Side-by-side contrast reveals why viral views don't equal followers

Opens with a shared metric (1M views each) that immediately creates tension , same result, wildly different outcomes. Middle runs a parallel comparison structure: each variable (content type, length, CTA presence) is isolated and contrasted one-by-one between the two characters. Close resolves the tension by naming the single highest-leverage fix (explicit CTA) and immediately demonstrates it in the same breath.

States the follow CTA explicitly in-video ('if you want them to follow you, you have to have a follow CTA at the end') then executes one live , modeling the exact behavior being taught.

- Open with two identical vanity metrics side-by-side in the first 5 words to force a 'wait, how?' reaction before any explanation is given. - Use a two-character dialogue format to externalize the contrast , one character holds the wrong assumption, the other corrects it; this removes the need for a direct lecture tone. - Isolate each variable (content type, length, CTA) as its own single-line exchange rather than bundling them , one variable per beat keeps the comparison scannable and skimmable. - Position entertainment content vs. educational/storytelling content as a conversion lever, not a quality lever , frame it as a strategic choice with measurable follower consequences. - Name the viewer's exact objection ('I didn't want to sound desperate') before correcting it , this neutralizes resistance before the fix lands. - End by demonstrating the advice in real-time inside the same video (live follow CTA after explaining why CTAs work) , proof-of-concept closes the loop. - Keep the corrective character's tone dismissive-but-factual ('Don't appreciate the dig, but...') to add personality friction without losing credibility.