Ottawa sits at the centre of Canada's federal government machinery, and with that comes one of the most scrutinized — and criticized — procurement systems in the world. A former public service executive is now speaking out about a persistent problem: the government contracting process is structurally biased against off-the-shelf software solutions, even when they'd cost far less and work just as well.
The Customization Trap
The argument, laid out by the former executive, is straightforward: when federal departments issue requests for proposals, the requirements are so tightly specified — often written with a particular vendor's capabilities in mind — that commercial off-the-shelf (COTS) products rarely fit the bill without significant modification. At that point, you're no longer buying a product. You're commissioning a bespoke build at product prices.
This isn't a bug in the system, the executive suggests — it's a feature of how procurement was designed. Departments write detailed requirements to reduce risk and ensure accountability. But the unintended consequence is that every project becomes a custom project, with custom timelines, custom costs, and custom failure modes.
Why COTS Gets Bypassed
Commercial software products are built for the 80% use case. They're designed to work out of the box for most organizations, with configuration options covering common variations. But government departments often insist they represent the other 20% — that their workflows, their data structures, their compliance requirements are genuinely unique.
Sometimes that's true. More often, critics argue, it reflects institutional inertia: it's easier to justify a custom build than to adapt internal processes to fit a proven product. The procurement system rewards the former and makes the latter surprisingly difficult to propose.
There's also a risk calculus at play. Public servants are accountable for failures in ways private-sector managers often aren't. A COTS product that underdelivers on a specific requirement is visible and attributable. A custom solution that over-promises is at least defensible — the vendor failed to deliver, not the department.
The Cost of Getting It Wrong
The consequences are well-documented in Ottawa's recent IT history. Major federal technology projects have run hundreds of millions over budget and years behind schedule. The Phoenix pay system debacle remains the most painful example — a project that was meant to modernize government payroll and instead left tens of thousands of federal employees underpaid or overpaid for years.
Smaller failures happen constantly and attract less attention. Departments spend millions on heavily customized systems that are immediately outdated, impossible to maintain, and locked to a single vendor indefinitely.
What Reform Would Look Like
The former executive's prescription isn't radical: change how requirements are written. Instead of specifying solutions, specify outcomes. Instead of locking in technical architecture upfront, allow vendors to propose how they'd meet a functional need — including with COTS products that might require departments to adapt their own processes.
This approach, sometimes called outcomes-based procurement, has gained traction in the UK and parts of the US federal government. In Canada, pilot programs have shown promise, but culture change in the public service is slow.
For Ottawa's large contractor ecosystem — firms that have thrived on complex, long-running custom builds — the stakes are high. A genuine shift toward COTS-first procurement would reshape the local tech economy. Whether that shift ever arrives remains an open question.
Source: Ottawa Citizen
