Promptivo Writing Guide

Most people who struggle with AI outputs assume the model is to blame. In reality, the problem is almost always the prompt. A language model is a precise instrument — give it a vague instruction and it will produce a vague result, reliably and without complaint. The good news is that prompt quality is not a matter of talent or intuition. It’s a set of measurable, learnable skills.

Promptivo evaluates prompts across 7 quality dimensions, each addressing a different failure mode. These are not a checklist to run through manually before every prompt — they’re a set of intuitions that, once internalized, change how you write. You don’t have to score 5 on all dimensions to get the output you need.

Clarity

Say what you mean

The most fundamental failure in prompt writing is ambiguity. Not the dramatic kind, where you contradict yourself, but the quiet kind — a pronoun without a referent, a task that could be read two different ways, an implied subject that the model has to guess. The test for clarity is simple: could a stranger read your prompt and know immediately what output you expect? If there's any room for interpretation, there's room for a bad answer.

Scoring scale

1No clear goal; multiple valid interpretations
2Intent guessable but requires inference
3Mostly clear with one ambiguous element
4Clear goal, minor wording could be sharper
5Crystal clear — zero ambiguity

Examples

Weak

“Explain it to me.”

⚠ "it" is undefined — the model has no idea what to explain.

↓ Improved
Improved

“Explain how HTTPS works to a junior developer. Cover the TLS handshake, certificates, and why it matters for security. Use simple analogies.”

✓ Topic, audience, sub-topics, and style all specified.

Score1/55/5
Weak

“Can you help me with the report and also suggest some improvements and maybe a new title?”

⚠ Three separate asks bundled together — each competes for focus.

↓ Improved
Improved

“Review the attached report for clarity and conciseness. List up to 5 specific improvements as bullet points. Then suggest 3 alternative titles.”

✓ Tasks separated, counts specified, output structure defined.

Score2/54/5
Weak

“Write something about climate change.”

⚠ "Something" gives the model no direction — length, angle, audience, and format are all undefined.

↓ Improved
Improved

“Write a 300-word explainer on the causes of climate change for a high-school audience. Use plain language and avoid jargon.”

✓ Length, topic focus, audience, and tone all locked in.

Score1/55/5

Quick checklist

  • Does every pronoun ("it", "they", "this") have a clear referent?
  • Is there exactly one main task? Split multi-part asks into numbered steps.
  • Could a stranger read this and know immediately what output you expect?
  • Is the subject of the prompt named explicitly (not implied)?
  • Would a 10-year-old understand what you are asking for?