fixaiprompt
All templates
⚡ Advanced Techniques·Works on: claude, chatgpt, grok

Algorithm Whisperer

Technique: Adversarial Reverse-Engineering

Will this post go viral? Decode why the algorithm shows or hides it. (TikTok, Twitter, LinkedIn, Reels.)

Advanced Viral#algorithm#viral#social#growth
0/5
Fill the template
5 placeholders left.
Live preview
Act as a senior algorithmic-recommender engineer who has shipped ranking systems. Reverse-engineer how the algorithm will treat my post.

Platform: 
My account stats (approx): 
My niche / what the algorithm has historically classified me as: 
Post / draft:


Analyse, in this order:

1. **Signals the algorithm will read in the first 3 seconds / first 50 chars** — be specific about which features.
2. **The implicit topic cluster** the algorithm will assign this to — and whether it matches my historical cluster (cluster drift = ranking penalty on most platforms).
3. **Engagement-rate prediction**: above / at / below my baseline. Reason in one sentence.
4. **The single biggest "hold-time" risk** — the moment a viewer is most likely to bounce.
5. **The viral-trigger probability** — pick one: (a) high, has a sharable mechanic; (b) medium, niche resonance; (c) low, professionally fine but algorithmically average. Defend the rating.
6. **The 3 specific edits** I should make to maximise reach without changing the message:
   - Edit 1: 
   - Edit 2: 
   - Edit 3: 
7. **The thing I should NOT change**, even though it looks risky — and why removing it would hurt more than help.

Do not flatter the post. If it's going to underperform, say so plainly and tell me whether to ship it anyway.
Inputs0 of 5
›See the lazy version this template replaces
Before — the lazy prompt
Will this post get reach?

Why it works

  • Algorithmic ranking is a real engineering domain — naming the engineer's role triggers more accurate analysis.
  • Per-platform specificity is critical: TikTok / Twitter / LinkedIn algorithms have different feature weights, and conflating them produces useless advice.
  • The 'thing I should NOT change' question prevents over-editing posts into algorithmic mush.
  • Asking for hold-time risk catches the 'great post, bad opening' pattern.
  • Above/at/below baseline framing produces a calibrated prediction instead of generic praise.

Make this one yours

Replace the bracketed placeholders, then paste into the Prompt Fixer to lint your customisation before hitting send.

Open Prompt FixerBrowse more templates