Parag Goswami, CEO, Clik.ai
Deal Screening on Steroids
Guest: Parag Goswami, CEO, Clik.ai
In brief:
- Most underwriting delays are caused by manual data extraction, not modeling complexity.
- Institutional CRE teams still rely on Excel, but increasingly automate the data layer beneath it.
- Parsing rent rolls and T12s accurately is now a scalability constraint, not a staffing issue.
- AI underwriting tools are being adopted first by lenders and servicers, not acquisition shops.
- Speed, consistency, and error reduction are emerging as underwriting risk controls.
Â
Why this conversation matters now
Artificial intelligence in commercial real estate is often discussed at the level of prediction, pricing, or market foresight. In this Demo-Day podcast with Parag Goswami, CEO and co-founder of Clik.ai, you'll discover something more prosaic and more urgent: the mechanics of underwriting itself.
Â
The core thesis is straightforward. The bottleneck in commercial real estate underwriting is not Excel. It is the labor-intensive, error-prone process of extracting data from documents and forcing it into Excel in the first place. Clik.ai does not replace spreadsheets. It automates everything that happens before the spreadsheet becomes useful.
Â
That distinction explains why Clik.ai's earliest and stickiest adopters are institutional lenders, servicers, and credit teams rather than opportunistic acquisition shops. At scale, underwriting is not about clever assumptions. It is about throughput, consistency, and control.
Â
What problem is Clik.ai actually solving?
As Goswami explains, underwriting begins long before assumptions are debated. A broker sends an offering memorandum. If interest remains, the next step is collecting rent rolls, trailing twelve-month statements, and historical financials. These arrive in PDFs or documents of wildly varying formats.
Â
"There has been no technology that can formalize all this information into a simplified way," Goswami says. "So this process of building even a preliminary underwriting model… takes hours."
Â
Traditionally, firms solve this in three ways:
- Hiring large in-house analyst teams
- Outsourcing data entry overseas
- Accepting that analysts will spend much of their time on manual work
All three approaches scale poorly. They introduce errors, create bottlenecks, and raise costs as portfolios grow.
Â
Clik.ai inserts itself at this exact choke point.
Â
How the system works in practice
Clik.ai uses client's existing Excel models. It does not impose a standardized template or replace internal underwriting frameworks.
Â
"Think of an AI that is sitting on top of your own Excel models to speed up the process," Goswami says.
Â
Users upload raw documents such as rent rolls, T12s, and operating statements. The system parses these documents regardless of format, normalizes the data, and maps each line item into the client's own chart of accounts.
Â
The output is not a new model. It is the client's model, fully populated.
As Goswami puts it, "Your own spreadsheet loaded with this data without manual work."
Â
This matters because underwriting conventions differ widely. One firm's definition of income, expense, or capital item may not match another's. Clik.ai is trained to mirror how the sponsor already underwrites, rather than forcing behavioral change.
Â
Why lenders and servicers adopted this first
The conversation makes clear that Clik.ai's core users are not casual investors screening a handful of deals. The typical customer manages roughly $500 million in debt portfolios, often across 100 or more assets but the tool can be used by much smaller teams too, particularly on the acquisition side.
Â
At scale, underwriting becomes infrastructure.
Â
Servicers reviewing quarterly statements across thousands of properties face the same problem repeatedly. Every reporting period, new PDFs arrive and analysts must confirm compliance, track variances, and ensure performance aligns with original underwriting.
Â
Goswami describes a servicer managing approximately 2,300 properties. Over five years, that firm grew assets under management from $23 billion to over $65 billion without adding headcount to its financial spreading team.
Â
That outcome is not about insight. It is about capacity.
Â
Speed as a form of risk management
A recurring theme is that speed is not merely a productivity benefit. It is a risk control.
"The more data you can look at analytically, that's where Clik comes into play," Goswami explains.
Â
Once data is normalized, it can be compared across portfolios, markets, and time periods. Clik.ai users can layer third-party data, such as Trepp, directly onto underwritten financials.
Â
This enables comparisons that are otherwise impractical. Revenue and expense comps. Payroll anomalies. Occupancy deviations. Debt yield comparisons across similar assets.
Â
As Goswami notes, lenders often lack "revenue and expense comparables," despite relying heavily on sales comps. Normalized underwriting data makes those comparisons feasible.
Â
Why Excel survives this transition
A notable aspect of the discussion is the deliberate decision not to replace Excel.
"Everyone has their own Excel model," Goswami says. "Our goal is to not change the process that customers do today."
Â
This is an important signal. Institutional underwriting is deeply embedded in spreadsheets, governance processes, and credit committees. Any tool that attempts to disrupt that layer faces adoption resistance.
Â
Clik.ai instead treats Excel as the destination, not the obstacle.
Â
Data flows into Excel through macros and integrations, preserving formulas, assumptions, and internal controls. This allows underwriting teams to move faster without losing transparency or auditability.
Â
What changes immediately for users
Â
When asked what changes the first time a user logs in, Goswami's answer is blunt. "A big time saver," he says. "Avoiding that mind-numbing work of inputting the data into Excel models."
Â
But there is a second-order effect. When analysts are no longer consumed by manual extraction, they can review more deals, ask better questions, and surface risks earlier.
Â
Teams score more deals, errors decline, and portfolio oversight becomes more consistent.
Â
Who this is most relevant for
This discussion is most relevant for:
- Institutional lenders and debt funds
- Loan servicers and asset managers
- Private equity platforms managing large portfolios
- Firms constrained by underwriting throughput rather than capital
- Acquisitions teams
Smaller acquisition-focused sponsors may find value, but the economics and onboarding effort make Clik.ai a better fit for scaled operations.
Â
Bottom line
The future of AI in commercial real estate underwriting is not about replacing judgment. It is about eliminating friction.
Â
Clik.ai demonstrates how automation at the data layer enables speed, scale, and consistency without forcing firms to abandon Excel or rewrite internal processes.
Â
In a market where deal flow is volatile and risk scrutiny is rising, the ability to underwrite faster and more accurately is becoming less of a competitive advantage and more of a baseline requirement.
Related to this episode:
