A federal jury has convicted Charlie Javice and Olivier Amar, executives at student financial aid platform Frank, for orchestrating a scheme that misled JPMorgan Chase into purchasing their startup for $175 million. The jury found that the two fabricated customer data to inflate the platform’s user base, misrepresenting the number of legitimate users by a factor of 10. The fraudulent data played a central role in persuading JPMorgan to proceed with the acquisition, which was aimed at expanding the bank’s access to college-aged clients. The case has spotlighted both the defendants’ actions and shortcomings in JPMorgan’s due diligence process during the deal.
How did Frank manipulate user data to deceive JPMorgan?
What role did internal whistleblowers and technical evidence play in the trial?
Javice and Amar reportedly hired a data scientist to fabricate a synthetic list of 4 million users, although Frank only had around 300,000 actual customers. After failing to convince Frank’s own head of engineering to create the list, Javice paid a mathematics professor $18,000 to generate the fake dataset. The doctored file was then handed over to a third-party verification company, which, relying on the volume of data fields rather than content validation, mistakenly confirmed the inflated customer count to JPMorgan.
Frank’s internal resistance to the scheme was documented during the trial. The platform’s head of engineering testified that he had refused to assist in creating fraudulent data. Despite this, Javice continued by outsourcing the task to an academic, reinforcing the deliberate nature of the deception. Prosecutors pointed to internal messages and transactions to demonstrate the planning behind the fraud.
Bank executives stated that they heavily relied on Javice’s direct assurances about the platform’s user base when finalizing the acquisition. Although some JPMorgan employees raised doubts, the deal proceeded. After closing the acquisition, the bank’s attempt to market services to Frank’s supposed user base revealed significant discrepancies, with only 28% of emails being successfully delivered.
“The acquisition depended largely on data representations made by Frank’s leadership,” said one JPMorgan executive during court testimony.
Javice’s legal team argued that the trial was improperly influenced, contending, “Defense attorneys for Olivier Amar operated as a second prosecutor, undermining a fair trial for our client.”
Similar concerns over Frank’s user metrics were voiced by JPMorgan soon after the acquisition in 2021, leading to a lawsuit against Javice by the bank in 2023. After regulatory scrutiny followed, additional charges were brought against her. The case has remained in the public and legal spotlight for over two years, gradually uncovering deeper layers of manipulation and revealing systemic vulnerabilities in high-stakes tech acquisitions.
Javice is expected to appeal the verdict, challenging the fairness of the proceedings. If the conviction stands, both her and Amar face potential decades-long prison sentences for their roles in one of the most notable startup acquisition frauds in recent years. The outcome of the appeal may also influence how corporate fraud cases involving tech startups are prosecuted moving forward.
Investor reliance on startup-provided data has become a growing risk factor in acquisition decisions. The Frank case underlines the importance of in-depth technical audits, especially for platforms claiming large-scale user bases. Institutions may now reexamine internal acquisition protocols to better detect misrepresented metrics. For entrepreneurs, the incident serves as a cautionary tale about the legal consequences of misrepresenting company performance to secure investment or acquisition deals.
