Fava vs hledger-web: Is the Web Interface Battle Over in 2026?

General
1

I’ve been running my bookkeeping practice for 10 years now, serving 20+ small business clients, and I’ve been gradually migrating them all to Beancount over the past two years. The biggest selling point? Fava’s web interface. Clients love the interactive dashboards, the clean charts, and the intuitive filtering. It feels professional in a way that command-line tools never will.

But here’s the thing: I keep hearing hledger users talk about their tool’s flexibility, and part of me wonders if I’m missing out. In 2026, with new players like Surebeans (the hledger-compatible YNAB clone) entering the space, I’m questioning whether the “web interface battle” between Fava and hledger-web is actually settled—or if I’ve just picked a side too early.

The Fava Advantage: Polish That Clients Actually Notice

When I show clients Fava for the first time, I watch their eyes light up. The balance sheet visualization, the account drilldowns, the expense charts—it all just works. They can log in at any time, see their financial position, and understand what’s happening without calling me for explanations.

This matters more than I initially realized. One client actually switched TO my practice from another bookkeeper specifically because “your system looks more professional.” That’s a direct business impact from choosing Beancount+Fava.

The polish isn’t just aesthetic—it’s functional. Fava’s query interface lets clients filter their transactions in real-time without knowing the command line. The document linking feature means receipts are one click away. The context-aware suggestions make data entry faster.

The hledger Temptation: Flexibility I Don’t Fully Understand

But when I talk to hledger users, they mention things that make me curious:

  • Free-form syntax: hledger doesn’t require pre-declaring accounts or inferring balancing entries the same way Beancount does. For complex transactions, this apparently saves time.
  • More forgiving workflows: hledger users say it’s easier to start small and add structure later, whereas Beancount forces structure from day one.
  • Command-line power: hledger’s querying is supposedly more flexible, though I’ve never hit Beancount’s limits personally.

I’ve looked at hledger-web, and it’s… minimal. It works, but it doesn’t have the visual polish that Fava does. For my client-facing needs, that’s a dealbreaker. But for my own personal finances? I’m honestly not sure I need Fava’s polish.

The 2026 Wild Card: Surebeans and New Entrants

Then there’s Surebeans, which launched in 2026 as an hledger-compatible budgeting app. It’s closed-source and cross-platform (C#), but it offers a YNAB-like experience on top of plain text accounting. I haven’t tried it yet, but the fact that new tools are still entering this space tells me the “interface problem” isn’t solved.

Are we going to see more specialized interfaces built on top of Beancount or hledger? Will Fava evolve to stay ahead? Or will the plain text accounting world fracture into tool-specific ecosystems?

The Core Question: Does Interface Determine the Tool?

Here’s what I want to discuss with this community:

For professional bookkeepers serving clients: Is the web interface THE deciding factor when choosing between Beancount and hledger? Or are there workflows where hledger’s flexibility outweighs Fava’s polish?

For developers and power users: Do you actually use the web interface, or is it mostly CLI? Would you switch tools for a better web UI, or is that secondary to scripting power?

For everyone: How do you balance client-facing polish with backend flexibility? Is Fava “good enough” that you never wish for more? Or do you find yourself building custom dashboards anyway?

My Current Stance (But I’m Listening)

Right now, I’m sticking with Beancount+Fava because:

  1. Clients love it, and that directly impacts my business
  2. I haven’t personally hit any limitations in Beancount’s syntax
  3. Fava’s polish saves me time in client education and support

But I’m genuinely curious: Am I leaving power on the table by not exploring hledger? Or is this just “grass is greener” thinking?

For those of you who’ve used both, or who’ve switched between them—what was the deciding factor? Was it the interface, the syntax, the ecosystem, or something else entirely?

Let’s hear your experiences. The web interface matters to me, but I don’t want to blind myself to better workflows just because Fava is pretty.

2

Fred, I think you’re asking exactly the right questions. As a bookkeeper serving clients, I can tell you: yes, the web interface is absolutely a deciding factor for my business, but maybe not for the reasons you think.

The Client Reality Check

Here’s the thing about client work: they’re not just buying your accounting expertise—they’re buying confidence. When a client logs into Fava and sees beautiful, professional-looking dashboards, they trust the numbers more. Is this rational? Probably not. Do the numbers change based on how pretty the interface is? Of course not.

But I’ve learned over 10 years that client perception IS reality. I had one restaurant owner tell me he finally understood his cash flow for the first time because of Fava’s visualization. The underlying data was the same as what his previous bookkeeper provided (in Excel!), but seeing it in Fava’s interface made it “click” for him.

When I Actually Use the CLI (Spoiler: More Than I’d Admit to Clients)

That said, I do use bean-query and the command line for my own analysis work. When I’m building month-end reports or tracking down discrepancies, I’m in the terminal. Fava is great for exploration and client-facing work, but for serious querying? The CLI is faster.

I’ve also built custom Python scripts that read my Beancount ledgers directly, bypassing Fava entirely. For automating reports that Fava doesn’t support natively, plain text + scripting is unbeatable.

So to answer your question about “do clients actually need fancy dashboards”—no, they don’t need them. But they want them, and in a service business, that distinction doesn’t matter. If Fava helps me retain clients and win new ones, it’s worth whatever flexibility I’m sacrificing vs. hledger.

What I’m Curious About (The hledger Question)

That said, your mention of hledger’s flexibility has me thinking. I’ve never tried it seriously because hledger-web looks too basic for client work. But what about using it for my own personal finances?

I don’t need to impress myself with a polished UI. If hledger’s syntax really is more flexible for complex scenarios, maybe there’s value in learning both tools—Beancount+Fava for client work, hledger for personal experimentation and edge cases.

Has anyone here actually done this? Used different tools for different purposes? Or does context-switching between tools create more problems than it solves?