The Ledger Stack

Stack guide · Bookkeeping

The complete tech stack for a solo bookkeeper

Every tool a solo bookkeeping practice actually needs — core ledger, receipt capture, client onboarding, payroll, and cleanup — with tested verdicts where we have run the books, and honest criteria where our reviews are still in the queue.

By Piper Tanaka Published Aug 10, 2026 Last verified Aug 19, 2026

Some links below are affiliate links — if you subscribe through one, we earn a commission at no extra cost to you. It never changes a verdict; every tool runs the same test books. Full disclosure

A solo bookkeeping practice runs on maybe six tools. The trap is that the industry will happily sell you sixteen. Every app in your stack is a monthly fee multiplied across clients, a login to manage, and one more thing that breaks during close week — so our rule for this guide is that every tool has to earn its seat, and our recommendations come in two honest flavors: verdicts from tools we’ve actually run our standard test books through, and — for the categories still in our review queue — the criteria that should drive your choice, so you can evaluate today rather than wait for us.

This page is the spine of The Ledger Stack: as each category gets the full test-dataset treatment, the criteria below get replaced with tested verdicts, and this guide gets re-verified. The date at the top is real — and every price this guide or any review quotes has a current, dated counterpart on the pricing tracker.

The stack at a glance

The layers, our status on each — updated as reviews publish
Layer Leading options Our testing status
Core ledger QuickBooks Online, Xero TESTED — full comparison below
Simple-client ledger FreshBooks TESTED — see our freelancer-client comparison
Receipt & document capture Dext, Hubdoc TESTED — Dext 26/32, Hubdoc 24/32; comparison below
Client document collection Content Snare In the test queue
Proposals & engagement letters Ignition In the test queue
Client payroll Gusto In the test queue
Cleanup & close checklist Double (formerly Keeper) In the test queue
Spreadsheets, security, file sharing See below Criteria only — commodity picks

The core ledger — tested

This is the one decision that shapes everything else, and the one we can speak to with evidence: we ran an identical quarter of deliberately messy books — planted duplicates, a refund, a feed gap — through both market leaders and scored them on a 32-point checklist. The full results are in QuickBooks Online vs Xero for a solo bookkeeping practice; the short version:

One stack-level fact from that review worth repeating here: Xero bundles Hubdoc (document capture) into its business plans — which means for some practices, choosing Xero quietly fills the next category for free.

A third option earned its own test since this guide first published: for freelancer and solo-service clients specifically — invoices out, cards in, no balance-sheet life — FreshBooks beat both platforms on duplicate detection and data-entry speed at nearly half the comparable price, while being structurally wrong for anyone with loans, equipment, or transfers. The boundary is the whole story: FreshBooks vs QuickBooks for bookkeepers with freelancer clients.

And the free end of the spectrum has now been through the same books: Wave’s $0 tier scored 22/32 — one point above QuickBooks — with penny-identical financials, real equity accounting, and the best refund handling of the four. Its boundary is sharper than FreshBooks’: the free tier structurally ends the moment a client grants their bookkeeper access, because every collaborator seat is paywalled. When a self-serve-Wave client hires you, that’s the moving day: when to move a client off Wave.

Receipt & document capture — tested

The shoebox problem: clients photograph receipts, forward invoices, and lose half of everything. Capture tools (Dext and Hubdoc lead this category) turn that stream into extracted, coded transactions pushed to the ledger.

Both have now been through our 15-document receipt gauntlet — staged photos, ugly scans, a German invoice, a duplicate pair — and the verdict came in: Dext, 26/32, the better machine (instant PDF extraction, and its Paperwork Match attaches documents to bank transactions the ledger already has instead of minting duplicate bills) against Hubdoc’s 24/32 with one decisive trump card — it’s free with every Xero plan. Both caught our duplicate receipt; both mangled something (Dext flips ambiguous US dates, Hubdoc surrendered on the foreign invoice). The stack advice survived testing intact: if you’re on Xero, start with the Hubdoc you already have, and move to Dext when capture volume makes its speed worth $23.92 per client.

Client document collection — in the test queue

Different problem from receipt capture: getting specific things out of clients — the bank statements, the loan agreement, the payroll reports — without twelve follow-up emails. Content Snare is the specialist here: checklist-driven requests with automatic reminders.

Criteria that matter: whether reminders actually move clients (the whole value proposition), effort to build reusable request templates (monthly close, onboarding, year-end), and whether the client-side experience needs a login (every login you impose on a client costs completion rate). The generic alternative — a shared folder and a spreadsheet checklist — is free and worth outgrowing deliberately, not preemptively.

Proposals & engagement letters — in the test queue

Ignition owns this category for accounting practices: proposal, engagement letter, e-signature, and automatic payment collection in one flow. The criteria: whether it genuinely ends scope creep (the engagement letter that clients actually read and sign), payment collection that doesn’t make you the collections department, and whether its cost pencils out below a certain client count — specialist proposal tools earn their seat at volume, not at client #3. Until then, a solid letter template and any e-signature tool cover the legal bases.

Client payroll — in the test queue

You don’t want to run payroll by hand; you want to resell a payroll service and manage it. Gusto is the name we hear most in this seat (and, full disclosure of a small irony: our own fictional test company’s books contain Gusto debit lines, because that’s what real small-business checking accounts look like). What the eventual review will weigh: multi-client dashboard quality, what the accountant-partner tier actually gives you, state tax registration handling, and the failure modes — what happens when a filing bounces.

Cleanup & close — in the test queue

Double — the bookkeeping close tool until recently named Keeper, not to be confused with the password manager of the same old name — structures the month-end: uncategorized-transaction review, client questions, file review, recurring close checklists. Our review will test it against the mess our dataset generates naturally — ten uncategorized transactions and an “ask the client” pile. Criterion zero: whether it beats the free alternative, which is a disciplined spreadsheet checklist and our own Ask-the-Client account convention.

The unglamorous connective tissue

Three seats in the stack are commodities — pick something reasonable and move on:

  • Spreadsheets. You have Excel or Google Sheets. Both are fine. The differentiator is your template discipline, not the brand.
  • A password manager is not optional. You hold credentials to clients’ financial systems; a breach at a solo practice is practice-ending. Any reputable manager (1Password, Bitwarden) with unique passwords per client system and two-factor everywhere clears the bar. This is the cheapest professional insurance you will ever buy.
  • Secure file delivery. Financial documents don’t belong in email attachments. A client portal (your capture/collection tools above often include one) or at minimum shared-folder links with access control.

What we deliberately leave out

No CRM at solo scale (your ledger’s client list and your inbox are the CRM). No time-tracking unless you bill hourly — and the case for fixed-fee is strong enough that we wrote it up: how to price bookkeeping services is a pricing decision before it’s a software one (and the pricing worksheet now does that article’s arithmetic live). No AI bookkeeping add-ons yet: our core-ledger test found the built-in automation both impressive and dangerous in equal measure — we’d rather test the add-ons against the same books before recommending any.

The honest starting stack

If you’re launching a practice this month, on our evidence so far: Xero as the default ledger (or QuickBooks where your client base demands it), the Hubdoc you get free with Xero for capture, a shared-folder checklist for document collection until reminders become your bottleneck, a password manager today, and patience on the specialist tools — let client count justify each seat. That’s roughly $25–55/month of software to run a real practice, with every upgrade path mapped above and each one getting the test-dataset treatment here as our review queue works through it.

Every tool named here will eventually face the same books. Affiliate links, where they exist, are disclosed and never touch a verdict — several tools above we have no relationship with at all. The test dataset behind our verdicts is public.

Filed under: tech stack · practice tools · quickbooks · xero · pillar