Performance Budget Receipts
Pricing
$80.00 / 1,000 url/device bundle (3 lighthouse runs)s
Performance Budget Receipts
Pricing
$80.00 / 1,000 url/device bundle (3 lighthouse runs)s
Rating
0.0
(0)
Developer
muazah
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
21 hours ago
Last modified
Categories
Share
A release gate, not a speed score: three sequential Lighthouse lab runs per URL, median and min/max evidence, and assertions against the budget you declare.
What it does
- Runs Lighthouse 3 times sequentially per URL (never concurrent audits on one machine), same cold-cache condition and fixed mobile or desktop emulation every run.
- Reports every raw run plus median and min/max range for LCP, CLS, TBT, FCP, Speed Index, total bytes, request count, script bytes, and the performance category score.
- Budget assertions in metric units (lcpMs, tbtMs, cls, scriptBytes, totalBytes, requestCount...): median vs your limit decides PASS or BUDGET_EXCEEDED. If the run range crosses your limit, you get a RANGE_CROSSES_LIMIT warning, because one threshold is not statistical proof of regression.
- Baselines with an environment fingerprint (tool version, Chrome version, device): comparison is refused as BASELINE_INCOMPATIBLE unless the fingerprint matches exactly. Deltas are reported in metric units, not score points. A failing run is never auto-promoted to baseline - you save baselines explicitly.
- Raw Lighthouse JSON for every run and an HTML report per URL in your run's key-value store, plus an assertions CSV.
- Missing metrics come back null with an UNVERIFIABLE warning, never zero.
Honest limits
- Lab metrics under simulated throttling are not field Core Web Vitals. This does not determine whether Google considers your page passing, and TBT is not INP.
- Lighthouse itself is variable: shared CPU and network move numbers. That is why you get the range, not just one run.
- Public HTTPS pages only, up to 5 URLs per run. You must be authorized to test the pages you point it at.
Input
Output
url_receipt rows (status SNAPSHOT / PASS / BUDGET_EXCEEDED / INCOMPLETE / BASELINE_INCOMPATIBLE), assertion rows, raw Lighthouse JSON/HTML artifacts, assertions.csv and a summary in the run's key-value store.
Billing
Pay per completed URL/device bundle of three runs. A bundle that could not finish all three runs is INCOMPLETE, uncharged, and its partial raw evidence is kept and marked.
