recorded controls out of 144 across the three comparable product records.
Original evidence snapshot / 2 Sep 2026
Nine products audited. Three comparable test records. No rankings disguised as evidence.
This release turns the public evidence ledger into a downloadable, reproducible coverage baseline. It measures what ServiceTech Signal has verified, not product quality, customer outcomes, or the field-service market.
EVIDENCE COVERAGE / 2 SEP5 / 9
products have bounded direct evidence
3use the same 48-control protocol
Coverage, not a leaderboard
Direct evidence exists for five products, but only three records are structurally comparable.
Jobber, ServiceM8, and Tradify each use the same S01-S08, 48-control protocol. Together they contain 144 recorded controls: 10 Pass, 69 Conditional, zero Fail, 64 Blocked, and one Not run. The large blocked share is a warning about evidence coverage, not a verdict about any product.
Nine-record population
Every product in the public ledger is counted once.
| Evidence state | Products | Counting rule |
|---|---|---|
| Direct | 5 | At least one bounded first-party observation; scope varies by product. |
| Documented | 2 | Official sources mapped to a test protocol; no hands-on workflow result. |
| Access blocked | 2 | The access boundary was observed; authenticated workflow evidence is absent. |
Comparable subset
The 144-control denominator is fixed and inspectable.
Only records that completed the same 48-control protocol enter this distribution. Zoho FSM and Jotform have bounded direct observations, but their scopes differ and are excluded from the comparable-control total.
recorded controls out of 144 across the three comparable product records.
recorded controls out of 144 across the three comparable product records.
recorded controls out of 144 across the three comparable product records.
recorded controls out of 144 across the three comparable product records.
Reproduction protocol
The frozen file and counting rules preserve the claim boundary.
Public evidence records frozen on 2 September.
- Nine products listed in the public evidence ledger
- Evidence state, scope, result, market, date, next gate, and record path
- Three complete 48-control records in the comparable subset
- Counts summed directly from each record’s published result
No industry performance claim.
- No customer production data or vendor telemetry
- No survey population or representative market sample
- No cross-product score, winner, savings, or outcome inference
- No partial protocol treated as comparable coverage
To reproduce the snapshot, download the CSV, filter included_in_48_control_subset to yes, and sum the five control-count columns. The result must equal 144. The product-state count uses all nine rows. The file’s SHA-256 is d5a981ab5f73dc2b06bee14294f33c34b10cfc527aca8ea24180a00d0ea497ec. The public evidence ledger links each row to its supporting record.
Legacy search snapshot / 13-22 Aug
The original growth signal remains useful, but it is not reproducible enough to become a benchmark.
The earlier Search Console note recorded 2,635 impressions, zero clicks, average position 79.8, and 31 visible page rows. Only the five highest-priority page totals were preserved below. Because the unmodified export and exact filter record were not retained, this historical snapshot is a prioritization record, not an industry benchmark, a traffic forecast, or a dataset for longitudinal claims.
| Page | Impressions | First intervention |
|---|---|---|
| Best field service management software | 1,866 | Rewrite the search promise, add a decision route, and connect the pilot scorecard. |
| Best cleaning business software | 269 | Make recurring crews and proof of service visible above the shortlist. |
| Best HVAC software | 109 | Clarify agreement, equipment, and peak-season decision paths. |
| Field service software pricing | 85 | Move the weighted cost control and scorecard into the next-step path. |
| Best landscaping software | 68 | Separate route maintenance from project production earlier. |
Future benchmark gate
A market benchmark will publish only after the evidence package is complete.
The minimum release package is an unmodified source export, a data dictionary, explicit dates and markets, a versioned protocol, a reproducible calculation, sample exclusions, and a review of denominator and missing-data treatment.
Retained unchanged with collection time, owner, and source-system settings.
Products, accounts, markets, dates, roles, devices, and exclusions are explicit.
Definitions and calculation steps reproduce every published number.
Coverage data cannot become a quality, outcome, or market-performance conclusion.
Use the evidence
Start with the record, then run the same pilot.
The baseline exposes where direct evidence is deep, partial, documented, or blocked. It does not replace a buyer’s own workflow, cost, role, device, finance, recovery, and exit checks.