Comparison Methodology
Updated August 6, 2026 · Applies to all scored comparisons, currently the HOA software comparison · August 6, 2026: the independence statement below was rewritten to name our two active commercial relationships; the rubric, criteria and weights are unchanged since July 3, 2026
The point of a published methodology is that you can check our work and predict our behavior. Criteria and weights are fixed before any product is scored; if we ever change them, the change and its reason get documented here with a date.
Who we score for
One reader: the volunteer board of a self-managed HOA or condo association, roughly 5–250 units, where a rotating, unpaid treasurer has to actually run the tool. Products excellent for professional management companies can score poorly here, that’s the rubric working, not a bug.
The rubric (v1, adopted 2026-07-03, before scoring)
| Criterion | Weight | 5 looks like | 1 looks like |
|---|---|---|---|
| Self-managed fit | 25% | Volunteer boards are the design center: dedicated self-managed positioning, no manager-scale minimums, features a board operates directly | Sold to management companies; boards are an afterthought or excluded by minimums |
| Core accounting | 20% | Native GL, invoicing/AR, bank sync/reconciliation, budgeting, tax-form support | No native accounting; money handled elsewhere |
| Price transparency & value | 15% | Full pricing published including usage fees; fair total cost at 5–250 units | Quote-only with undisclosed minimums and implementation fees |
| Resident portal & payments | 15% | Owner portal, ACH/card payments, communication residents actually use | No owner-facing layer |
| Ease for volunteers | 15% | Self-serve trial without a credit card, no long lock-in, survivable setup for a rotating treasurer | Demo-gated sales, annual auto-renewing contracts, professional-grade learning curve |
| Support & continuity signals | 10% | Documented onboarding/support, company scale and stability signals | Minimal support footprint or unclear organization behind the product |
Each product gets 1–5 per criterion. Weighted total = Σ(score × weight), maximum 5.00. Ties break toward the higher self-managed-fit score. The full score matrix is published in the comparison table itself, so every total can be recomputed by hand.
Data sources, and the honesty label
- Scores are currently based on vendor documentation and public sources as of July 3, 2026, not hands-on testing. Every comparison page carries this label. When we complete hands-on trials, tested products will be labeled as tested and the methodology will gain trial-based criteria (setup time, data export quality).
- Every price traces to the vendor’s own published pricing page, with the URL linked and the verification date shown per row. Facts a vendor does not publish are reported as “not public,” never estimated. Third-party price claims that don’t appear on the vendor’s site are excluded or explicitly flagged as unverified.
- Vendor selection: we scored the platforms in our research database that plausibly serve a 5–250-unit volunteer board. Manager-only platforms are listed as unranked reference points with the reason stated.
Rescoring cadence
- Quarterly: pricing and program facts on commercial pages are rechecked; each recheck updates the visible page date.
- Annually, or on material change: the full rubric is rescored, sooner if a vendor materially changes product, pricing, or terms, or if a reader-reported error survives verification.
- Material scoring changes are noted on the affected page with the date and reason.
Money and scores never touch
As of August 6, 2026, CommonKeel has no active affiliate, referral, sponsorship, or other paid relationship with any scored vendor. That is a narrower statement than "no relationships at all", so here is the whole picture: we have two active commercial relationships, a QuickBooks (Intuit) affiliate arrangement and a Google Workspace referral arrangement, both listed on our disclosure page. Neither vendor is scored. QuickBooks appears only in the unranked "also in the market" section of the software comparison, where it is disclosed inline, and Google Workspace appears in no scored comparison at all. Any future relationship with a scored vendor would be disclosed on every affected page, and it could not move a score, because the rubric inputs (published facts, weights fixed in advance) leave no discretionary lever to sell. If you ever find a discrepancy between this promise and a page, report it.
Calculator methodology
The reserve contribution calculator publishes its full model, straight-line full-funding formulas, allocation assumptions, interest treatment, and hand-verified test cases, directly on the tool page and in its page source, so the arithmetic is auditable.
Related: Editorial standards · Disclosure · See the rubric in action