Quick Answer
The most effective way to increase your underwriting capacity without adding headcount is to eliminate the data entry bottleneck. Most acquisition teams spend 60 to 70 percent of their underwriting time on manual data extraction (typing rent rolls, T12s, and OMs into Excel), leaving only 30 to 40 percent for actual analysis. Automating the extraction layer lets your existing team evaluate 2 to 3 times more deals using the same judgment and process.
The CRE underwriting capacity problem
Most acquisition teams screen only 30 to 40 percent of their inbound deal flow. The other 60 to 70 percent gets triaged based on headline numbers, location, or gut feel. Not because the team lacks judgment, but because there is no time to open every OM and build a model.
The math is straightforward. A typical analyst can fully underwrite 15 to 20 deals per month. If your team receives 40 to 60 OMs per month, only half get a serious look. The rest are killed without full evaluation. As one multifamily acquisitions principal put it: "There are a lot of times where a deal is marginal, and we don't bring it upstream because it's a lot of effort."
The hidden cost is not the time wasted on deals you did evaluate. It is the deals you never opened. What if your best acquisition this year is sitting in the pile that never got underwritten? According to McKinsey's research on CRE operations, top-quartile acquisition teams differentiate on deal selection quality, not deal volume. They close the same number of deals, but from a larger evaluated pipeline.
The aspiration gap
One self-storage acquisitions director described the gap clearly: "20 to 50 deals a month would be my ideal flow." He was doing 10 to 20. Not because of capital constraints or market conditions, but because he had to sit and type in rent rolls by hand.
Where underwriting time actually goes
The data entry layer consumes the majority of analyst time on every deal. Here is a typical breakdown for a single acquisition underwrite, from OM receipt to investment committee memo.
| Task | Time per deal | Automatable? |
|---|---|---|
| Data entry from rent roll | 30 to 60 min | Yes |
| Data entry from T12 operating statement | 20 to 40 min | Yes |
| Cross-referencing OM figures | 15 to 30 min | Yes |
| Market research and comps | 30 to 60 min | Partially |
| Financial modeling and analysis | 30 to 60 min | No (human judgment) |
| Investment committee prep | 15 to 30 min | No (human judgment) |
The first three rows, data entry and cross-referencing, total 65 to 130 minutes per deal. That is 60 to 70 percent of the total underwriting time on a typical deal. These tasks require zero judgment. They require a human to look at a PDF, find a number, and type it into a cell.
As one acquisitions VP described the pain: "The manual data linking within Excel models" is where all the time goes. Another put it more bluntly: "I have to sit and type in rent rolls." Yardi exports, in particular, are universally hated as a data source because of their inconsistent formatting.
5 approaches to scaling underwriting capacity
There are five realistic ways to increase your team's deal throughput without hiring a full-time analyst. Each has different cost, quality, and scalability tradeoffs.
1. Outsourcing to offshore analysts
Offshore analysts (typically in the Philippines or India) cost $10 to $15 per hour and can handle rent roll data entry, T12 population, and basic model setup. Many CRE teams use this approach, and it works. One acquisitions director described maintaining a "working relationship with a team of Filipinos at $10 an hour."
The tradeoffs: quality variance between analysts, time zone coordination overhead, training time on your specific model, and the need for your team to QA every output. Outsourced analysts also cannot reconcile conflicting data across documents without guidance, because that requires understanding your investment criteria.
2. Standardized templates
Building a standardized pro forma template reduces the modeling time per deal, because analysts are not rebuilding the spreadsheet from scratch each time. This is table stakes for any serious acquisition team.
The limitation: templates solve the modeling step, not the data entry step. Your analyst still needs to type the rent roll, T12, and OM figures into the template by hand. The format of models that come from brokers varies wildly, which means even a great template does not eliminate the extraction bottleneck.
3. AI document extraction tools
AI-powered document intelligence tools eliminate the data entry step entirely. You upload the OM, rent roll, and T12. The tool extracts the data, maps it into your Excel model, and returns your model populated and ready for analysis. The best tools also cite every extracted cell back to its source document and page, so you can audit any number in seconds.
The key advantage: consistent quality on every deal, instant turnaround, and no training overhead. The key consideration: you need a tool that maps into your model (not its own), handles the document types you actually receive (Yardi exports, broker OMs, hand-scanned rent rolls), and provides a real audit trail. General-purpose AI like ChatGPT cannot do this. You need a purpose-built CRE extraction tool.
4. Process optimization (tighter buy box, faster kill decisions)
Tightening your investment criteria reduces the number of deals that reach the underwriting stage. If you can kill more deals at the screening stage (before any data entry), you free up analyst time for the deals that matter.
This approach is free and effective. The risk: you may miss "cuspy" or marginal deals that would have been winners. Narrowing your buy box increases efficiency but reduces optionality. According to Deloitte's 2026 CRE Outlook, the firms generating the highest risk-adjusted returns are those with the broadest evaluated pipeline, not the tightest screening criteria.
5. Hybrid approach: AI for extraction, analysts for judgment
This is what top acquisition teams do. AI handles the extraction layer (rent rolls, T12s, OMs into your Excel model). Analysts handle the analysis layer (market assumptions, business plan, IC narrative). The analyst's job shifts from data entry to deal selection.
The hybrid model works because it preserves everything that makes your team valuable (judgment, relationships, market knowledge) while eliminating the work that does not require those skills (typing numbers from PDFs into spreadsheets). One acquisitions VP described the ideal state: "I give you my model, upload docs, get my model back exactly the way I want it, like an analyst would do."
Cost comparison: four ways to underwrite
Here is how the numbers break down across the four most common approaches. The cost per deal is what matters, because it determines your breakeven on scaling deal flow.
| Approach | Monthly cost | Deals/month | Cost/deal | Quality control |
|---|---|---|---|---|
| Junior analyst ($65K/yr) | $5,400 | 15 to 20 | $270 to $360 | High |
| Offshore team ($10/hr) | $1,600 | 15 to 20 | $80 to $107 | Variable |
| AI extraction (Primer) | Flat fee/mo | Unlimited | Low (flat fee) | Consistent, cited |
| DIY ChatGPT ($25/mo) | $25 | 5 to 10 | $2.50 to $5 | Low (no audit trail) |
The DIY ChatGPT approach looks cheapest per deal, but it breaks down at volume. ChatGPT cannot reconcile conflicting numbers across a rent roll, T12, and OM. It cannot map into your specific Excel template persistently. And it produces no audit trail linking outputs to source pages, which means your analyst still verifies every number manually.
Calculate your team's data entry cost
Use this formula to estimate what you spend today on the extraction step alone.
x $50/hr loaded cost x 52 weeks
= Annual cost of manual data entry
A two-analyst team reviewing 10 deals per week, spending 1.5 hours on data entry per deal, costs approximately $78,000 per year in analyst time on data entry alone.
What the best teams do differently
The best acquisition teams do not save time per deal. They see more deals. The goal is not to underwrite each deal 20 minutes faster. The goal is to evaluate 100 percent of your pipeline instead of 40 percent, so you are selecting winners from the full opportunity set.
One storage operator started putting LOIs out on deals they would have triaged out before. Not because the deals looked better on paper, but because they could finally afford to open the OM, build a quick model, and make an informed decision. Several of those "skip pile" deals turned out to be strong acquisitions.
Another acquisitions VP framed it this way: "We want to look at every deal as much as possible... look at the art of each deal rather than typing in an Excel sheet." The insight is simple. When you remove the data entry cost, the calculus on which deals to evaluate changes completely. Marginal deals get a fair look. And some of them win.
This is the critical point that NAA operational benchmarks confirm: institutional buyers are capital constrained, not deal constrained. They close a fixed number of deals per year. The value is not closing more deals. It is closing better deals, picking the best opportunities from a larger evaluated pipeline instead of settling for "good enough" from a narrow one.
The dead deal tax: the hidden cost nobody tracks
Roughly 80 percent of deals that cross your desk will be killed. That is normal. The problem is that each one still requires analyst time before you make the kill decision.
Every dead deal costs 20 to 30 minutes of analyst time: opening the OM, skimming the rent roll, checking the location, confirming it does not fit your buy box. At 10 to 15 new OMs per week, that is 8 to 12 deals that consume 20 to 30 minutes each before dying. The total: 4 to 6 hours per week on deals that never advance past initial screening.
Dead deal tax: weekly cost
As one platform CEO managing $2.4 billion in AUM described it: "The operational cost isn't in the 20% that advance. It's in the 80% you have to touch before killing." Reducing the kill-decision time from 20 to 30 minutes down to 5 to 10 minutes reclaims 2 to 3 hours per week. That is 100 to 150 hours per year returned to analysis, sourcing, and deal selection.
There is also a psychological cost. When killing a deal requires 30 minutes of work, teams develop sunk cost bias. They rationalize advancing weak deals because "we've already invested the time." AI that provides clear, dispassionate data on every deal helps overcome this bias by making the kill decision faster and easier to justify.
How to get started: 3 concrete steps
You do not need to overhaul your process overnight. Start with a one-week audit and build from there.
Audit your current process for one week
Track time per deal across your team. Break it into extraction (typing data into your model), analysis (assumptions, market research, judgment calls), and admin (formatting, IC prep, CRM updates). Most teams are surprised by how much falls in the extraction bucket.
Identify the extraction bottleneck
If more than 50 percent of your underwriting time is data entry, automation has the highest ROI of any change you can make. If most of your time is in analysis and market research, your bottleneck is different and automation will not help as much.
Test with a real deal
Use your own documents, your own model, your own workflow. Any tool worth using should be able to handle a recent deal you have already underwritten, so you can compare outputs. If the tool cannot work with your specific Excel template and your specific document types, it will not solve the problem. See our multifamily underwriting guide for a framework on what to test.