Methodology: how every number on this site gets verified
What exactly do we verify, and how often?
Software pricing (weekly, Mondays): we fetch each vendor’s public pricing page, extract plan names, monthly and annual prices, user limits, and per-seat fees, and store a snapshot. Each new snapshot is diffed against the previous one; any difference becomes a public entry in the change log with before and after values. Tracked vendors:
- Jobber — https://www.getjobber.com/pricing/
- Housecall Pro — https://www.housecallpro.com/pricing/
- ServiceTitan — https://www.servicetitan.com/pricing (quote-gated: tracked for structural changes; prices are crowdsourced — see below)
- Workiz — https://www.workiz.com/pricing/
- FieldPulse — https://www.fieldpulse.com/pricing
- Service Fusion — https://www.servicefusion.com/pricing
Material prices (weekly, Mondays): a fixed basket of 6 tracked SKUs (concrete mix, interior paint, architectural shingles, PVC, Romex, mulch) priced at Home Depot and Lowe’s store-level product pages across 20 metros — 240 price points per week. National medians are computed across metro medians.
How do the “verified as of” stamps work?
A stamp is the timestamp of the last run where both the fetch and the parse succeeded for that source. If a vendor redesigns their pricing page and our parser fails, the stamp does not advance — you’ll see an honestly-old date instead of a silently-wrong number. Parse failures alert us the same day; broken selectors get fixed, and the stamp only moves when real data flows again.
Weekly is the promise. We never claim “real-time” — nobody checking shelf prices hourly is telling the truth.
How is user-reported data handled?
ServiceTitan’s pricing is quote-gated, so published-price verification is impossible. Instead we aggregate real quotes: from public forum posts (cited) and reader submissions via the intake form on the ServiceTitan pricing page. User-reported figures are always displayed with a purple label and a sample size (n=…), rendered visually distinct from verified retail pricing, and never mixed into verified cells. Medians, not means, so one outlier quote can’t skew the number.
How do rankings stay independent of affiliate payout?
Four disclosed criteria, applied in order, for every best-of list and verdict:
- Trade fit — does the tool's feature set match how this trade actually works?
- Verified 5-tech cost — the like-for-like monthly price at a 5-person shop.
- Pricing transparency — published pricing ranks above quote-gated at equal fit.
- Contract terms — month-to-month availability ranks above annual lock-in.
Structural guarantees: every comparison matrix includes at least one vendor we have no affiliate relationship with (currently Service Fusion), affiliate links are disclosed on every page and in the disclosure, and a vendor’s payout never enters the ranking math. When a non-affiliate vendor wins a row, it wins the row.
What’s the current verification status?
Site status: pre-launch. The numbers currently shown carry the Aug 24, 2026 research-baseline stamp, compiled by hand from vendor pricing pages and public sources while the automated pipeline is stood up. From launch, this section reports every run publicly: run date, targets checked, successes, and failures. A week where 11 of 12 sources verified will say “11/12” — radical transparency is the product.
Corrections
Found a number that’s wrong? The change log is append-only — we don’t rewrite history, we append corrections with their own dates. Corrections land within one verification cycle of being confirmed.