Most acquisitions teams respond to a rising pile of offering memorandums by trying to look at more of them, faster. That is the wrong fix. Precision deal screening, not raw throughput, determines whether a team's time goes to deals that can close. A screen that lets the wrong deals through wastes exactly as many analyst hours whether it processes twenty OMs a week or two hundred. The volume framing treats screening as a bottleneck of quantity. It is a bottleneck of discrimination: how well the filter separates deals worth a model from deals that were dead on arrival.
Key Takeaways
Precision deal screening measures how many of the deals that clear the initial screen fit the buy box. A screen that lets ten deals through with four real fits has 40 percent precision, regardless of how many total deals it processed.
Raising screening speed without raising precision moves the same error rate through the pipeline faster. It does not reduce wasted analyst hours; it can increase them.
CBRE projects U.S. commercial real estate investment volume will rise 16 percent to $562 billion in 2026, which means more inbound deal flow hitting the same screening capacity, not less.
A low-precision screen and a low-recall screen fail in opposite, equally expensive directions: one burns hours on deals that never fit, the other kills deals that would have.
The fix is a testable buy box and a feedback loop from closed and dead deals, not a faster read of the same loose criteria.
What does precision mean when a deal screen is the classifier?
A deal screen is a classifier: it sorts inbound OMs into "advance" and "reject." Precision is the share of the deals it advances that belong there, meaning they genuinely fit the buy box once underwritten. A screen that advances ten deals a week, of which four turn out to be real fits and six get killed after a model is built, is running at 40 percent precision no matter how many total OMs it looked at to get there.
This is a different failure from a capacity failure. A team that cannot look at enough deals has a throughput problem, and more software or more analysts fixes it. A team that looks at plenty of deals but keeps advancing the wrong ones has a precision problem, and more throughput only processes the same bad ratio faster. Confusing the two is why firms buy tools that summarize OMs quicker and see underwriting hours barely move.
Why do acquisitions teams default to volume as the fix?
Teams default to volume because deal flow is the visible symptom: inboxes fill up, analysts fall behind, and "we need to see more deals faster" is the obvious complaint to make in a Monday pipeline review. The actual leak, false positives clearing the screen and false negatives getting killed too early, is invisible unless someone tracks outcomes against the reasons a deal was advanced or rejected in the first place.
Symptom | Volume framing | Precision framing |
|---|---|---|
Analysts underwater | Hire more analysts or buy faster extraction | Find out how many advanced deals were never real fits |
Deals arrive faster than they can be reviewed | Speed up the initial read | Tighten the criteria the initial read applies |
Good deals feel like they slip away | Widen the top of the funnel | Check whether the screen is too aggressive, killing fits early |
IC sees weak deals reach the memo stage | Add more IC members to catch it | Fix the screen that let a non-fit advance in the first place |
The volume framing is comfortable because it does not require anyone to admit the criteria themselves are the problem. It is easier to ask for another analyst than to audit why six of the last ten deals that reached a model never should have.
How much does a low-precision screen cost?
A low-precision screen costs analyst hours in direct proportion to its false positive rate, and the cost compounds every week the criteria stay loose. Consider a team that receives 40 inbound OMs a week and advances 10 of them past the initial screen, a 25 percent advance rate. If that screen runs at 40 percent precision, four of the ten advanced deals are real fits and six are false positives that get built into a model before anyone realizes they were never in the buy box.
If each false positive consumes three hours of analyst time before it dies, meaning the deal gets modeled, discussed, and killed rather than rejected on the initial read, that team loses 18 analyst hours a week, or roughly 78 hours a month, to deals the screen should never have advanced. Raise precision to 70 percent on the same ten advanced deals and false positives drop from six to three, cutting the monthly waste to about 39 hours. Precision moved, the OM count did not, and the team recovered nearly forty hours a month it can now spend on deals that were real fits from the start.
That is the quotable version of the argument: a faster read of a bad filter produces bad decisions faster. Speed multiplies whatever precision the screen already has, for better or worse.
What raises precision without cutting recall?
Precision only improves without cost if it does not come at the price of recall, meaning the screen cannot simply reject everything to avoid false positives; it has to keep catching the deals that are real fits. The two failure modes pull in opposite directions, shown here:
Failure mode | What happens | Who feels it first |
|---|---|---|
Low precision | Deals that do not fit the buy box clear the screen and consume underwriting hours before being killed | Analysts, in wasted model-building time |
Low recall | Deals that do fit the buy box get rejected at the initial screen, often on a loose or outdated criterion | Principals, in deal flow that quietly disappears |
Raising precision without sacrificing recall requires two things a fast read cannot substitute for. First, the buy box has to be written as testable criteria, not adjectives: specific ranges for basis, unit count, submarket, vintage, and deal size rather than "value-add" or "good location." A criterion that cannot be checked against the OM's actual numbers cannot be screened against consistently, and inconsistency is what produces both false positives and false negatives from the same loose rule. Second, the team needs a record of why each deal was advanced or killed, so that when a deal dies downstream after clearing the screen, someone can trace which criterion let it through and tighten it. Without that record, a low-precision screen never improves; it keeps making the same mistake at whatever speed the team can now read at.
Why is rising deal flow making the precision problem worse, not smaller?
Rising deal flow raises the stakes of a low-precision screen because the same error rate now runs against a larger number of inbound deals. CBRE's 2026 North American Investor Intentions Survey found that 74 percent of investors plan to buy more commercial real estate this year, and CBRE projects total U.S. investment volume will rise 16 percent to $562 billion, nearly matching the 2015 to 2019 pre-pandemic average. JLL's 2026 investor survey found a similar imbalance in the retail sector, with 64 percent of investors planning to increase acquisitions against only 48 percent planning to sell.
More capital chasing deals means more OMs landing in acquisitions inboxes across the market, not a change in how good any single team's screening criteria are. A screen running at 40 percent precision against 40 OMs a week will run at 40 percent precision against 60 OMs a week; the absolute number of wasted hours simply grows with the inbound count. Teams that treat this moment as a volume problem will buy faster tools and process more bad deals at the same ratio. Teams that treat it as a precision problem will tighten the filter before the volume increase hits, and absorb more deal flow without a matching increase in wasted underwriting time.
Frequently Asked Questions
What is the difference between deal screening precision and deal screening volume? Volume is how many deals a team can look at in a given period. Precision is the share of the deals that clear the screen that turn out to fit the buy box. A team can raise volume and still have a precision problem if the ratio of real fits to false positives never changes.
How do you measure the precision of a deal screen? Track every deal that clears the initial screen through to its outcome: closed, killed at underwriting, or killed at IC. Precision is the number that closed or remained live divided by the total that were advanced. Most teams can reconstruct this from six months of deal logs even without a formal tracking system.
Does adding more analysts ever fix a precision problem? No. More analysts increase how many deals a team can process, but if the screen still advances the wrong deals at the same rate, more analysts build more models on deals that were never going to close. Fix the screen first.
What is the fastest way to raise screening precision without losing good deals? Write the buy box as testable numeric criteria rather than descriptive language, and track the reason every deal was killed against the criterion that killed it. That record shows which criteria are too loose, too strict, or simply wrong, without requiring new tools.
Conclusion
A deal screen's job is to discriminate, not to process. Every hour an acquisitions team spends underwriting a deal that clears the screen but was never a real fit is an hour a precision problem cost, not a volume problem. As inbound deal flow rises with the broader market, the teams that come out ahead will not be the ones that read OMs the fastest. They will be the ones whose screen was already precise before the flow increased, because precision, not speed, is what a busier market tests.
Related reading: what a buy box is and why yours cannot screen a deal, why screening 500 deals a week is a reading problem, not a headcount problem, why extraction accuracy needs precision and recall, not a single number, and the deal screening memo that records why a deal was killed.