Appendix 1: The Impact GEO self-check list

Use this list to check your own practice on a schedule. Brands, GEO service providers, content creators and platform governors are all covered. Run it quarterly, about 30 minutes each time. Six sections map onto the core modules in the book. You do not need full marks. You do need to know where the gap is.

1. Content transparency

Goal: make sure everything you publish survives the test of traceable origin and identifiable interests.

1. Does every piece state the author's identity and professional background?
2. Is commercially sponsored content labeled as advertising or sponsorship?
3. Does every core factual claim carry a traceable primary source?
4. Are factual statements and opinions distinguished in the text itself?
5. Have unverifiable phrases such as "industry-leading" or "the best choice" been avoided?
6. Is the brand knowledge base stored separately by zone: fact, opinion and marketing?
7. Do factual claims in marketing copy match the fact-zone version?

Self-check conclusion: if item 3, 4 or 6 is a no, fix it first. Those three are the base of transparency.

2. Source credibility

Goal: make sure core information appears in channels AI systems trust, and says the same thing in each of them.

1. Are core factual claims anchored to Tier A or Tier B sources (government websites, mainstream media, academic papers, industry whitepapers)?
2. Has the brand's core factual information been published in at least one high-weight source channel?
3. Does the brand website show complete qualifications, contact details, ICP license information and an update record?
4. Is factual consistency across sources checked regularly, covering dates, figures and definitions?
5. Have bulk postings of duplicate content on low-quality channels been avoided?
6. Have you taken part in industry research or supplied data and cases third parties could cite?

Self-check conclusion: if item 1 or 2 is a no, the brand's initial citation weight is low. Establish at least one core factual source in a high-weight channel first.

3. The five scoring dimensions

Goal: rate a single piece of content against the five dimensions that set AI citation weight.

Media tier and endorsements (25%)

• Does the publishing channel have verifiable qualifications and an ICP license?
• Is there a record of editorial review or peer review?

Content expertise (22%)

• Is there a named author with a verifiable professional background?
• Does every thousand words carry at least 5 verifiable statistics and at least 2 authoritative citations?
• Is the core conclusion stated in the opening paragraph?

Site health (18%)

• Does the site have HTTPS, mobile responsiveness and structured data markup?
• Is page speed within an acceptable range?
• Is there a violation history or a mass of broken links?

Social and citation signals (20%)

• Has the content been cited by other high-weight sources?
• Is there expert endorsement or a recommendation from a professional community?
• Are user comments positive and specific?

Content freshness (15%)

• Has the core content been updated within two years?
• Do the figures state when they were published and how long they hold?
• Has the site responded promptly to events in the field?

Self-check conclusion: of the five, raise content expertise and social and citation signals first. The first is the most controllable; the second only comes with time.

4. Creator reputation

Goal: see how much long-term credibility has actually accumulated, and where reputation building is weakest.

1. Do you keep producing verifiable content in your core field?
2. When AI cites your content, does the citation express the original information correctly?
3. Have partners invited you back for repeat work?
4. Can your professional credentials be verified independently (education, title, career history, body of work)?
5. Do you keep your public biography and record of results updated?
6. Have you proactively corrected content AI cited wrongly?
7. Have you contributed to shared knowledge in at least one professional community?

Self-check conclusion: if item 1 or 2 is a no, accumulation has not started. From this week, sign every piece and label its sources.

5. Ecosystem impact

Goal: judge the net effect your work has on the AI information ecosystem, and spot negative externality risks.

1. In the last three months, did you publish anything unverifiable?
2. Have you taken part in generating similar content in bulk?
3. Have you embedded hidden instructions or prompts in content?
4. Have you proactively corrected content AI cited wrongly?
5. Have you supplied reusable public knowledge to the field (method documents, industry data, case studies)?
6. When you found a competitor poisoning information, did you report it to a platform or an industry body?
7. Does your content distinguish where the brand's interest ends and the public interest begins?

Self-check conclusion: if the answer to item 1, 2 or 3 is yes, stop and correct it immediately. A history of negative externality can depress a brand's long-term weight inside AI systems.

6. The 90-day roadmap

Week 1

• Add bylines and source labels to the last three pieces you published.
• Check the brand website's HTTPS, mobile responsiveness and structured data.

Month 1

• Inventory the brand's verifiable assets and draft the fact zone.
• Fix 3 to 5 core queries and test current citation results on several AI platforms.
• Publish at least one piece of core factual content in a high-weight source channel.

Month 2

• Build a source matrix: set the publishing strategy and priority for each tier of source.
• Run three-zone separation: store the brand knowledge base by fact, opinion and marketing.
• Start an AI citation monitoring log recording source, context and accuracy.

Month 3

• Optimize at least three core pieces against the five scoring dimensions.
• Take part in one discussion on an industry standard or self-regulation rule, and put front-line practice on the table.
• Complete the first ecosystem impact check, and correct what it turns up.

After 90 days

• Re-test the core queries and compare against the figures from three months ago.
• Iterate on assets, sources and expression from what the monitoring says.
• Write the next 90-day plan, and move up a service capability tier or deepen reputation accumulation.

7. The one-question check

If only one question stays with you, make it this one:

My content, once AI cites it: does this make AI's answers more trustworthy, or less?

If the answer is more trustworthy, keep going. If it is less, stop and start again.

How to use this checklist: this is not an exam, and full marks are not the point. Its value is showing you which plank is shortest. Pick the weakest section each quarter and work on it alone. Look back in a year and you will find you are already inside the virtuous cycle of ecosystem health.