How to Run an SEO Content Audit: Inventory and Scoring

Introduction: how to run a content audit in 2026 to optimise what you already have (and complement your SEO audit)

If you have already structured your SEO audit, the next step is often to deepen the analysis of your published pages. A well-executed content audit transforms a collection of URLs into clear editorial decisions (update, consolidate, remove), grounded in evidence (Search Console, Analytics) and a consistent qualitative framework.

In 2026, it is also a pragmatic way to protect performance in increasingly "closed" SERPs (zero-click, AI-assisted answers): according to Semrush (2025), 60% of searches end without a click. Optimising existing content (clarity, structure, evidence, intent) becomes a priority lever—often more cost-effective than publishing "more and more" new pages (HubSpot).

Content audit: definition, objectives and scope (qualitative + quantitative)

An SEO-focused editorial audit is a systematic review of already-published pages to identify what performs, what underperforms, and what competes with what. It combines:

• A quantitative view (visibility, engagement, conversions, trends) using Google Search Console and Google Analytics.
• A qualitative view (clarity, completeness, freshness, evidence, structure, role in the journey) using an evaluation grid.

The goal is not to "make it longer" for the sake of it, but to make each page useful, unique, aligned to an intent and actionable for your roadmap (HubSpot, Semrush). As a rough benchmark, some methodologies recommend prioritising in-depth content beyond 800 words when the intent requires it—without turning this into an arbitrary rule.

What this analysis covers (and what it is not designed to replace)

This analysis focuses on what already exists: inventory, performance, quality, overly thin content, overly similar content, and consolidation decisions. It is not meant to replace a technical diagnosis or a full competitive analysis: it draws on them, but remains focused on page-by-page editorial decisions and their measurable impact.

Which page types to audit: blog posts, landing pages, categories, resources and FAQs

To avoid a "blog-only" bias, include pages with clear business value and pages likely to generate SEO side-effects:

• Blog posts, guides, glossaries, resource pages.
• Landing pages (acquisition, demo, contact), offer pages.
• Category and listing pages (depending on your model), FAQ pages.
• Archives, tags, variants, "print" pages or templates likely to duplicate titles and H1s.

Define the scope before collecting URLs (Semrush, HubSpot), even if you start with a subfolder (e.g. "/blog") for a first cycle.

Expected deliverables: an existing content inventory, quality scoring, prune/merge/redirect decisions and a prioritised backlog

A useful audit typically produces:

• A documented inventory (spreadsheet) listing URLs and key attributes.
• Quality scoring (quantitative + qualitative) with explicit rules.
• Operational decisions per page: prune, merge, redirect (and/or simple updates).
• A prioritised backlog (impact × effort × risk) plus before/after validation criteria.

Step 1 – Build the existing content inventory (the audit's foundation)

The inventory is the bedrock. Without a reliable URL list, you risk making changes at random, missing variants, or overlooking structural duplicates.

Consolidate URLs: indexed, non-indexed and variants

Aim for a single list of URLs to audit by aggregating:

• Pages in your sitemap (useful for a structured list).
• Pages surfaced in Google Search Console (queries → pages generating impressions and clicks).
• SEO entry pages identified in Google Analytics.
• Obvious variants (parameters, path duplications, very similar pages) to prepare for proximity analysis.

On large sites, this step quickly reduces noise (secondary URLs, old campaign pages, near-empty pages) and secures everything that follows.

Normalise attributes: page type, topic, persona, intent and status

To make the analysis workable as a team, standardise your columns. Example of a minimal baseline (HubSpot, Semrush):

• URL, type/template, theme/cluster, page goal.
• Target persona, dominant intent (inform, compare, act).
• Publish date / last update.
• Main target query (if known), provisional status (to be qualified).
• Metadata: title, meta description, H1 (useful for spotting repetitions).

This standardisation then speeds up arbitration (near-twin pages, editorial orphans, scattered coverage).

Step 2 – Measure performance: what to collect and how to read it without bias

The classic trap is judging a page purely on traffic. A page can be low in sessions but critical for conversion, or highly visible but poorly aligned (lots of impressions, few clicks). Cross-referencing Search Console and Analytics remains the most robust foundation.

Google Search Console: impressions, clicks, CTR, queries and pages

From the "Search results" report, collect per URL:

• Impressions (visibility), clicks (captured demand).
• CTR (promise quality and SERP fit), average position (a broad signal).
• Main queries (and dispersion) to detect cannibalisation and intent mismatch.

A page ranking positions 6–20 can be a strong "quick win" lever if editorial quality is there (consistent with the prioritisation approach described in the parent article).

Google Analytics: engagement, journeys, conversions and entry pages

In GA4, evaluate the page as a step within a journey:

• Entry pages (SEO landing pages), engagement (time, scroll/events if configured), internal navigation.
• Conversions or micro-conversions depending on the goal (lead, demo, sign-up, CTA click).
• Common mismatches: high traffic but low conversion (offer, evidence or CTA issue) vs low traffic but high conversion (discoverability issue).

UX context to keep in mind: according to Google (2025), 53% of users leave a mobile page if it takes longer than 3 seconds to load.

Segment the analysis: by template, theme, maturity and funnel role

Segment to avoid misleading "average" conclusions:

• By template (article, landing page, FAQ, listing) to identify template-level issues (duplicated titles and H1s, repeated sections).
• By theme/cluster to spot where coverage is too scattered (duplicates) or too thin (missing angles).
• By maturity (new page vs evergreen) so you do not penalise recent content.
• By funnel role (awareness, consideration, decision) to select the right KPIs (Semrush).

Step 3 – Quality scoring: build an actionable content evaluation framework

Quality scoring has one purpose: to make decisions repeatable and debatable. It removes "gut-feel" arbitration and speeds up prioritisation.

Quantitative scoring: visibility, traffic contribution and upside potential

A simple model assigns points across three axes:

• Visibility: impressions, average position, presence on strategic queries.
• Captured demand: clicks, CTR, share of organic traffic per page.
• Business outcome: conversions, assisted contribution (when measurable).

Add a potential indicator (e.g. pages close to the top 10, or pages with strong impressions but weak CTR) to steer quick wins.

Qualitative scoring: clarity, completeness, evidence, freshness and structure

Build a short, scored grid (e.g. 0–2 or 0–5) based on auditable criteria (HubSpot):

• Immediate clarity: definition upfront, explicit promise, appropriate vocabulary.
• Completeness: does it cover expected sub-questions (steps, use cases, limits, mistakes) without drifting?
• Evidence: sourced numbers, examples, methodology, trust signals.
• Freshness: up-to-date stats, current examples, non-outdated information.
• Structure: clear H2 and H3 headings, lists, scanability (also helpful for being cited by generative engines).

Intent alignment: when the page attracts clicks but does not meet the need

A strong signal in an editorial audit is the gap between "what Google shows" and "what the page delivers". Typical patterns include:

• The page ranks for informational queries but pushes a CTA too quickly, without answering clearly.
• The page targets a comparison intent but offers no criteria, tables or scenarios.
• The page captures a specific query but stays generic (blurred angle, unfulfilled promise).

Common situations: low CTR, stagnant rankings, weak conversions

• High impressions + low CTR: the promise needs revisiting (title and meta description), or the format does not match what the SERP expects. According to MyLittleBigWeb (2026), an optimised meta description can increase CTR by 43%.
• Stagnation: the content is decent, but not the "reference" (missing evidence, depth or structure).
• Weak conversions: not enough reassurance, a poorly placed CTA, an unclear next step, or a promise that does not align with the offer.

Step 4 – Thin content evaluation: identification, causes and operational thresholds

"Thin" content is not just about length. It is content that does not provide enough useful, differentiating information to justify its place (and indexing) against competitors' pages—and sometimes against your own pages.

Thin content vs useful short content: how to decide without an arbitrary rule

Use a practical rule: if a short page fully satisfies the intent (simple definition, precise FAQ, short procedure), keep it. However, when the intent implies a decision, a comparison or a methodology, a page that is too short often becomes thin because it fails to cover expected sub-questions (HubSpot, Semrush).

Common signals: weak topical coverage, low differentiation, low engagement

Combined signals (verify together):

• Coverage is insufficient: terms are not defined, steps are missing, no examples.
• Differentiation is low: the page repeats another article, or relies on generic statements.
• Engagement is weak: time on page far below expected reading time, high bounce rate, little internal navigation (indicators used in Analytics-led editorial audits).

Action plans: enrich, refocus, re-brief or deindex

• Enrich when the page has legitimate intent and upside (positions 6–20, impressions). Add definitions, evidence, "common mistakes" sections, FAQ.
• Refocus when the page mixes multiple intents (it attracts, but "doesn't land").
• Re-brief if the angle is wrong: a new outline beats a patchwork.
• Deindex / remove properly if the page has no role, no value and creates noise (to be arbitrated with the safeguards in Step 6).

Step 5 – Duplicate content analysis: detect duplicated and overly similar content

Duplication is not just copy-paste. In practice, issues often come from "too similar" pages that compete and dilute signals (impressions, clicks, internal links).

Internal duplication: similar pages, templates, variations and structural repetition

Common cases:

• Near-twin pages on close topics, published at different times.
• Templates duplicating title and H1 or entire blocks across dozens of URLs (tags, archives, listings).
• Variants (parameters, facets, filters) creating near-identical pages.

The right reflex is to determine whether the similarity is legitimate (specific value, distinct intent) or parasitic (same useful information, same promise).

External duplication: risks, diagnostics and editorial control points

The audit should also check risks of external reuse (plagiarism) and partially copied content, because it complicates attribution of the "reference" page. Editorial checkpoints include uniqueness of viewpoint, the presence of original examples, and the ability to prove claims (method, numbers, named sources).

Limit SEO impact: clarification, canonicalisation, rewrites and consolidation

Your options depend on the cause:

• Clarify each page's intent (one intent = one reference page whenever possible).
• Canonicalise when legitimate variants exist but one version should carry the main signal.
• Rewrite to truly differentiate (angle, use cases, persona, evidence).
• Consolidate when two pages answer the same need.

Step 6 – Prune/merge/redirect decisions: consolidate, preserve and avoid losses

Once pages are qualified, the challenge is safe execution: preserve what has value (traffic, links, conversions), reduce internal competition, and avoid regressions.

When to remove (prune) without losing value: criteria and safeguards

Remove with caution. Typical criteria:

• Outdated page (campaign ended) with no evergreen value.
• Page with no role in the journey, no traffic, no conversion, and no editorial usefulness.
• Page creating noise (obvious duplicate) and cannot be differentiated at reasonable cost.

Safeguards: verify it does not receive important internal links, does not serve as an SEO entry page, and plan a coherent redirect where needed (Semrush).

When to merge: group by intent and preserve the strongest signal

Merge when multiple pages target the same intent. A solid pattern:

• Select a canonical page (most relevant, most stable, or strongest signals).
• Bring across the best of each page (examples, sections, FAQ, evidence) whilst avoiding repetition.
• Update internal linking to point to the consolidated page.

When to redirect: relevance rules, targeting and post-change controls

A redirect (often 301) transfers value from a removed or merged URL to the closest intent match (Semrush). Avoid "catch-all" redirects to the homepage or an irrelevant category: this damages the experience and confuses signals.

After publishing, check in Search Console: indexing, errors, impression and click evolution, and associated queries.

Pre-execution checklist: internal links, anchors, content, tracking and monitoring

• Update internal links (anchors, source pages, hubs).
• Review title, H1 and meta description on the target page (the promise).
• Keep (or reinstate) sections that captured long-tail queries.
• Validate tracking (events/CTAs) so you can compare before and after.
• Set measurement checkpoints at day 14, day 30 and day 90 (depending on volume).

Step 7 – Prioritise actions after the audit: an impact/effort-driven method

At the end, you will have too many possible actions. Prioritisation prevents editorial effort from being spread too thin and aligns teams around a rational execution order.

Impact × effort × risk matrix: organise quick wins and strategic workstreams

Score (1–5) across:

• Impact (SEO, conversions, funnel role, business importance).
• Effort (simple update vs rewrite vs multi-page consolidation).
• Risk (traffic loss, technical dependencies, intent uncertainty).

This is a straightforward way to produce an initial roadmap of 15–20 actions, then iterate.

Prioritise by cluster: editorial coherence, internal linking and pillar pages

Prioritising by cluster prevents you from improving an isolated page with no editorial support. The goal is to strengthen a reference page and its satellites, using clear internal linking (hubs → child pages → specialist pages), which also reduces cannibalisation.

Plan production: sequence updates, consolidations and new pages

A robust sequence often looks like:

1. Update high-stakes pages (existing visibility, conversions) and fix promises (CTR).
2. Consolidate close-intent duplicates (merge + redirect).
3. Only then produce new content to fill genuine gaps (content gaps), so you do not recreate internal competition.

Interpret results: turn analysis into measurable decisions

An audit is successful when it produces traceable trade-offs and success criteria (before and after). Semrush highlights the value of comparing post-change performance against a baseline to judge whether the effort was justified.

Read the signals: visible but not clicked, converting pages, high-upside pages

• Visible but not clicked: rework the promise (title and meta description), clarify intent, enrich expected sections.
• Converting pages: protect, strengthen and improve discoverability (internal linking, structure, completeness).
• High-upside pages: mid positions + clear intent = strong candidates for priority editorial optimisation.

Avoid false positives: seasonality, query mix and SERP changes

Before labelling a page "problematic", check:

• Seasonality (normal peaks and troughs) over a comparable period.
• SERP changes (new formats, AI Overviews) that shift CTR and clicks.
• Query mix: a page may perform on unexpected secondary queries.

Set up monitoring: KPIs by page, by intent and by period

Define three management layers:

• By page: impressions, clicks, CTR, conversions.
• By intent: information vs comparison vs action (with adapted KPIs).
• By period: before and after at day 30 and day 90, then quarterly reviews of strategic pages.

To establish realistic benchmarks and feed internal comparisons, you can use our SEO statistics.

Tools and automation: structure a content audit without bloating the process

Done manually, the exercise quickly becomes time-consuming (multiple sources, arbitration, consolidation). The aim is to keep editorial control whilst automating repetitive work.

Minimum stack: Search Console, Analytics and an audit grid

The recommended minimum stack (Semrush, HubSpot) is threefold:

• Google Search Console (visibility and queries).
• Google Analytics / GA4 (engagement and conversions).
• An inventory sheet plus a qualitative scoring grid to make decisions traceable.

Automate with a personalised-AI SEO & GEO platform: what must stay under control

Automation makes sense for extracting metadata, identifying thematic proximity, suggesting clusters and preparing briefs. However, some decisions must remain with your team: removing URLs, choosing the canonical page, and validating promises (title and H1) against brand positioning.

If you want an equipped foundation, the Incremys platform centralises analysis, planning and production with personalised AI, whilst leveraging Search Console and GA4 data.

Incremys focus: audit what exists and produce faster—without compromising quality

The value of industrialising this approach is to run a full cycle: inventory → diagnosis → decisions → backlog → production → measurement.

Automate detection: thin content, similar content, cannibalisation and opportunities

Incremys helps structure large-scale analysis by highlighting:

• Pages whose coverage is too weak for the intent (signals of insufficient content).
• Similar pages likely to compete (cannibalisation or dilution risk).
• Improvement opportunities on already-visible pages (high impressions, improvable CTR).

The goal is to turn signals into documented decisions, not into alert lists that are hard to action.

Identify content gaps and generate SEO & GEO briefs with the Content Production module

After consolidating what exists, creation becomes more efficient when it fills real gaps. The content production module is designed to identify gaps and generate structured SEO & GEO briefs (angle, intent, outline, evidence elements), so you can scale production without losing coherence.

Prioritise in an editorial plan: SEO/GEO potential, effort and expected ROI

Once the backlog is established, execution is the real challenge. Incremys helps you run an editorial plan that accounts for each topic's SEO and GEO potential, the effort needed for updates or consolidation, and performance monitoring that stays ROI-focused (pages improving, pages stagnating, pages needing re-qualification).

When to use the SEO audit module to cross-check content, technical factors and competitive signals

When "content" signals are not enough (stagnation despite a strong page, major SERP volatility, doubts about signal consolidation), you can cross-check with a broader diagnosis via the SEO audit module. The objective remains the same: evidence first, then a prioritised roadmap—without multiplying marginal optimisations.

To place this work back into the overall methodology, you can also revisit the content component within the full SEO audit.

To go further and get a complete framework (inventory, scoring, decisions and prioritisation), see also our resource on content audit scoring.

FAQ: common questions about content audits

What exactly is a content audit?

It is a structured process that reviews a website's existing pages to identify what should be kept, improved, consolidated or removed. It combines data (Search Console, Analytics) with a qualitative grid to produce a measurable action plan (HubSpot, Semrush).

What are the key elements to check during an editorial audit?

• Dominant intent and a consistent promise (title and H1).
• Uniqueness and differentiation (avoid overly similar pages).
• Completeness (definitions, steps, use cases, limits, FAQ).
• Evidence (sourced numbers, examples, method).
• Freshness (up-to-date information).
• Performance (impressions, clicks, CTR, conversions) and role in the journey.

How do you run a content audit step by step?

1. Define objectives, scope and decision rules.
2. Build the inventory (URLs + metadata + normalised attributes).
3. Collect Search Console and Analytics data.
4. Score (quantitative + qualitative) and detect thin and overly similar content.
5. Decide per page (update, merge, remove, redirect).
6. Prioritise (impact × effort × risk), execute and measure before and after.

Which tools should you use for a content audit?

The minimum baseline is Google Search Console and Google Analytics, plus an inventory sheet and a qualitative grid (HubSpot, Semrush). To industrialise detection (thin content, proximity, gaps) and generate briefs, a platform such as Incremys can automate part of the workflow whilst keeping human validation.

What deliverables should you expect from a content audit?

In practice: an inventory (spreadsheet) with attributes and metadata, actionable scoring, a status per page (keep, update, merge, remove/redirect), a prioritised backlog and before and after measurement criteria.

How should you interpret the results of a content audit?

Interpret at a macro level first (trends by template and cluster), then case by case:

• High impressions + low CTR = the promise or format needs revisiting.
• High traffic + low conversion = lack of evidence and CTA or intent mismatch.
• Low traffic + high conversion = improve discoverability (internal linking, structure, consolidation).

How do you prioritise actions after a content audit?

Use an impact × effort × risk matrix, then group by cluster to strengthen a reference page and its satellites. The best first workstreams are often already-visible pages (mid rankings) with a clear qualitative gap, because the upside is faster to measure.

How do you evaluate thin content without getting it wrong?

Avoid an "X words" rule. Evaluate instead: (1) intent (what the user expects), (2) completeness (sub-questions covered), (3) differentiation (unique value), (4) engagement signals. A short page can be excellent if it fully answers the need.

How do you diagnose duplicated content with a duplicate content analysis?

Start by detecting internal proximity (same promises, same outlines, same titles and H1s), then check whether pages truly target different intents. If not, consolidate (merge) into one canonical page and redirect the old URLs cleanly. Then monitor in Search Console how impressions and clicks reallocate.

When should you make prune/merge/redirect decisions?

As soon as you observe: (1) pages with no role and no value, (2) near-twin pages competing, (3) outdated content, or (4) clusters where multiple URLs carry the same intent. The aim is to reduce noise, concentrate signals and make the editorial structure clearer.

What are common mistakes when running a content audit?

• Auditing only the blog and forgetting business pages.
• Deciding based only on traffic (ignoring conversions and intent).
• Removing pages without a coherent redirect or internal linking updates.
• Creating new content before consolidating duplicates.
• Not defining a qualitative grid, so decisions are not repeatable.
• Measuring too early (no before and after window) or without segmentation.

How much does a content audit cost in 2026?

Cost mainly depends on URL volume, the depth of qualitative review (simple triage vs detailed consolidation), and the level of deliverables (backlog + briefs). In-house, cost is largely measured in people time (collection, review, arbitration). With a supplier, pricing varies widely by scope and output; the most reliable approach is to request a quote based on: number of URLs, expected scoring depth, and the number of "merge and redirect" decisions to define.

How often should you run a content audit?

An annual cadence is often considered a minimum (Semrush, SEO editorial approaches). According to Semrush's "State of Content Marketing 2023" report, 61% of marketers run audits twice a year or more. For large sites or fast-moving sectors, a quarterly rhythm on strategic pages (evergreen, business pages) can be relevant, with a less frequent full review.