Comparison: two ways to do one thing, side by side, same task, same clock
The split-screen comparison that is fair enough to be believed: identical task, identical start, a visible timer, a verdict earned on screen. Prompt returns the script and the fairness checks.
Write a side-by-side comparison script: {option A} vs {option B} for {task}. The format works when the viewer believes it was fair, so the fairness is part of the script. Use this format when one option replaces something slower or more expensive; use a generic name for the other option ('the usual way', 'a desktop editor'), never a competitor's brand. Structure (40 s, 9:16 split top/bottom, or 16:9 split left/right): - 0-3 s: the question, spoken and on screen: '{A} or {B} for {task}?' Both halves show the starting state, identical. - 3-6 s: the rules, on screen as a 3-line list: same file / same start / timer on both. Spoken: 'Same photo, same start, clock running.' - 6-30 s: the task, both halves running at once. A timer in each half. Spoken lines only when one side reaches a step: 'A is done cutting out. B is still selecting.' Max 6 lines, max 12 words each. - 30-36 s: results side by side, timers frozen. Spoken: the two times and one difference you can see. - 36-40 s: verdict in one sentence, with the case where the loser is still the right choice (one clause). No download ask; the result is the argument. Comment ask: 'what would you compare next?' Word budget: 80-90 words at 165 wpm. On-screen: the question, the rules list, two timers, the two final times, the verdict in 5 words. Return: the fairness checklist (what must be identical), then a table (time, left/top half, right/bottom half, spoken line, on-screen text). Option A: {option A} Option B: {option B} Task: {task} What I expect to find: {honest guess, including where B wins}
In Claude Code, paste the prompt as a message.
Example
A: a background-removal photo app on a phone. B: the usual way, a desktop editor with a selection tool. Task: three product photos on white.
Fairness checklist: same three photos sent to both devices, both start from the file open, timers start on the first tap or click, no pre-made selections, both exported at the same size.
| Time | Top (phone app) | Bottom (desktop) | Spoken | On-screen |
|---|---|---|---|---|
| 0-3 s | Photo 1 open | Photo 1 open | Phone app or desktop editor for three listing photos? | PHONE vs DESKTOP |
| 3-6 s | Finger hovering | Cursor hovering | Same photos, same start, clocks running. | SAME FILES · SAME START · TIMER |
| 6-12 s | Tap, background drops | Selection tool tracing the handle | Phone’s done with one. Desktop is still on the handle. | 0:06 / 0:06 |
| 12-20 s | Photo 2, 3 done | Still refining the edge of photo 1 | Two and three. Desktop, edge of number one. | 0:20 / 0:20 |
| 20-30 s | Export, done, timer stops | Photo 2 selection | Phone stops at thirty-one seconds. | 0:31 / 0:30 |
| 30-36 s | Three results | Three results, timer stopped later | Thirty-one seconds against four minutes ten. Look at the handle edge: the desktop one is cleaner. | 0:31 vs 4:10 |
| 36-40 s | Phone for a listing. Desktop when the edge is the product. What should I compare next? | PHONE FOR LISTINGS |
Tips
- Name the case where the loser wins. A comparison with no concession reads as an ad and gets argued with instead of shared.
- Run the slower side for real and speed up the footage with a visible label. Cutting it short makes the result a claim instead of a measurement.
| works with | Claude Code: Runs as a skill or a prompt in the terminal; can read your files and write the result. · Claude: Paste into claude.ai or the Claude app. · ChatGPT: Paste into ChatGPT; works with any GPT model. · Any LLM: Model-agnostic: Gemini, Mistral, a local model. |
| cost | Free. No API, no key. |
| status | Unverified: written from the technique, not yet run end to end by us. |
| license | CC BY 4.0 for the text here; skills are MIT. |
| updated | 2026-11-01 |
| for agents | side-by-side-comparison-format.md |
Related
Before/after: the 20-second reveal with the 'after' shown first
The before/after format with the twist that keeps it from being skipped: open on the finished result, then show how bad it was, then the change. The prompt returns a timed script and text plan.
'I tried it for 7 days': a diary cut with one metric logged every day
The experiment-diary format that stays honest because one number is measured daily and shown on screen; the prompt plans the daily clip, the tally and a verdict that includes what did not work.
Storytime with a twist: a 60-second told story with the turn at 40 seconds
The talking-to-camera story that holds because the opener promises a turn and the turn arrives late. The prompt plans the beats, the b-roll inserts and the one line of text per beat.
Build in public: a 45-second behind-the-scenes cut around one decision
The founder diary format that earns trust because it shows the decision, not the launch: the prompt structures a week of real screen and desk footage around one choice and one measured result.
Day in the life: a 45-second cut from real footage with one through-line
A recipe for the vlog-style day that keeps working because the viewer waits for one promised payoff; the prompt turns your clip list into a timed script with text overlays.
Founder explainer: 'I built this because' in 40 seconds, to camera, one problem
The founder-to-camera format that converts because it names the specific moment the problem became unbearable, then shows the product once. The prompt bans origin-story clichés and budgets the words.