How we test
Every review on this site rests on one method: identical sample books, run through every tool. Not feature-page summaries, not vibes from a demo — the same messy quarter of a small business's finances, imported into a real trial account of each product, scored on the same checklist.
The client: Cedarline Landscaping LLC
Our test books belong to a fictional three-person landscaping company — an owner and two crew running weekly mowing routes, patio builds, and sprinkler installs. It's exactly the kind of client a solo bookkeeper carries: real revenue, thin process, books assembled from bank feeds at month-end. (Cedarline is entirely invented; so is every customer and employee in its books.)
- A full quarter: April–June 2026, so month-over-month reports mean something
- 162 transactions across a business checking account and a business credit card
- 18 customer invoices, three left unpaid so receivables aging has something to catch
- Biweekly payroll through a payroll service, an equipment loan, an owner's draw
The traps
Real client books are messy, so Cedarline's books are deliberately messy. Five traps are planted, and how each tool handles them is scored, with screenshots:
- A duplicated charge — the same $342.17 hardware-store purchase posted twice by the bank. Does import catch it?
- A refunded deposit — $1,500 in, then back out a job cancellation later. Does income get counted twice?
- An uncategorizable vendor — a cryptic descriptor no category fits. Does the tool ask, guess loudly, or guess silently?
- A foreign-currency charge — a EUR parts order plus its FX fee.
- A two-week bank-feed gap — the checking feed simply stops mid-June. Does reconciliation surface the hole, and how painful is backfilling from a statement?
Scoring
Each tool works through a fixed 16-item checklist — import quality, the five traps, everyday workflows (auto-categorization, transfer matching, bank rules), and reporting — scored 0–2 per item, 32 points total. Two rules keep it honest: every score needs a screenshot or a written observation, and the checklist is filled during the trial, not reconstructed afterward. Reviews quote scores and show the evidence.
Prices are re-checked on the date shown as "last verified" on each article — what a tool actually charges at checkout, not what the pricing page implies.
What we don't do
- We don't claim practitioner experience. Working bookkeepers' opinions appear as attributed quotes, sourced from communities and interviews.
- We don't take payment for coverage or verdicts, and no vendor previews a review.
- We don't give investment, tax, or licensed financial advice — reviews cover tools and process.
Run it yourself
The dataset is public. Import it into any tool's free trial and check our work:
- cedarline-master.csv — Master ledger — every transaction, both accounts, traps annotated
- cedarline-invoices.csv — 18 customer invoices (A/R)
- cedarline-qbo-checking.csv — Checking feed, QuickBooks import format
- cedarline-qbo-creditcard.csv — Card feed, QuickBooks import format
- cedarline-xero-checking.csv — Checking feed, Xero import format
- cedarline-xero-creditcard.csv — Card feed, Xero import format
- cedarline-statement-jun-gap.csv — The missing feed weeks, as the backfill statement
Questions
Why should I trust a site that earns affiliate commissions?
You shouldn’t trust claims — you should check evidence. Every verdict traces to the same public dataset, scored on a published checklist, with screenshots. Commissions are disclosed on every page that carries a link, and we publish negative verdicts, including about tools that pay us.
Are you bookkeepers or accountants?
No — and we never claim to be. We are testers: we run the software the way a lab runs hardware. Where a review needs the practitioner’s view, we quote working bookkeepers and cite where the quote came from.
Do vendors pay for reviews or see them early?
No. We accept trial access, never payment for coverage. No vendor sees a review before you do.
Can I run the same test myself?
Yes — the full dataset is downloadable below. Import it into any tool’s free trial and compare your results to ours.
Something we should be testing that we aren't? Tell us.