SEOGPT-5.6 readyClaude Fable 5.1 ready

Internal Linking Strategy Prompt

A detailed prompt template with named context, evidence rules, a clear output contract, and a quality check — built from this original task:

Create an internal linking strategy for our website content, identifying opportunities to link between relevant pages and articles to improve the user experience and boost SEO performance. Our website is: .

Prompt variables

Add your context

0/6

Fill in what you know. Empty fields stay as editable placeholders in the prompt.

<url_or_page_type>

<target_audience>

<primary_query_and_related_intents>

<conversion_or_information_goal>

<search_console_analytics_serp_or_research_data>

<technical_editorial_or_compliance_limits>

Live prompt

Your entries update the template instantly.

You are a technical SEO and content strategist who validates current search guidance and user intent.

Complete the task below using the supplied context and only the sources you are authorized to use.

<task>
Create an internal linking strategy for our website content, identifying opportunities to link between relevant pages and articles to improve the user experience and boost SEO performance. Our website is: .
</task>

<context>
Site or page: <url_or_page_type>
Audience: <target_audience>
Query or topic: <primary_query_and_related_intents>
Business goal: <conversion_or_information_goal>
Evidence: <search_console_analytics_serp_or_research_data>
Constraints: <technical_editorial_or_compliance_limits>
</context>

Working rules
1. Identify the likely search intent and the job the page must complete for the visitor.
2. Use current first-party search documentation for claims about crawling, indexing, structured data, or ranking systems.
3. Separate observations from hypotheses and proposed experiments. Do not promise rankings or traffic.
4. Prioritize fixes by user impact, technical dependency, confidence, and implementation effort.

Output contract
1. A concise diagnosis of intent, page purpose, and the largest opportunity or risk.
2. A prioritized recommendation table with evidence, impact, confidence, effort, and owner.
3. The requested SEO deliverable in copy-ready or implementation-ready form.
4. A measurement plan with baseline, leading indicators, outcome metrics, and review date.

Quality check
Before finalizing, verify the result against each criterion below. Fix any gap you can resolve from the supplied context; list anything that still needs human input.
1. Recommendations improve the visitor experience, not only keyword placement.
2. Technical claims are current and source-backed.
3. The plan avoids doorway pages, deceptive tactics, thin content, and unsupported guarantees.
4. Success can be measured against a defined baseline and time window.

Return the finished deliverable first. Keep private reasoning internal. Do not add a generic introduction.

Review the completed prompt before using it. After the first answer, make one targeted change at a time so you can see what improves the result.

Why this works

Enough structure, without scripting the reasoning

  • It combines the query with the visitor's actual job, reducing mechanical keyword output.
  • It asks for evidence and a baseline before proposing a large backlog.
  • It treats SEO recommendations as measurable hypotheses instead of guarantees.

Model notes

GPT-5.6

Use medium reasoning as a practical starting point. The prompt defines the outcome, evidence boundary, constraints, and final answer while leaving the model room to choose an efficient approach.

Claude Fable 5.1

Start with high effort, then test lower levels against your own quality bar. The XML sections, explicit source rules, and output contract match current Claude prompting guidance without asking for visible chain-of-thought.

Guidance checked September 1, 2026.

After the first draft

Continue with one focused follow-up

Keep the original brief in the same conversation, then use the next instruction that matches what you actually need.

1

Critique the internal linking strategy above against the quality check. List only material gaps, then provide a corrected version.

2

Create a second version for <alternative_audience_or_scenario>. Preserve verified facts and explain the three most important changes.

3

Turn the result into a 30-day action plan with owners, dependencies, success signals, and a weekly review cadence.

Ground the prompt in your own business

Train Cody on approved company knowledge, then run detailed prompts with answers grounded in the context your team trusts.

Get Started Free