Director Chair
All groups

Compare reach and engagement with the account's own medians, never with another account.

Test and publish

Read two numbers, then decide

You know whether to fix the opening, fix the body, repeat or drop.

Use when
A post that has settled, about 72 hours after it went out.

Prompt instructions

Give me a step-by-step plan I can follow this week, for a short video about [your product], for [your audience]. The technique is read two numbers, then decide: you know whether to fix the opening, fix the body, repeat or drop. How it works: compare reach and engagement with the account's own medians, never with another account. It is right for: a post that has settled, about 72 hours after it went out. Reach without engagement: a fake-news opener had 490,000 plays at 1.0 percent likes and 0.18 percent saves, and led nowhere. Engagement without reach: the three weakest posts reached 7,600 to 11,700 views yet were saved at 0.9 to 3.4 percent, as much as the hits. Be concrete: numbers, seconds and settings, no theory. Ask me one question first if something you need is missing.

The same prompt works in Claude, ChatGPT and Gemini.

Watch out

Notes

The four outcomes

Engagement at or above the medianEngagement under half the median
Reach 3 times or moreA winner. Post the next one on the same promise within 48 hoursThe opening works, the body or payoff does not. Keep frame 0, fix what follows
Reach under halfThe body works, the opening did not stop people. New frame 0, re-cut, post againDrop the idea

Reach is views divided by the median of the account's ten posts before it. Engagement is likes, saves and shares per view against the account's own medians.

Related

Sources

Own words from Director Chair field notes on short-form video (one TikTok account, 233 posts and 27 measured clips, read 30 Oct 2026) Checked 2026-11-13.

For agents: this card as markdown, as JSON.