matters — because a decision to tolerate includes a review date and a cost cap. Default tolerance has neither.

Common Questions

How do you calculate the true cost of a legacy system?

Three layers: direct costs (licence, support, hardware), hidden workaround costs (staff time compensating for what the system can't do), and opportunity costs (revenue and capability you can't access). Most businesses only see the first layer. The second and third typically double the number.

When should you replace a legacy ERP?

When the workaround cost over 3 years exceeds the replacement cost, when the system is blocking a commercial capability you need, or when the security risk is unacceptable. Don't replace because it's old — replace because the business case is clear.

What happens if you keep delaying replacement?

Workaround costs compound. Support resources become harder to find. Workarounds become more embedded and harder to remove. The eventual migration becomes more complex. Delay has a cost — it's just deferred rather than immediate.