Underwriting has a reputation problem. People picture a spreadsheet with twenty tabs and a confident IRR at the bottom, and they assume the skill is in the modeling. It is not. The spreadsheet is the last step. The actual work of underwriting a deal is everything that happens before the numbers get typed in: walking the property, understanding why it is underperforming, and being brutally honest about what fixing that costs in money and time.
I have been on both sides of this. I have looked at deals where the seller's pro forma was a work of fiction, and I have caught myself building the model that argues for the deal I already wanted to do. The second one is more dangerous, because nobody is lying to you except you. Here is how I actually run the process on a Michigan deal.
Start with the basis, not the upside
The first number I care about is basis: what I paid, plus what the work truly costs, plus the carry while the work happens. Not the flattering version of that number — the real one. If a roof is near end of life, it is in the basis. If the parking lot needs to be milled and repaved in the next three winters, it is in the basis. Michigan is unforgiving on deferred maintenance; freeze-thaw cycles turn small exterior problems into structural ones on a schedule.
Everything downstream is judged against basis. A property that cash flows beautifully on a light basis is a good deal. The same cash flow on a heavy basis is a mediocre one. This sounds obvious, but most bad deals I see trace back to a basis that was optimistic on day one, and no amount of clever operations fixes an entry price that was wrong.
Name the cause before you price the cure
Every underperforming property is underperforming for a specific reason, and "the market" is almost never the whole reason. Vacancy has a cause. A rent roll below the neighborhood has a cause. Sometimes the cause is fixable — bad management, tired units, zero marketing. Sometimes it is not — a floor plan nobody wants, a location on the wrong side of a corridor, a tenant mix that repels the tenants you need.
My rule is simple: if I cannot name the specific, fixable problem, I am not buying an opportunity, I am buying somebody else's exit. The whole point of local knowledge is that it lets you tell those two situations apart. That is the same lens I apply across real estate investing generally, and it is why I focus on markets I actually know. You cannot underwrite the cause of a problem in a city you have only seen in a data pack.
Underwrite the downside case first
I run three cases like everyone else, but I care about them in reverse order. The bad case is the deal. If rents flatten, if a unit takes four months to turn instead of six weeks, if the capex number comes in twenty percent high — does the deal still hold together? Does it cover its debt, keep its reserves, and leave me the option to simply hold?
A deal that only works in the good case is not a deal, it is a bet, and bets belong in a different bucket of your life than investments. The base case should be boring and achievable. The good case is upside you are happy to receive but never needed. Real estate rewards this ordering because the asset is illiquid: you cannot change your mind cheaply, so you have to be right on the way in.
Time is a line item
The cost most models undercount is time. A value-add plan that takes eighteen months of your attention is not free just because the checks clear. I try to be honest about what a project will demand operationally, because attention is the scarcest resource I have, and every hour on a troublesome asset is an hour not spent on the companies and properties that are working.
This is where buy-fix-hold and building companies rhyme. When I evaluate a move at Detour, I ask the same question: what is the true cost in money and in months, and does it survive the bad case? The habit transfers cleanly because the underlying discipline is identical.
Walking away is a result
Most deals I look at, I pass on, and I consider that the system working. The point of underwriting is not to find a reason to buy. It is to find the truth early, while it is still cheap. A pass costs you some diligence time. A bad buy costs you years.
I am Travis Hermiz. I invest in Michigan because local knowledge is the one edge that does not transfer, and underwriting is where that edge actually gets used. If you want the broader story of how I think about building and holding, the about page is the short version, and the post on what real estate taught me about building companies is the long one.