TL;DR

• Run this checklist every 90 days; align it with the GEO Authority Signal Engineering Framework cadence.
• 40 items across six tracks: Technical, Content Structure, Entity & Authority, Citation Tracking, Competitive Drift, Remediation.
• Each item has a single owner, observable evidence, and a pass/fail criterion.
• Output: a one-page scorecard and a remediation backlog scoped for the next quarter.

Why quarterly

Daily and weekly cadences catch tactical issues (bad deploys, broken schema). Quarterly is the right cadence to detect:

• Entity drift (your canonical claim getting overwritten by competitors).
• Citation share trend changes across engines.
• Index coverage erosion in Bing or Google.
• Authority signal decay (broken sameAs links, stale Wikidata).
• Engine-side algorithm shifts that change which content shapes get cited.

If you only audit once a year, by the time you find a regression you have lost two quarters of citation share. If you audit weekly, you drown in noise. Quarterly hits the sweet spot.

Roles

• Audit owner. Content ops lead. Drives the schedule, owns the scorecard.
• Technical contributor. SEO/web engineer. Runs technical track items.
• Editorial contributor. Managing editor. Runs structure and entity tracks.
• Analytics contributor. Data analyst or marketing ops. Runs tracking and drift tracks.
• Reviewer. GEO/content director. Approves remediation backlog.

The 40-point checklist

Track 1 — Technical AI readiness (8 items)

• [ ] 1.1 llms.txt exists, returns 200, and lists priority sections. Evidence: curl -I /llms.txt + manual diff vs. last quarter. Owner: Technical.
• [ ] 1.2 llms-full.txt (or per-page .md URLs) exists and is current. Evidence: spot check 5 pages. Owner: Technical.
• [ ] 1.3 robots.txt does not block major AI crawlers (GPTBot, PerplexityBot, ClaudeBot, Google-Extended) unless intentionally. Evidence: robots.txt review. Owner: Technical.
• [ ] 1.4 Sitemap is fresh and submitted to Google Search Console + Bing Webmaster Tools. Evidence: GSC/BWT screenshots. Owner: Technical.
• [ ] 1.5 Bing index coverage ≥ 90% of priority URLs. Evidence: BWT coverage report. Owner: Technical.
• [ ] 1.6 Google index coverage ≥ 95% of priority URLs. Evidence: GSC coverage. Owner: Technical.
• [ ] 1.7 Core Web Vitals pass on priority URLs. Evidence: PageSpeed Insights. Owner: Technical.
• [ ] 1.8 No JS-rendered-only critical content (HTML-first text for priority pages). Evidence: curl raw HTML check. Owner: Technical.

Track 2 — Content structure & groundability (8 items)

• [ ] 2.1 Top 25 priority pages have an answer-first paragraph in the first 150-200 words. Evidence: spot review. Owner: Editorial.
• [ ] 2.2 Top 25 pages have a TL;DR or summary block. Evidence: spot review. Owner: Editorial.
• [ ] 2.3 Top 25 pages include a FAQ block with 3-5 answer-first Q&A. Evidence: spot review. Owner: Editorial.
• [ ] 2.4 No buried answers ("the answer is below the marketing copy"). Evidence: LLM grader prompt asking the canonical question only from the page. Owner: Editorial.
• [ ] 2.5 Heading hierarchy clean (single H1, sequential H2/H3). Evidence: HTML lint. Owner: Editorial.
• [ ] 2.6 Internal linking graph healthy: every priority page receives ≥3 internal links. Evidence: link graph export. Owner: Editorial.
• [ ] 2.7 Comparison and how-to pages include tables / steps that retrieval can extract. Evidence: spot review. Owner: Editorial.
• [ ] 2.8 Misconceptions section present where the page corrects common errors. Evidence: spot review. Owner: Editorial.

Track 3 — Entity & authority (7 items)

Reuse outputs from the GEO Authority Signal Engineering Framework.

• [ ] 3.1 Wikidata entity exists for brand and primary product, with sameAs to your canonical URL. Evidence: Wikidata diff. Owner: Editorial.
• [ ] 3.2 Schema.org markup validates on priority pages (Article, Product, FAQPage, HowTo as appropriate). Evidence: Schema validator. Owner: Technical.
• [ ] 3.3 sameAs cluster across LinkedIn, GitHub, Crunchbase, etc., is consistent. Evidence: manual diff. Owner: Editorial.
• [ ] 3.4 ≥1 third-party citation per priority claim in the past 90 days. Evidence: mention monitor export. Owner: Analytics.
• [ ] 3.5 No conflicting entity statements between docs, marketing, and product surfaces. Evidence: terminology audit. Owner: Editorial.
• [ ] 3.6 Author bylines + reviewed_by are populated on priority pages. Evidence: spot check. Owner: Editorial.
• [ ] 3.7 Retraction trail / changelog is current; corrections from last quarter are documented. Evidence: changelog page. Owner: Editorial.

Track 4 — Citation tracking (6 items)

• [ ] 4.1 Tracking set of 25-50 priority queries is current and stratified across topics. Evidence: tracking sheet. Owner: Analytics.
• [ ] 4.2 Citation share measured per engine (ChatGPT, Perplexity, AI Overviews, Gemini, Copilot, Claude). Evidence: tracking export. Owner: Analytics.
• [ ] 4.3 Quarter-over-quarter citation share trend documented. Evidence: trend chart. Owner: Analytics.
• [ ] 4.4 Citation surfaces mapped to the SERP feature citation map. Evidence: surface coverage doc. Owner: Analytics.
• [ ] 4.5 AI traffic referrals reconciled in GA4 / analytics platform. Evidence: referral report. Owner: Analytics.
• [ ] 4.6 Brand-prompt sentiment + accuracy spot-checked across engines. Evidence: brand prompt log. Owner: Analytics.

Track 5 — Competitive drift (5 items)

• [ ] 5.1 Top 5 competitors' citation share per priority query measured. Evidence: tracking export. Owner: Analytics.
• [ ] 5.2 Competitor entity claims monitored for overwrites ("who is the leading X" prompts). Evidence: prompt log. Owner: Analytics.
• [ ] 5.3 New competitor llms.txt / MCP coverage detected. Evidence: manual scan. Owner: Technical.
• [ ] 5.4 New SERP features in priority queries logged. Evidence: SERP map diff. Owner: Analytics.
• [ ] 5.5 Competitor freshness cadence benchmarked. Evidence: sample of competitor change-logs. Owner: Analytics.

Track 6 — Remediation & planning (6 items)

• [ ] 6.1 Failed items grouped into themes (technical, structural, entity, tracking). Evidence: scorecard. Owner: Audit owner.
• [ ] 6.2 Each theme assigned an owner and target date. Evidence: backlog sheet. Owner: Audit owner.
• [ ] 6.3 Citation Confidence Score recomputed for top 25 priority pages. Evidence: scoring sheet. Owner: Analytics.
• [ ] 6.4 Pages with CCS < 0.6 enter the next-quarter rewrite queue. Evidence: rewrite list. Owner: Editorial.
• [ ] 6.5 Tracking set adjusted for next quarter (add new priority queries, retire dead ones). Evidence: tracking sheet diff. Owner: Analytics.
• [ ] 6.6 Executive one-pager produced (score delta, citation trend, top 3 risks, remediation plan). Evidence: one-pager. Owner: Audit owner.

Pass thresholds

A quarter passes overall if every track clears its threshold. Anything below = automatic backlog item for the next quarter.

Scorecard template

GEO Quarterly Audit — Q? 20??

Overall: PASS / FAIL

Tracks (P/F):

Technical: ?/8

Structure: ?/8

Entity: ?/7

Tracking: ?/6

Drift: ?/5

Remediation: ?/6

Citation share trend: +x.y% QoQ

Top 3 risks:

1. …
2. …
3. …

Next-quarter focus: …

Keep the scorecard short. Detail belongs in the backlog and tracking sheets, not the one-pager.

Common pitfalls

• Auditing before the tracking set is stable. A churning query set produces noisy citation trends. Lock the set at the start of each quarter and resist mid-quarter edits.
• Treating the audit as one-and-done. Each quarter's failures must enter the next quarter's plan, otherwise the same items recur indefinitely.
• Skipping competitor drift. Citation share is zero-sum on most queries. Tracking only your own pages gives you a flattering but false picture.
• Confusing tools with evidence. A dashboard screenshot is not the same as a documented pass criterion. Always record the value and the threshold.

How to apply

1. Schedule. Block 1-2 days per quarter on the audit owner's calendar.
2. Pre-work. One week ahead, ask each contributor to refresh their inputs (sitemaps, tracking exports, schema reports).
3. Run. Walk the checklist top-to-bottom in a shared doc; mark P/F + evidence link inline.
4. Synthesize. Same day, draft scorecard and remediation backlog.
5. Review. Within 5 business days, review with the GEO/content director and lock the next-quarter focus.

FAQ

Q: Can I run this checklist monthly instead?

You can, but most items (entity drift, competitive citation share) move on a quarterly time scale. Monthly cadence yields diminishing returns and burns analyst time. Keep the deep audit quarterly and instrument lighter weekly checks for breakage (broken llms.txt, schema validation failures).

Q: We are a small team — can we skip tracks?

Skip Competitive Drift first if forced to cut. Never skip Tracking or Remediation; without them you are auditing without learning.

Q: How does this differ from a one-off GEO audit?

One-off audits diagnose. The quarterly checklist governs. The output is not a report but a backlog the team commits to executing in the next 90 days.

Q: How does this checklist tie into the Citation Confidence Score?

Track 6 ends with a CCS recomputation. Pages with low CCS feed directly into the rewrite queue, closing the loop between audit and content production.

Q: What if our citation share looks flat?

Flat citation share masks two scenarios: (a) you are stable in a stable market, (b) you are stable while competitors gain. Always compare against competitor share trend before declaring "flat = healthy."