Why Most Investors Analyze Multifamily Deals Wrong (And What I Do Differently)

A few months ago I sat across the table from an investor who'd just closed on a twelve-unit building. Good location, decent bones, and on paper, a strong return. He was proud of it, and honestly, he should've been. He'd done more homework than most people do before buying a property.

Except when I asked him where the rent numbers in his pro forma came from, he said "the seller's rent roll." When I asked about the expense figures, same answer. When I asked what vacancy rate he'd underwritten to, he told me the building was fully occupied, so he'd assumed close to zero.

Eighteen months later, that same investor called me. Two units had turned over and weren't renting for anywhere near what he'd projected. Property taxes had jumped after the sale triggered a reassessment. And a boiler that "still had a few good years left" didn't.

None of that is bad luck. It's what happens when you underwrite a deal to someone else's story instead of your own numbers.

The pattern I see over and over

I've looked at a lot of multifamily deals at this point, and the ones that go sideways almost always share the same root cause. Not a bad market. Not bad timing. A pro forma that was never conservative enough to survive contact with reality.

Here's what that usually looks like in practice: the rent roll gets treated as fact instead of a starting point. The seller's expense numbers get copied over without adjustment. The vacancy assumption reflects the building's current, fully-leased snapshot instead of what's actually happening in that market. And somewhere in there, everyone quietly assumes nothing major will break for the next few years.

Individually, none of these feel like a big decision. Together, they're the difference between a deal that works and one that just looks like it works.

Why this matters more right now than it used to

Canada's rental market has shifted in a way that makes sloppy underwriting a lot more expensive than it used to be. The national vacancy rate for purpose-built rentals climbed to 3.1% in 2025, up from 2.2% the year before, and it's now sitting above the ten-year average. Vancouver's vacancy rate hit its highest point since 1988. Toronto crossed 3% for the first time since the pandemic.

For years, you could underwrite a deal a little loosely and the market would bail you out. Rents kept climbing, vacancy stayed tight, and even an optimistic pro forma had a decent shot at working out. That's not the environment we're in anymore. When vacancy is rising and rent growth is slowing, the gap between what you projected and what actually happens shows up fast, and it shows up in your bank account.

What I actually do differently

I don't look at a deal's projected return until I've rebuilt the numbers myself. Every time.

That means confirming actual in-place rents against signed leases and bank deposits, not the rent roll a seller hands over. It means using current, not peak, comparable rents when I'm projecting what a vacant or below-market unit could achieve, and being honest about how long it'll actually take to get there. It means pulling trailing twelve-month actual expenses, adjusting property tax for what it'll become after reassessment at the new sale price, and building in a real contingency line for the repairs that always seem to show up right after closing.

And I underwrite vacancy to what CMHC is actually reporting for that specific market, not to whatever the building happens to be running at the moment I look at it. A building sitting at full occupancy today tells you almost nothing about what happens at the next round of lease renewals.

Interestingly, this is basically the same standard CMHC applies to its own multifamily financing programs. Their underwriting relies on independently verified rental, vacancy, and expense benchmarks rather than taking a borrower's projections at face value, and if a submission looks overly optimistic, they'll apply more conservative assumptions themselves. If the institution insuring the mortgage doesn't trust the seller's story, I'm not sure why I would.

The uncomfortable part

Here's the thing nobody likes to hear: when you underwrite this way, a lot of deals that looked exciting stop looking exciting. That's the point. I'd rather find out a deal doesn't actually work while I'm still reviewing a listing than find out eighteen months after closing, the way that investor I mentioned earlier did.

Conservative underwriting isn't about being pessimistic. It's about being honest with yourself before the numbers become permanent.

If you want the exact framework I use to rebuild a deal's numbers before I ever make an offer, along with the full four-step process I walk every acquisition through and a checklist you can run against your next deal, I put it all together in The Real Estate Deal Analysis Guide. It's free, it's built specifically for the Canadian market, and it's the same process behind every deal we underwrite at Delftrise.