Trullion tackles applying AI to accounting workflows


We’re excited to bring Transform 2022 back in person on July 19 and virtually July 20 – August 3. Join AI and data leaders for insightful conversations and exciting networking opportunities. Learn more

Most of us don’t think of accounting as a separate industry. But accounting is big business in itself.

The global accounting services market is expected to grow from $544 billion in 2020 to $573 billion in 2021 at a compound annual growth rate (CAGR) of 5.4%, according to Research and Markets’ Global Accounting Services Market Report. The market is expected to reach $736 billion by 2025 at a CAGR of 6%.

This is a sizeable market and a target for digital transformation. The accounting software market size is expected to reach $22.9 billion in 2027, from $14.2 billion in 2020, growing at a CAGR of 7.1% during the forecast period 2021 to 2027, according to Valuates Reports.

The account software market includes companies such as Microsoft, Oracle, and SAP to Intuit and Xero, plus many others. Still, there’s something that none of these names fit, says Isaac Heller: AI in accounting. That is why Isaac Heller and Amir Bolto co-founded Trullion in 2019.

Trullion seeks to “unify the unstructured and structured worlds of accounting by reading PDF or Excel-based contracts and translating them into financial workflows, connected to the data source.”

Trullion has more than 100 clients, from pre-IPO to multinational Fortune 500s and audit firms that source directly for their clients, and has crossed the seven-digit recurring revenue mark.

In February, Trullion announced it was closing $15 million in Series A financing co-led by Aleph and Third Point Ventures, with the participation of existing investors Greycroft and veteran financial executives, including Bob Mylod, Artie Minson, Jody Padar and Guzel Lumpkin.

Heller met with VentureBeat to discuss Trullion’s use of AI to innovate in the accounting software market.

Accounting is disconnected

Heller lives in the world of finance and accounting. After earning an MBA and working as a teaching assistant and visiting lecturer at the University of Texas at Dallas, he learned about generally accepted accounting principles (GAAP) and how those great standards are used. After that, he made his way into the corporate world.

Here he encountered more challenging standards and business models – things like revenue recognition, revenue accounting and lease accounting. After a stint with Visual Lease, a lease optimization software provider, Heller was inspired to do the same for accounting. Heller describes accounting as three parts that are disconnected:

The first part is the world of corporate accounting. They may be running some of their processes in spreadsheets or offline. The second part is ERP — SAP, Oracle and WorkDay. And then the third part is the audit leaders and the audit team.

There are many reasons why the ecosystem is becoming more manual and risky. “The question is: how could you make it more accurate and transparent? And the way to do that, in our view, would be to find ways to connect those three parts together, meaning everyone in all parties could connect to one source of truth,” Heller said.

“But to do that, if you tried to come out with a heavy tool that did everything, it wouldn’t work. Or you would have to spend $100 million to get started. So what we do is focus on a pain point. In this case, we focus on things like lease accounting and revenue recognition,” said Heller.

Accounting and finance teams are still tied to legacy processes, including manual data entry, spreadsheets, and PDFs, meaning they are at high risk for critical and costly errors. Meanwhile, continuously evolving standards — including new lease accounting standards — have added an estimated $2 trillion in lease liabilities to S&P 500 balance sheets, further adding to the complexity and potential inaccuracy of data.

Trullion emphasizes changing standards as a way to raise awareness, drive innovation and create a sense of urgency. Heller compared esoteric accounting standards such as ASC 842 and IFRS 16 to the GDPR in that regard. The end goal, he said, is greater balance sheet transparency.

Diving deeper into ASC 842 reveals the reasons why Trullion thinks it’s a good idea to apply AI to accounting, especially lease accounting. ASC 842 defines leases as contracts, or portions of contracts, that grant “control” over an identifiable asset for a period of time in exchange for payment.

Trullion notes that this means contracts must be scanned carefully to identify leases as per the definition above. That also includes contracts that are not necessarily about leasing. That poses an interesting problem, where AI can lend a helping hand.

Applying AI to Accounting

Before we get to AI, there are a lot of data pipelines and data stores that need to be in place — in other words, creating the “source of truth” that Heller was referring to.

Often, when you rely on an ERP or an old database as a source of truth, it doesn’t contain any of the actual source, Heller said. With source-based accounting, that means more than PDF documents, he said. It can also be Excel files, or systems like Salesforce or QuickBooks. Trullion takes data from all those sources and normalizes it into one source-based accounting system.

Of course it’s not as easy as it sounds. Relevant data from all those sources must be parsed. In many cases, that includes scanned handwritten documents, which is why Trullion has been working to improve its OCR. But identifying lease definitions in a contract, for example, requires advanced technology.

The Trullion white paper mentions the name NER (Entity Recognition), a technique used to identify specific instances of interest in data sources. To build this, and even normalize ingested data into a single source of truth, a domain entity model is needed. The Trullion white paper refers to combining deep learning with domain-specific ontologies.

In fact, as Trullion co-founder and CTO Amir Bolto confirmed, it takes a lot more than NER to identify leases in contracts. Bolto referred to pre-processing, which aids identification by passing the data through multiple steps such as data cleansing, data integration, data transformation, and more before going through NER.

Bolto said this helps improve accuracy and reduce errors in the final result, even though there are currently no publicly available performance stats or ontologies to share. Trullion’s solution for identifying leases in contracts has been in development for about a year.

“There are a lot of start-ups that make sales software, marketing software and customer support software, and those are use cases that we all have experience with in startup companies. But bookkeeping – that’s big business. Those are the mature companies. It’s not very well represented with startups,” Heller said.

There are some use cases where AI is applied, such as accounts payable and invoice automation that are more common with us. “I’m not sure why, but we just haven’t seen much AI. We saw a lot of people talking about it. People wrote very nice papers about it, but we didn’t see it applied in practice. It’s definitely harder than anyone thinks,’ Heller said.

Trullion does not claim to wave a magic wand and extract 100% of a document. But if you can connect to the actual finance workflow and get something like a 70% improvement, that’s still a big deal, Heller said.

Heller said what Trullion does is complementary to solutions like QuickBooks, Xero and ERP software, in that lease accounting is a hole in their ecosystem. The difference with more specialized suppliers is the use of AI. Trullion, which has a presence in New York and Tel Aviv, Israel, will use its Series A funds to hire people for various technical, sales and marketing roles, as well as accelerate product development.

VentureBeat’s mission is to be a digital city square for tech decision makers to learn about transformative business technology and transactions. Learn more

Leave a Reply

Your email address will not be published. Required fields are marked *