Most freelance SEO audits are written for the consultant, not the client. They run 40 to 80 pages, quote every GSC chart available, restate industry benchmarks, and end with a 200-item "recommendations" spreadsheet that nobody — neither the client nor the consultant — will ever work through in order.

This checklist is the opposite. It produces one page. Three lists. Zero industry commentary. It is the audit we run inside SEO Triage on every free audit request, and it is designed to complete in under 2 hours per client site while surfacing work that pays for itself inside a single retainer cycle.

Why the 2026 calibration matters

SEO audits written before 2024 assumed every ranking click converted to a visit. That assumption broke twice — first when AI Overviews started pulling informational click-through below historic baselines, and again when Google's own click data (post the 2024 leaks) confirmed that positions 4 to 10 had shifted expected CTR lower than what the public Advanced Web Ranking benchmarks had reported for a decade.

The methodology page publishes the updated expected-CTR lookup table calibrated in March 2026. The short version: expected CTR at position 12 is 1.5% (not 2.5% as older benchmarks still show), and the composite score applies a 15% haircut to any keyword whose SERP shows an AI Overview block. Audits that skip those two updates will over-report winnable keywords by 20 to 30%.

If you are still running pre-2024 expected-CTR tables, every Quick Win your audit produces is probably a false positive. The fix is mechanical: update the lookup table, re-score, re-tier.

Step-by-step — the 10 steps, expanded

Step 1 — GSC access first, always

Tool-based audits that do not include real GSC data are guessing. DataForSEO can tell you the SERP composition, average volume, and ranking position. Only GSC can tell you the actual CTR, the actual impressions, and whether the ranking is stable or drifting. No substitute. If the client has not granted access yet, stop the audit and get access before scheduling any optimization work — a $200 audit ticket runs cheaper than a $2,000 client engagement predicated on guessed CTR.

Step 2 — Composite score tiering in one pass

Export the GSC Performance report as CSV. Join against a SERP composition source (DataForSEO, your preferred SERP API, or SEO Triage's audit endpoint) to fetch the top-10 per keyword. Apply the composite formula deterministically:

Quick Win: score ≥ 0.50
Opportunity: score ≥ 0.30
Monitor: score ≥ 0.15
Ignore: score < 0.15

The glossary defines each tier precisely. The thresholds are fixed in product logic — we do not tune per-client because tuning is what makes audit outputs non-comparable across engagements.

Step 3 — At Risk detection before opportunity hunting

Every audit should surface the At Risk list before the Quick Win list. The rationale is defensive: a keyword that used to rank position 4 and now ranks position 11 is losing clicks continuously. Recovering the lost position often takes a 20-minute edit and produces 5 to 10x the recovered-click impact of a net-new Quick Win at the same effort.

Criteria: position dropped ≥ 3 slots in the last 30 days, impressions held within ±15% (to rule out seasonal drops in demand), and the page itself has not been deleted or redirected. These three conditions isolate actual ranking regressions from noise and from legitimate consolidation.

Step 4 — Title tag and meta description audit

This is the fastest leverage in any audit. For every page ranking top-20 on a commercial query:

Does the title contain the exact-match keyword? If not, that is a 5-minute fix with ~10% CTR uplift expected
Is the title under 60 characters (truncation threshold)? If not, rewrite
Does the meta description include a specific verb-led outcome promise? "Learn more about X" does not count; "Compare prices on X in under 60 seconds" does
Is the meta description under 160 characters?

A 100-page site audited against these four questions usually surfaces 20 to 40 fixable title tags and 15 to 25 meta descriptions worth rewriting. Batch them into a single sprint — they individually take 5 minutes, collectively take 3 to 5 hours, and ship the first measurable CTR movement inside 10 days.

Step 5 — AI Overview sampling (new for 2026)

AI Overviews appear on roughly 15 to 25% of informational commercial queries in English-speaking markets as of early 2026 — exact share varies by vertical. You do not need to audit all of them. Sample 10 of the client's most commercially valuable queries manually. If AI Overview appears, multiply the keyword's composite score by 0.85 (the measured click-suppression factor on impacted queries from the March 2026 calibration).

Do not attempt to "optimize for AI Overview" in the audit deliverable. That is a separate engagement with different tactics and different measurement — see the CTR optimization guide for what can be done today, and what cannot.

Step 6 — Indexation sanity check

Open GSC Pages report. Compare indexed URL count against sitemap URL count. If indexed is within 10% of submitted, skip the rest of the indexation section — nothing actionable. If the gap is larger, note it as a checklist item but do not debug in audit time. Indexation debugging eats audit budgets and rarely surfaces a Quick Win on its own; isolate it as a separate ticket if the composite score step surfaced a Quick Win on an unindexed page.

Step 7 — Orphan Quick Win detection

Run this count against the Quick Win list from step 2: for every target page, count internal links pointing at it from other pages on the same domain. Anything with zero internal links is an orphan — the target page is winnable but the site is not directing equity to it.

The fix is adding 2 to 4 contextual internal links from high-authority pages (homepage, cornerstone content, related category pages). This is a 15-minute task per orphan page and amplifies every other optimization on the same page.

Step 8 — Opportunity isolation

Not every strong-signal keyword is a Quick Win. Some are Opportunities — genuinely winnable, but requiring content volume, new pages, or schema changes that exceed the sprint budget. Flag the top 3 Opportunities with the highest composite score. Do not ship them this sprint. Log them for the next retainer quarter and estimate hours.

Step 9 — The one-page output

Every audit closes with exactly three lists. Nothing more, nothing less:

Quick Wins — ranked by clicks-per-hour, shipped this sprint
Opportunities — ranked by score, queued for next sprint
At Risk — ranked by lost-click velocity, defended this sprint

No charts. No dashboards. No screenshots of the GSC UI. Clients do not need more data; they need the next decision.

Step 10 — Schedule the re-score cadence

The audit is not a one-time deliverable. Schedule a weekly re-score (the composite formula re-applied to the updated GSC data — 5 minutes per client) and a full re-audit every 90 days or sooner if a ranking cliff or algorithm update demands it.

What this checklist deliberately omits

Backlink audit. For freelance retainers under $3,000/month, backlink audits rarely justify their cost. Run them quarterly at the portfolio level, not per-audit.
Technical SEO deep-dive. Unless indexation failed the 10% check in step 6, there is no audit-time signal worth chasing. Clients on hosted platforms (Shopify, Webflow, WordPress managed) have technical SEO handled by the platform.
Competitor analysis. Competitors matter as SERP composition (step 2), not as a separate audit dimension. The scoring formula already incorporates them via SERP weakness.
Content gap analysis. That is a topic-cluster workshop, not an audit. Different engagement, different deliverable.

Run the audit in under 2 minutes

If you want the tiered output without running the 10 steps manually, drop a client domain into the free audit — the scoring pipeline produces the same three lists this checklist targets, deterministic, in under a minute. Use it as a first pass, then layer the manual steps (AI Overview sampling, internal link orphan detection, title tag review) on top.

For the prioritization logic that follows the audit, see how to prioritize SEO tasks across multiple client sites. For why some tools produce different outputs, see SEO Triage vs Semrush vs Ahrefs.