Most owners and controllers can tell you what they spend on software licences. Fewer can tell you what an aging system costs in people. That cost shows up as a role you cannot fill, a contractor you cannot replace, or an analyst who spends Fridays fixing spreadsheets.
This is a staffing issue as much as a technology one. Here is where it lands.
The talent pool shrinks to fit the system
When your core application runs on an older language, a heavily customized platform or a vendor product nobody teaches anymore, you are recruiting from a small group. Many of those people are senior, well paid and not looking.
The result is predictable:
- Searches take longer because few candidates match.
- Offers run higher because the few who match know it.
- Strong generalists pass on the role because it will not build their skills.
If a job posting reads like a museum exhibit, good people notice. Candidates ask what they will be working on and what it does for their next job.
Knowledge sits in two or three heads
Plenty of mid-sized companies have a system that one person built or has maintained for 15 years. Everything works because that person knows which report to run first and which field to never touch.
Then they retire, take another offer or go on a long leave. A new hire arrives with no documentation, learns by breaking things, and takes months to become useful.
This is not limited to IT. It happens in:
- Operations, where scheduling or inventory logic lives in an old tool and one planner.
- Finance, where month-end close depends on macros that only one accountant understands.
- Manufacturing, where machine data and quality records are held in systems that do not talk to each other.
- Sales, where the CRM was customized years ago and nobody trusts the pipeline numbers.
Your best people spend time on workarounds
When a system cannot do what the business needs, people fill the gap by hand. They re-key orders, merge exports, chase approvals by email and rebuild the same report every week.
That is paid capacity going to work a better setup would do automatically. It also wears down retention. Skilled people leave jobs where they feel like human middleware.
When you calculate the cost of an old system, count those hours. A coordinator losing 6 hours a week to manual entry is close to a sixth of a full-time role.
AI and data work stalls on bad foundations
Many employers now want automation, better reporting or AI tools. Those projects need clean, connected data, and old systems rarely supply it.
Before you hire a data scientist or an AI engineer, ask whether they will spend their first year cleaning and connecting data instead. If so, the hire you need first may be a data engineer or an integration specialist, not the role on your wish list. Hiring the wrong profile first is a common and expensive mistake.
Contractors become permanent
When no one on staff can maintain a system, companies bring in a contractor. That is a reasonable move. It becomes a problem when the contractor is the only person who understands the system and the engagement quietly runs for years.
Contract help works best with a defined purpose: document the system, migrate it, train staff or cover a peak. Set an end point and a handover plan at the start.
A quick test for which systems cost you most
For each major application, ask:
- How many people truly understand it, and what happens if two leave?
- How long did the last opening take to fill for someone who works on it?
- How many manual steps surround it?
- Is it blocking a reporting, automation or AI project you want?
- Do you depend on an outside contractor to keep it running?
Two or more troubling answers on one system tells you where to start. You do not need to replace everything at once. Documenting a system and cross-training a second person is a low-cost first step that reduces risk right away.
What to do next
Pick your two most fragile systems this month and write down who knows them. Then decide what you need: a second person trained, a short contract to document and migrate, or a permanent hire with modern skills who can lead the change. Build the plan before the key person gives notice, not after.
A Traverse recruiter can help you scope those roles, whether they are contract, permanent or a mix, and find people in operations, finance, data and AI across Canada and the US.