SEO Prompt Engineering: Reusable Briefs, Guardrails, and QA

A good SEO prompt is not a magic sentence. It is a compact work order: a defined task, known inputs, source boundaries, an output schema, and a test for acceptance. This approach makes model output easier to review and less likely to hide unsupported assumptions behind polished prose.

Use a seven-part prompt contract

1. Role: define the practical perspective needed, without inventing credentials or experience.
2. Goal: state the reader or analyst task and the decision the output should support.
3. Audience: name prior knowledge, context, locale, and constraints.
4. Inputs: provide pages, data, source notes, examples, and the date window.
5. Rules: define allowed sources, forbidden claims, tone, and how uncertainty must be shown.
6. Output: specify fields, order, length range, and labels.
7. Verification: require claim-source mapping, missing-data flags, and checks a person can repeat.

A reusable SEO analysis prompt

TASK
Create a content brief for [reader task].

AUDIENCE
[Who they are, what they already know, and what they need to decide.]

INPUTS
Use only the supplied Search Console export, page inventory, and source URLs.

RULES
Separate observations from recommendations. Do not invent search volume, rankings, quotes, tests, or product behavior. Mark missing evidence as UNKNOWN.

OUTPUT
1. Intent statement
2. Evidence table: observation | source | date | confidence
3. Proposed outline with purpose for each section
4. Internal-link candidates with destination and reason
5. Risks and verification checklist

ACCEPTANCE TEST
Every factual claim maps to an input. Every recommendation includes a mechanism and a post-publication metric.

Prompt for evidence extraction before drafting

Separate source work from prose generation. First ask for a claim ledger containing claim, source URL, date, exact supporting location, scope, and uncertainty. A human should open each source and approve the ledger. Only then ask the model to draft from approved claims. This two-pass design makes unsupported leaps easier to catch.

Prompt for metadata candidates

Generate five title and meta-description pairs for the page below.

For each pair include:
- title
- description
- primary reader promise
- evidence in the page that supports that promise
- risk of mismatch or overstatement

Constraints:
- preserve the page's actual scope
- no guarantee language, fabricated numbers, or keyword lists
- descriptions must be page-specific
- do not assume Google will use the supplied title or description

Prompt for internal-link suggestions

Supply a list of source passages and candidate destination pages. Require output fields for source URL, exact passage, destination URL, proposed anchor, reader benefit, relevance score, and rejection reason. The rejection field matters: a system that must explain why a suggestion might be inappropriate is easier to audit than one optimized only for volume.

Make uncertainty visible in the schema

Common prompt failures and fixes

Score the output before using it

• Does every material claim have a source or a visible uncertainty label?
• Does the output solve the stated audience task rather than merely include keywords?
• Are observations, inferences, examples, and recommendations clearly separated?
• Can another reviewer repeat the analysis from the provided inputs?
• Would the page still be useful if search engines did not exist?
• Has a human checked the final rendered page and accepted accountability?

Primary sources

• OpenAI: Prompt engineering guide
• Google: Guidance on generative AI content
• Google: Creating helpful, reliable, people-first content
• Google: Spam policies for web search