UC-11 Any time SEO leadExecAnalyst

Is this number trustworthy?

A number is about to go into a deck. Before it does, you want to know whether it would survive someone asking how you know.

What you'd type

Is this change statistically significant?
Did conversion rate really move, or is that noise?
Can I put this number in the board deck?

You never name a skill or a tool. The assistant routes on what you asked. Run it on your own data →

What comes back

Use case: Is this number trustworthy? Simulated · illustrative

Conversion rate is up 12%, can I put that in the board deck?

Analysis plan, every step, resolved for you

Routed to the governed workflowsignificance-referee
Dates bound explicitlyJul 2026
1 analysis queuedpop_significance
Honesty gates armedsignificance α=0.05 · scope echo · freshness

Not yet. The lift is there in the raw numbers, but the sample cannot separate it from normal month-to-month variance.

Conversion rate: 3.41% ▲ +12.2% goal: Trial started
Sessions: 48.2K the denominator, and the reason the test cannot resolve

Jul 2026 vs Jun 2026organic

⚠ Observationalas of Aug 1, 2026 (web analytics loads nightly)IllustrativeOpen in Quattr ↗

The interval spans zero: a 6% decline and a 31% gain are both consistent with this data.

Verdict: Noise, not signal
p-value: 0.19
95% CI: −5.8% to +31.4%

At this session volume, roughly nine more weeks would be needed to resolve a lift this size.

Jul 2026 vs Jun 2026organic

⚠ Observationalas of Aug 1, 2026 (web analytics loads nightly)IllustrativeOpen in Quattr ↗

Report it as flat, or report it with the interval attached. "Up 12%" on its own is a claim this data cannot support.

A simulated conversation on fictional data. Scope, freshness and the ⚠ Observational marker are attached by the same director that renders every demo on this site.

The answer is sometimes no

A significance test can return three verdicts, and "cannot tell yet" is one of them. When the sample is too small to distinguish a real change from variance, that is what the card says, it does not round the verdict up to make the slide easier to write.

What's behind it

Analyses it runs

significance_check
pop_significance
correlate_domains

Sources it reads

The skills that own this

The dashboard rituals it replaces

Going deeper

Next use case The exec readout →

See this use case run for real

All runs →