Numbers don’t lie. But for modern, high-volume businesses like D2C brands, e-commerce marketplaces, or multi-location F&B chains, financial numbers often don’t agree. At every “month-end close”, their finance team takes up the unenviable task of staring at multiple screens, trying to tally and reconcile the ghosts of transactions past while the rest of the company races ahead.
For instance, the company’s order system may display a million dollars in monthly sales. But the payment gateway report confirms only a deposit of $950,000. The logistics partner’s invoice may claim fees on orders the customer says were never delivered. Meanwhile, Amazon’s sales report is a complex puzzle of commissions, FBA fees, and marketing deductions. All the while, the company’s bank statement, the ultimate source of truth, shows deposits that match none of these figures.
The truth is in there, somewhere, buried under hundreds of thousands of rows of transactional data. But finding it is a brutal and soul-crushing exercise in digital archaeology. The finance team is left to bridge these gaps and reconcile the accounts manually, a Sisyphean task that is slow, error-prone, and a massive drain on resources that could be used for strategic analysis.
This is the Gordian knot of modern commerce. Revenue leakages go unnoticed, chargebacks are missed, and the simple question, “Did we pay and get paid the right amount?” becomes agonizingly difficult to answer.
For Ankit Narsaria, the founder of ZenStatement, this is a puzzle he has spent his career wrestling with, and is now decisively cutting through. His company, ZenStatement, is an ambitious endeavour to build an AI-powered Finance Operating System, designed to bring order, clarity, and automation to the financial chaos of high-volume businesses.
Founded in 2023 by Ankit and Sourabh Nolkha, ZenStatement recently raised $1.62 million in a seed round led by 3One4 Capital and Boldcap VC. For founders and fintech leaders, Ankit’s story building ZenStatement is a lesson in tackling a fundamental operational challenge in finance with a truly modern, AI-native approach.
A Natural Founder-Market Fit
While Ankit’s resume includes a B.Tech and an MBA from IIM, his real education came from building products on the front lines. His journey began at the women’s apparel e-commerce platform, Voonik, where he built operational systems from scratch and learned to combat revenue leakage from complex logistics and payment failures.
This hands-on experience deepened in the B2B fintech space, where he was instrumental in building the entire UPI reconciliation platform for JustPay, a system designed to handle the colossal transaction volumes of India’s top retail giants.
But the catalyst for ZenStatement came during his time at ShopUp, a company with intertwined logistics, commerce, and fintech verticals. There, he worked directly with the finance department, witnessing their daily struggle firsthand.
"In JustPay, the data was flowing across multiple systems. I noticed that attributing a single payment back to a specific order was incredibly difficult."
He saw a universal pattern here. Finance teams were under-resourced and drowning in a level of data complexity that legacy tools were never designed to handle. He realized this wasn’t a niche issue but a fundamental, unsolved problem for modern commerce.
This insight, combined with the expertise of his co-founder, Sourabh Nolkha, a 16-year finance veteran, created the perfect founder-market fit. Ankit was the builder who had felt the pain, and Saurav was the finance operator who had lived it. Together, they set out to build the solution.
The Problems Solved by ZenStatement
ZenStatement’s ideal customers are businesses grossing over ₹50 crore annually and processing more than 50,000 transactions a month. Their clients are businesses like The Good Glamm Group, Pine Labs, and MPL, companies operating at a scale.
Ankit breaks down the core challenges that his platform is built to address in these businesses where the financial landscape is a minefield:
1. Multiple, Disparate Revenue Channels: A modern D2C brand like The [Pant] Pro-ject or a giant like The Good Glamm Group doesn’t just sell from its own website. They sell on Amazon, Flipkart, Myntra, and now, even through quick commerce platforms. Each channel has its own fee structure, payment cycle, and reporting format.
2. Complex Financial Transactions: An order is never just an order. “If I am purchasing a product from Amazon for ₹100, Amazon will not give ₹100 to the seller. They might give ₹50. The remaining ₹50 is a mix of commission, logistics fees, technology fees, and other XYZ charges,” Ankit illustrates. Manually untangling these deductions for thousands of orders is a nightmare.
3. Fragmented Systems & Data Integrity Issues: The order originates in an Order Management System (OMS). The payment is processed by a payment gateway. The delivery is handled by a logistics partner. The final settlement appears in a bank statement. As data moves between these systems, the data’s sanctity is often lost. A unique identifier in the OMS may not exist in the bank statement, making a direct match impossible.
4. The Limits of Existing Tools: Traditional ERPs and accounting software are essential for high-level accounting, but they might not be suitable for this granular, order-level reconciliation. “If I want to track my receivables at an order level in an accounting platform, I cannot do it today. Everything goes in a summarized way,” Ankit notes.
And spreadsheets? “If I’m trying to analyze a data of five lakh rows in Excel and put in some formulas, Excel will start hanging, whatever laptop you’re using.”
These four core problems in financial operations – multiple channels, complex transactions, fragmented data, and inadequate tools – create a perfect storm. The result is a finance team bogged down in manual work, an inability to get a real-time view of cash flow, and significant, often hidden, revenue leakage. This is the area where ZenStatement excels.
The First PoC: From Python Script to Platform in Four Months
ZenStatement’s journey from concept to product was a testament to the principles of lean startup methodology, supercharged by deep domain expertise. In September 2023, with no platform and no developers, Ankit and his co-founder approached gaming giant MPL, a company processing millions of transactions.
“MPL shared almost two to three months of data with us. It was millions and millions of transactions. An Excel sheet would have crashed just trying to open the file,” Ankit says.
Using their proprietary R and Python scripts, Ankit’s team processed the massive datasets. Their MVP at this stage wasn’t a flashy UI. The singular focus was on accuracy. They had to process and match this massive transaction data with more than 99.99999% accuracy. For a finance team, a 0.01% error on millions of transactions can translate into a huge financial discrepancy. Within a month, they presented their findings to MPL.
MPL’s reaction was ZenStatement’s first major validation. “The partner MPL was working with at the time took six months to arrive at the same first-cut result. It was a proof point for us that whatever we are solving for, and whatever our thought process is, it is working.” Ankit says.
This early win defined ZenStatement’s DNA. They needed to build a platform that was both incredibly fast to deploy and robust enough to handle immense scale from day one. This is a dichotomy that plagues many startups, forcing a choice between velocity and scalability. ZenStatement needed both.
Building for Velocity, Scale, and Intelligence
“Today, our platform is handling more than 100 different use cases across payment gateways, banks, marketplaces, retail, food and beverages, international payment gateways, warehouse management systems… You just name it, and our platform can handle it,” says Ankit.
The secret to this rapid expansion lies in the platform’s core architecture. ZenStatement has a foundational layer of reusable, configurable logic that can be used for various financial data use cases. Ankit compares the versatility to Excel, the most ubiquitous finance tool of all.
“When you use Excel, it has different functions like VLOOKUP, XLOOKUP, MATCH, and INDEX. What we have done is, we have made our system in such a way that we are using all these different functions inside our modules. You can just drag and drop, or I can just write the function, and the function does all the heavy lifting,” he explains.
“The foundation is built in such a way that if any use case, however complex it is, comes to us, we can enable that use case in less than a day,” Ankit claims.
The User’s Journey - Designed With a “Finance-First” Mindset
For all its back-end complexity, the ZenStatement platform is designed for the finance user, not the engineer. The interface is clean and intuitive, abstracting away from the backend heavy lifting.
The journey begins with data ingestion. A client can upload raw financial data files (the platform can process a single 40GB file in under 10 minutes), or set up automated feeds via SFTP (Secure File Transfer Protocol), email, or APIs. The platform then takes over, automatically identifying the file type, transforming the data into a standardized format, and running the pre-configured reconciliation rules.
Within minutes, the user can see the results. Reconciled transactions are cleared, while discrepancies are flagged as “open items” in a dedicated dashboard. Here, a built-in workflow allows teams to investigate issues, assign tasks, and track resolutions.
Ankit highlights that the platform requires raw financial data for accuracy. By taking raw, unmanipulated data files directly from banks, marketplaces, and internal systems, ZenStatement becomes the single source of truth, eliminating the “garbage in, garbage out” problem that plagues so many systems. At the end of every process, a “completeness check” runs to ensure every number reported is 100% accurate.
“This is something I learned during my time working in MatchMove,” Ankit reflects. “We were doing RBI reporting. Once I sent the report, I could not change those numbers. The same kind of philosophy is used in building this product.”
The platform also features customizable dashboards tailored to different user personas. An AR executive might see a detailed breakdown of outstanding invoices, while a CFO gets a high-level view of cash flow, revenue leakage trends, and overall financial health.
Perhaps the most significant feature ZenStatement is building is not technical, but philosophical.
Ankit says, “Finance is a very conservative vertical. We are not coming in and changing the way the finance teams work. We tell them, ‘You continue using your Tally or Zoho Books. We will not touch your internal processes.’ We sit on top of their existing workflows.”
“We are using AI and technology to enable finance teams, not replace them. We are taking the 30% of their time spent on data crunching and making it zero. That is the biggest value proposition of ZenStatement,” says Ankit.
The AI-Native Advantage: From Bolted-On to Built-In
While many legacy SaaS companies are now trying to “bolt on” AI features, ZenStatement was born in the AI era, giving them the opportunity to weave it into their product development journey. But Ankit’s approach to AI is pragmatic and multi-faceted, focusing on tangible benefits rather than hype.
Ankit details ZenStatement’s three-pronged AI strategy:
1. Supercharging Development: The first application of AI was internal. They have a budget for AI tools for every employee. The team uses AI for code generation, writing unit tests, and, crucially, rapid prototyping.
“If a client asks about a feature that’s six months down the road, we can build a workable prototype in a couple of days using AI tools and show it to them. This validates our approach and builds immense confidence.”
2. Intelligent Data Ingestion: The platform uses AI models from OpenAI and Claude to run a “sanity check” on raw data even before it’s ingested. This is where AI tackles the messiness of real-world data.
“Take a date format like ‘04-05-2025’,” Ankit illustrates. “Is that May 4th or April 5th? Our date function is a self-evolving AI model that analyzes patterns within that file and across all files for that client to determine the correct format with a high degree of confidence.”
“A finance team member will be able to talk to their data in plain English. They can ask, ‘Show me all orders from the last quarter where the payment gateway fees were higher than 2%’ or ‘What’s my outstanding receivables from Amazon for the month of May?’ Our AI agent will understand the query, pull the data, perform the analysis, and present the answer. This eliminates the dependency on a dedicated data engineering or BI team for ad-hoc analysis.”
Go-to-Marketing Gyan for Fintech Players
“The conventional GTM strategy, like cold calling, emailing, or LinkedIn campaigns, is not very fruitful for us,” he says.
Instead, ZenStatement is pursuing a more nuanced, relationship-driven strategy:
1. Curated, High-Value Events: They are launching small, intimate “breakfast events” for 10-15 finance leaders to discuss their real-world problems, creating a community and establishing ZenStatement as a thought leader.
2. Strategic Partnerships: Recognizing that many companies outsource financial work, they are partnering with accounting and CPA firms. They are also working on a partnership with one of India’s top accounting platforms, similar to the QuickBooks-Amazon collaboration, to get their solution in front of the right users at the right time.
3. A Long-Term Content Play: They are building a deep repository of content that addresses the complex, evergreen problems of finance. The goal is to become the go-to resource for any finance professional trying to understand the intricacies of, for example, marketplace reconciliation.
This unconventional approach demonstrates a deep understanding of their target persona and a commitment to building trust over chasing vanity metrics.
ZenStatement's The North Star(s)
As Ankit looks to their next phase of growth and a future Series A fundraise, his focus is crystal clear and hinges not on one, but two important metrics that he says are his North Star(s).
The first is transaction volume.
“I care about the number of transactions my system is processing. It’s the ultimate proof of our scalability and shows that we can compete with players much bigger than us.”
Currently processing over 100 million transactions a month, this number is their benchmark for technical prowess.
The second is velocity and feature adoption.
“We are launching features every week, but we track their usage obsessively. We don’t want to be a feature warehouse.”
