SENTINELLAT

Validation Report

One synthetic scenario. Baseline vs system. — illustrative sample

VENDOR (fictional, for illustration)

Acme Threat Response

SCENARIO

Time-to-contain a simulated lateral-movement event

BASELINE

Vendor's current manual/legacy containment process

PRODUCTION DATA USED

None — synthetic telemetry only

Methodology

One agreed synthetic scenario, run against a fixed baseline, under the same conditions used for any vendor. No production or customer data was used. The scenario was run three times to check the result held up on repeat, not just on a single favourable run.

Results

Latency — mean time to contain

Run	Baseline	System	Reduction
Run 1 (20 trials)	4.80s	1.35s	71.9%
Run 2 (20 trials)	4.85s	1.42s	70.7%
Run 3 (20 trials)	4.78s	1.29s	73.0%

Reliability — successful automated containment

Run	Successful	Rate
Run 1 (20 trials)	20 / 20	100%
Run 2 (20 trials)	19 / 20	95% — see note below
Run 3 (20 trials)	20 / 20	100%

ANOMALY NOTED — disclosed, not hidden

In Run 2, 1 of 20 trials failed to complete automated containment inside the test window and fell back to manual handling. Reported as found; not smoothed over or excluded.

Repeatability — spread across repeat trials

Run	Std. deviation	Read as
Run 1	± 0.08s	Consistent
Run 2	± 0.11s	Consistent, one outlier above
Run 3	± 0.07s	Consistent

What this means

Across three repeat runs, the system held a latency advantage over the baseline of roughly 71–73%, with one disclosed reliability anomaly in Run 2. The result is consistent, not a single best-case run — and the anomaly is reported rather than excluded.

Why trust the number

Every benchmark runs against a fixed, agreed baseline, under the same conditions we'd use for any vendor.

The vendor decides whether to publish a result. SentinelLat does not. This sample is published only to show report structure, not as a real engagement outcome.