Munger's most cited advice is deceptively simple: invert, always invert. When a problem resists the head-on approach, turn the question around and answer that instead.
Deals typically reach an acquisitions team through brokers or relationships. Many of our clients receive more brokered deals than a team can read carefully, which is why we built fit scores: a qualitative and quantitative read of every incoming deal against the firm's investment criteria. Non-fit deals are out in seconds, and the rest get real attention.
Outbound-sourced deals, however, are the true prize in real estate: no competition and lower prices. Teams want to source proactively, but manually sifting county records and transaction databases is painfully inefficient, and the volume you have to get through to find one good deal is enormous. Inbound deal filtering is systemized. The outbound half is not.
It stayed manual because filtering the entire universe of transactions against a buy box was infeasible. Until now. Inverted, the criteria that define exactly what you want to buy also tell you exactly where to look, before those deals ever hit the market.
Here is what that looks like in Starboard, end to end. Your buy box, run as a query against the market, returns every mandate-fit asset on record. Timing signals cut that universe to the owners likely to transact right now. The context engine ranks those down to the handful worth a call this week.
Layer 1: Turn Your Buy Box Into a Search
Your buy box (size range, submarket, asset class, vintage, occupancy threshold) is already a precise description of what you want to own. Run against inbound deals, it is a gate. Run through the universe of all MSCI properties, it is a map of potential deals.
If your mandate is 20,000–35,000 SF light industrial, infill Phoenix, sub-5% vacancy, built 2000 or later, that's a specific, countable set of buildings, each with a known owner, loan maturity schedule, and transaction history. But even that filtered list is far too long to work through by hand. The map needs narrowing: which of these owners is actually likely to sell?
Layer 2: Monitor the Signals That Precede a Sale
Starboard scores transactability, a measure of how likely an owner is to sell, reading across the owner (fund vintage, hold period, portfolio pressure), the asset (debt, occupancy, capex position), and the market (pricing, liquidity). Every asset in your mandate-fit universe runs through this screen automatically, so what surfaces is only the small slice that is highly relevant right now.
Loan maturity is the clearest signal. A fund with $40M of floating-rate debt maturing in nine months, on an asset that hasn't been refinanced or listed, is most likely a motivated seller, whether or not they're officially "on the market." MSCI Real Capital Analytics often surfaces this data. Most acquisitions teams look at it when evaluating an OM. The inverting move is to run a loan maturity screen on your target property universe every day, and surface motivated owners before a broker does.
The asymmetry is significant. When a high-transactability signal appears on an asset that fits your buy box, you usually have a 2–4 month window before the owner decides to formally list, engage a broker, or take another path. That window is the competitive advantage.
A 47% transactability score pursued proactively beats a 95% fit score pursued reactively. Fit tells you what you want. Transactability tells you what will close.
Layer 3: Let the Context Engine Learn What You Actually Buy
The first two layers produce a list. The third makes the list better every week, without anyone tuning it.
Every deal your team touches in Starboard (viewed, saved, pursued, passed) becomes part of your firm's context engine, the same firmwide memory that already holds your past deals and underwriting. Over time, the engine learns the difference between what your firm says it buys and what it actually chases. Stated preferences are the criteria you wrote down. Revealed preferences are the pattern in your decisions.
How Starboard Actually Runs It
Everything above is live inside Starboard today, running alongside your inbound pipeline.
You set your buy box once, through the Thesis Wizard: write the mandate in plain language, and Starboard asks the questions that sharpen it (which counties, what deal size, what floor) until the thesis is precise enough to query. Each answer visibly narrows the universe, so you watch the market shrink toward your mandate as you type.
From there, Starboard runs the thesis continuously against MSCI Real Capital Analytics data: more than $40 trillion of commercial property transactions, linked to over 200,000 investor and lender profiles, tracked since 2000. That is the outbound universe your buy box queries, and it stays current as buildings trade, lease up, and refinance. Timing signals update as loan maturities approach and occupancy shifts, so assets move in and out of the motivated slice on their own. And because your team already works its pipeline in Starboard, the context engine keeps sharpening the ranking in the background.
What lands on your desk is a weekly call list: a handful of owners, each with the reason they surfaced, each traceable to your own criteria.
And when an owner conversation turns into a live opportunity, it enters your pipeline like any other deal: full analysis, underwriting, and the IC presentation, all built on the same context engine that found it.
Outbound sourcing is available now for every Starboard client, and it takes less than 10 minutes to see your first wave of deals. If you want to see it running on your own buy box, schedule a demo.
In CRE deal sourcing, Munger's inversion is a specific, executable instruction: take the criteria you use to filter what comes to you, and run them in the other direction. The deals worth finding are already in the data. They just haven't been sent to you yet.
Data insights courtesy of MSCI.