A person uses the ArriveCan app on Feb. 9.Laura Proctor/The Globe and Mail
Sarah Niedoba is outreach manager at Code for Canada, a tech nonprofit. Laura is director of innovation and outreach at Code for Canada.
If Canadians feel angry reading the revelations last week about ArriveCan, they should.
Auditor-General Karen Hogan’s report on the federal government’s app for travellers during COVID-19 comes on the heels of countless reviews and committee hearings, after The Globe’s reporting found that the app’s price tag had ballooned to at least $54-million.
Ms. Hogan wrote that the Canada Border Services Agency (CBSA), Public Health Agency of Canada (PHAC) and Public Services and Procurement Canada (PSPC) had “repeatedly failed to follow good management practices in the contracting development and implementation of the ArriveCan application.”
That mismanagement included contracts more-or-less developed by the winning contractor and documentation so incomplete the Auditor General had to guess at the actual cost of the app. The groups involved mismanaged the project and disrespected public funds.
But there’s more at stake here than one poorly developed app.
This is a crucial moment for the federal government – will they finally address a broken procurement system that inevitably leads to more ArriveCans and Phoenix pay systems? Or will they finally make the changes needed to successfully procure, manage and develop the digital solutions Canadians deserve? It doesn’t look like it.
Let’s look at what went wrong. The Globe’s reporting detailed how GCStrategies, a two-person remote-based company, won millions of dollars of government contracts, only to outsource the work to other IT contractors while collecting a sizeable commission for themselves.
It’s reasonable to ask, why would a company like this exist? But the answer comes from the government’s own Procurement Ombudsman, Alexander Jeglic, who told MPs: “The complexity of the procurement animal is so great that you do have entities that exist only for helping other firms comply with the set of regulations.”
In other words, the existing procurement process is so complicated and tedious that only a select group of companies have the time, resources and expertise to compete for contracts – and a player like GCStrategies can charge other companies a commission simply to secure the work.
Currently, there are two main ways the federal government procures tech vendors. Either enormous error-riddled requests-for-proposals that require hundred-page responses or something called “task-based informatics professional services,” which entails paying tech vendors for their work on a “per diem” basis, incentivizing them to lengthen projects to bill for longer periods.
Many IT companies fully capable of successfully developing an app like ArriveCan at a reasonable cost simply don’t have the time, resources or knowledge to work within either of these systems. And so, the same vendors continue to be awarded for projects that arrive poorly designed, over budget and over deadline.
Yet, in response to the Auditor General’s report, the CBSA and PSPC have suggested unproductive modifications to the dismal status quo.
This week, the CBSA announced a “Procurement Directorate” to handle all procurement decisions and a “contract review board” to approve contracts and authorizations around specific contract tasks. Beginning March 31, all CBSA staff must report every interaction with private vendors.
Meanwhile, PSPC, the department responsible for government-wide contracting, said it is updating its guidelines for procurement staff so that all contracting-related documents will need clear descriptions of completed work.
None of these proposed changes will make it any easier for an IT company not already overly familiar with the federal government to apply for government work. They also won’t address the elephant in the room –the lack of in-house digital know-how that allowed the ArriveCan failure to unfold in the first place.
The federal government must embrace changes that will make a meaningful difference in developing and launching digital products and services. They need to adjust their procurement system for agile digital development – shorter contracts allow a project to be scoped, launched and continually improved based on public feedback.
They must invest in hiring and training public servants to work on digital projects in-house to reduce their reliance on outside contractors. This means creating job titles for roles that have existed outside the public service for decades, like software developers, product managers and user experience designers. It means adjusting woefully low salaries for digital roles in the public service instead of paying exorbitant fees to outside contractors. It means improved training for existing leadership so they can scope and manage digital projects (and know when one is going off the rails).
The public sector is famed for being risk-averse, and there is good reason for it to approach its work differently than the world of “move-fast-break-things” private sector tech. But the true risk in this instance is not to do things differently. Unless something changes, another ArriveCan bungle in the headlines isn’t a risk – it’s a certainty.