WorkSitePass

The protagonist is the site superintendent at 7am, not the safety manager at a desk. That distinction turned out to be the whole product.

Discovery: what contractors actually told me

Around a dozen conversations with site superintendents, health-and-safety managers and contractors across Canada and the US. Not usability sessions — discovery calls, where the goal was to find out whether the problem I imagined was the problem they had.

How I ran the calls:

• Questions, not a pitch. “When a worker shows up and someone has to confirm their tickets are current — how does that actually happen?”
• Ask about the last failure. “When it slipped last time, did work stop, or did it get caught later? What did it cost?”
• Test whether it's a system. “Is that a system, or one person being vigilant?” — the question that consistently opened people up.

Competitive analysis: where the gap actually was

Salus and SiteDocs are good products. They're built for a health-and-safety manager running a program from a desk — and priced and scoped for companies big enough to have one.

The gap wasn't a missing feature — it was a missing user. Every tool in this market serves the desk; the decision happens at the gate.

This is the part outsiders miss, and it's why the problem is worth solving commercially. A lapsed certificate isn't an admin nuisance — it's bid eligibility and cash.

Same data. Different people, different moments.

The reframe that shaped the product: this isn't a tracking problem, it's a visibility problem. The same certification data is needed by different people, in different places, at different moments — and the existing tools only serve one of them.

Competitors serve the H&S manager well. Nobody was serving the superintendent.

Four decisions that exist only because of a discovery conversation. None of them were in my first version of the idea.

The third row is the one I'd have got wrong on my own — I assumed adoption would be all-or-nothing across every subcontractor.

The readiness flow, end to end

One worker, one path. Everything in the product exists to make step four take under ten seconds, outdoors, on a phone.

1. Worker uploads — certificates added once, from the worker's own phone.
2. Verified at source — credentials checked against the issuing body, not a photo taken on trust.
3. Assigned to a site — the site defines its requirements; the worker is matched against them.
4. Cleared at the gate — green or red, in seconds, before anyone walks on.
5. Expiry flagged ahead — renewals surfaced 30 days out, so they never become a gate problem.

How it changed: from a wallet to site readiness

It started as a certificate wallet — an app where a worker keeps their tickets. It became site access readiness, ending in check-in at the gate. That shift came from the conversations, and it changed what the product is for.

There is no “worker profile” page in WorkSitePass, and that's deliberate. The worker owns their identity, the phone is the primary record, and the dashboard is an operational view of that identity — not a CRM of people.

A company never asks “who is this person?” It asks: is this worker ready · which site are they on · what's missing · when does it expire.

• Status you can read at arm's length. Green or red as the primary signal, always paired with a label — bright sun, gloves, a phone held at an angle.
• One screen answers the gate question. No navigation to the answer. The readiness state is the landing state.
• Same system, three surfaces. Admin console, superintendent mobile view and worker wallet share one component set, so the data reads identically in all three.
• Setup that a superintendent can do. Every configuration screen is designed on the assumption that nobody will read documentation.

The pass itself

The worker's wallet and the superintendent's check are two sides of one object. The worker uploads once and carries it between general contractors; the super scans and gets an answer without needing the worker's paperwork at all.

One wallet per worker satisfies every general contractor's requirements — which is the part subcontractors actually care about.

What I'd measure next: trial-to-paid conversion and where trials stall, time to first cleared worker after signup, the share of gate checks done in-app vs. by phone call, and expiries caught before the gate vs. at the gate.

What surprised me, and what I'd do differently

I sent around a thousand cold emails and got essentially no replies. The same message, asked as a question in a real conversation, worked almost every time. The problem was never the audience — it was that I was pitching before I'd earned the right to.

My last two products taught me the same thing from opposite directions: find out whether you're right before you spend a year being confident.