AI Skills Marketplace / Email & Automation

Send-Time Model Explainer

The Send-Time Model Explainer is a free Claude Skill that builds a recency-weighted hour x weekday engagement histogram with a minimum-sample-size guard, and recommends a specific send window with a plain-English rationale.

↗ View Source ⬇ Download Full Repo (.zip)
MIT
License
168 bins
Decision branches
min-sample-size guard
Escalation rules
Yes
Multi-tool support
Aug 14, 2026
Last updated
12/12 passing
Tests
README.md View raw ↗

What this is (and isn’t)

A histogram and recommendation function over engagement events you supply, not a hosted service or an ESP integration.

Files

FilePurpose
explainer.jsexplainSendTime() — zero dependencies
test.js12-test suite covering the minimum-sample-size guard, recency weighting, and weight scaling
SKILL.mdFull skill definition, loadable in Claude Code/Desktop

Why the minimum-sample-size guard matters

A single lucky click at 3am shouldn’t recommend sending at 3am. This skill only recommends a bin once it has enough raw events to trust, falling back to the best trusted bin with confidence flagged as "low" rather than silently recommending a fluke.

FAQ

Is this skill free to use?

Yes. Free to download and run yourself. MV3 charges $175/hr only if you want it wired into your real ESP so recommendations update automatically.

What counts as "insufficient data"?

When no single hour-by-weekday bin reaches your configured minimum sample size, the skill returns null with an honest explanation rather than guessing.

Can opens and clicks be weighted differently?

Yes, every event accepts an optional weight, so a click can count for more than an open.

Does my engagement data get sent to MV3?

No. The histogram runs entirely in your own environment.

Which ESPs does this work with?

Any platform you can export open/click timestamps from, including Klaviyo, Mailchimp, and ActiveCampaign.

JR
Jordan Reeves
ABM & Outbound Pipeline, MV3 Marketing