Most multifamily buyers review individual deals constantly and review their pipeline almost never.
The deals get scrutiny because there’s a deadline attached. The pipeline is the thing that quietly stops working, and you don’t notice for two or three quarters because every individual week feels productive. You’re taking broker calls. You’re underwriting. You’re busy.
Then a year goes by and nothing closed.
This is the review to run every quarter. It takes an afternoon. Six checks, in order.
1. Sort your last 20 deals into three columns
Start here, because everything downstream depends on the answer.
Pull the last 20 deals that reached your desk. For each one, mark it: lost on price, lost on terms, or never made an offer.
The distribution tells you which problem you have.
Heavy on “lost on price” means you’re seeing the right product and losing a competition. That’s a pricing, capital, or market problem.
Heavy on “never made an offer” means your criteria and your market aren’t matched. That’s a completely different problem with a completely different fix, and it gets misdiagnosed as the first one constantly.
Can’t fill out 20 rows at all means your deal flow is the constraint, not your strategy. Go work on broker relationships and direct-to-seller volume before you touch anything else in this list.
2. If you’re losing on price, find out who’s winning
The instinct when you keep getting beaten by 20 or 30% is to assume your underwriting is too conservative. Somebody is paying that number, so they must see something you don’t.
Sometimes they do. Often they just need something different from the asset than you do.
I went through this and eventually started talking to the buyers beating me. Their targets were around a 10 to 12% IRR and a 5% cash on cash. Mine were 17 to 18. They weren’t buying a wealth-building vehicle. They were parking capital and buying depreciation, making their money in another business entirely, and treating apartments as something closer to a bond position.
Nothing was wrong with my model. We were buying the same building to solve different problems, and their problem let them pay more.
Two honest caveats. First, this is a comfortable story, and at some point “everyone else is wrong” becomes a way of never transacting. Test it by naming the actual buyer profile. If you can’t, you may simply be low.
Second, if that buyer profile dominates your market, that market is structurally expensive for you. You can accept lower returns, or you can go somewhere the competition needs what you need.
3. If you’re not making offers, check whether your product trades
This is the check almost nobody runs.
I once shifted my buy box up to 1990s vintage, 100 to 150 units, and told my brokers what I was looking for. They laughed at me. Texas went through a bank and loan crisis that produced close to a decade with almost no development. There’s a great deal of 70s and 80s product here, close to nothing from the 90s outside of a few boutique builders, and then construction resumes in the mid 2000s.
I was shopping for a product my market never built.
So run the check. Over the last two quarters, how many properties actually traded in your market that fit your vintage, size, and class? Not listed. Traded. Your broker can tell you in one phone call.
If the answer is a handful, one of two things has to move. Either the box or the market. I moved the box, splitting it into a C-class 1980s bucket and a B/A-class 2000s bucket rather than holding out for a vintage that doesn’t exist here.
The risk in this check is that it becomes permission to loosen your standards indefinitely. The line is whether the product is absent or merely expensive. Absent means change something. Expensive means you’re back in check two.
4. Confirm your screening criteria are still in the right order
Screening order determines where your underwriting hours go, and most people never revisit it.
The first thing I look at on any deal that comes in is median household income for the area, at a 1-mile and 3-mile radius. We hold a floor and I don’t make exceptions on it anymore. After that comes crime, distance to major employment centers, population, population growth, and job growth. Price doesn’t enter the conversation until all of that clears.
Income moved to the top of that list relatively recently, and the reason is worth understanding because it isn’t financial.
We started our own property management company about three years ago. Once you’re watching daily operations across assets in genuinely different income bands, the difference shows up as management hours per unit. Lower-income properties absorb disproportionate time, and time is the actual constraint on everything else you’re trying to build. The deal you never found because you spent the quarter putting out fires is a real cost that appears on no statement anywhere.
Run this check by asking a simple question: over the last quarter, what percentage of your underwriting hours went into deals that failed on a criterion you could have checked in five minutes?
If it’s high, your screen is in the wrong order.
5. Refresh your capital stack assumptions
The fastest-decaying inputs in your model are the ones a lender gave you, and almost nobody updates them on a schedule.
Call every lender you’d realistically use and get four numbers on paper for each: maximum LTV they’re doing right now, current spread over the index, the DSCR they’re underwriting to, and the debt yield they’re pricing to. Then ask whether they’re holding their terms or backing off proceeds.
This used to be a quarterly task and honestly isn’t often enough anymore. Rates have moved enough inside single 30-day windows to change a deal materially. If your model is running on a quote from two quarters ago, your offer price is wrong and you’ll find out after acceptance.
Your brokers are worth asking too. They watch the same indexes and they’d rather tell you now than lose a closing later.
6. Recheck your all-in cost, not your price per door
Price per door is the number everyone quotes and the number that fools people in a distressed market.
When cash gets tight, owners stop fixing things. Deferred maintenance compounds for two or three years. Then the lender takes it back, and lenders aren’t operators. A special assets group or a court-appointed receiver has one real objective, which is getting the property to cover the loan long enough to sell it without a loss. Nobody in that chain is spending money on the roof.
So the property you’re being shown at an attractive price per door may need a great deal of work that the price per door doesn’t reflect.
Your all-in number should include purchase price, capex, reserves, and closing costs. And reserves scale with the reposition. A stabilized B or A class asset might carry three months. A heavy value-add can need six months or considerably more, and whatever timeline you think it will take to stabilize, add six months to a year to get the real number.
Then set your exit against something real. My check is where the previous market cycle topped out. If I’m all in around $100,000 to $110,000 a door on C-class product here, I’m underwriting an exit somewhere around $135,000 to $140,000 in five to seven years, and then testing that against how this market actually performed through the last run-up. As long as I’m not projecting past the previous market top, the exit assumption is defensible.
What to do with the output
The point of this review isn’t a document. It’s picking one thing.
Most quarters, one of these six checks comes back clearly worse than the others. That’s the quarter’s work. Fix the screening order. Refresh the lender numbers. Have the uncomfortable conversation about whether your product actually trades where you’re looking.
The failure mode isn’t picking the wrong fix. It’s four quarters of being busy without ever running the review.
Put it on the calendar the same way you’d schedule anything else that only matters if it actually happens. First week of the quarter, one afternoon, six checks in order. If you’ve never done it, the first one will take longer and will probably turn up something you already suspected.
And if you run it and the answer comes back that your product doesn’t trade in your market, or that the buyers beating you don’t need what you need, those are hard calls to make alone. I’ve made both of them wrong before making them right. If you want a second read on what your last 20 deals are telling you, get in touch. Happy to look at it or not.
Vince
