Hey everyone,
I need to tap into the collective wisdom here because I’m facing a challenge that I suspect others might be dealing with too.
The New Reality for CPAs in 2026
As of this year, I’m seeing a significant shift in client expectations at Thompson & Associates. Privacy audits are no longer a “nice-to-have”—they’re becoming an integral part of the financial audit process. Several of my clients (particularly those in California dealing with CPPA requirements) now expect me to assess their data security practices alongside their financial reporting.
This isn’t just informal advice anymore. I’m being asked to:
- Identify data security gaps that could affect financial reporting
- Assess compliance with GLBA (Gramm-Leach-Bliley Act) for financial institutions
- Verify adherence to state privacy laws
- Document data handling procedures
- Provide written privacy compliance opinions
The Plain Text Dilemma
Here’s where it gets interesting. Over the past few years, I’ve been quietly converting several clients to Beancount for their internal bookkeeping. The transparency, version control, and audit trail have been game-changers for my work.
But now I’m second-guessing this approach.
When I conduct privacy audits, I’m discovering that plain text accounting creates some unique compliance considerations:
The Good:
Complete version history via Git—I can see every change, who made it, when
Easy to audit access logs and repository permissions
Transparent data handling—nothing hidden in proprietary formats
Client owns their data completely, not stored on third-party servers
The Concerning:
Financial data stored in human-readable format vs. encrypted databases
How do I prove deletion compliance when Git history is immutable?
Access controls rely on repository permissions rather than application-level encryption
Regulatory expectations seem to assume “encrypted database” not “version-controlled text files”
A Specific Example
One of my California-based clients asked me point-blank: “Is my Beancount ledger GLBA-compliant?”
Honestly, I wasn’t sure how to answer. They keep their ledger in a private GitLab repository with SSH key authentication. By most practical measures, it’s secure. But when I try to map it to GLBA Safeguards Rule requirements (encryption at rest, access controls, audit logging), I’m finding that the terminology doesn’t quite fit.
The client wants Git for transparency. The regulators want encryption for security. How do we satisfy both?
The Core Question
I keep coming back to this: Is readable = auditable = compliant? Or is readable = exposed = risky?
Part of me thinks that plain text + proper repository security (private repos, SSH keys, encryption at rest on the host) is actually more secure than trusting QuickBooks Online with client credentials. At least with Git, I have complete visibility into who accessed what and when.
But another part of me worries that I’m rationalizing because I love the workflow, and that a regulatory examiner would take one look at .txt files and immediately cite a violation.
What I Need From This Community
Has anyone else dealt with privacy compliance requirements for plain text accounting?
Specifically:
- How are you handling GLBA or similar regulations?
- What security controls have you implemented?
- How do you document compliance for auditors or regulators?
- Have you had any actual regulatory examinations of your setup?
I’m not looking to abandon plain text accounting—I genuinely believe it’s superior for accuracy and transparency. But I need to figure out how to make it demonstrably compliant with 2026 privacy standards.
Any experiences, frameworks, or even cautionary tales would be incredibly helpful.
Thanks,
Alice