Most website audits start with opinions. The homepage feels dated. The service pages seem thin. Nobody knows whether the 2019 webinar recap still gets traffic, but nobody wants to delete it either.
A content inventory replaces that guesswork with a working record of every URL, what it does, who owns it, and what should happen next. It is useful before a redesign, during a CMS migration, after an acquisition, or whenever a website has grown faster than the team managing it.
This resource gives you the exact columns, formulas, scoring rules, and workflow needed to build one. You can copy the field list into Excel, Google Sheets, Airtable, or your project system and start today.
What is a website content inventory?
A website content inventory is a structured list of the pages and files published on a website. Each row represents one URL. Columns record facts such as status code, page title, owner, traffic, conversions, last update, canonical URL, accessibility issues, and the recommended action.
An inventory is not the same thing as a content audit. The inventory is the evidence. The audit is the judgment you make from it.
That distinction matters. A crawler can tell you that a page returns 200, has no internal links, and received 14 organic clicks in 12 months. It cannot tell you whether the page contains a contractual promise, supports a sales conversation, or must remain online for regulatory reasons. Those decisions need a human owner.
Google’s own guidance supports this evidence-plus-judgment approach. Its people-first content self-assessment asks whether content provides original value, demonstrates first-hand expertise, has a clear audience, and leaves the reader satisfied. Those questions cannot be reduced to word count or traffic alone.
Copy this 32-column website content inventory template
Use one row per indexable page at first. Add PDFs, videos, downloads, and other files in a second pass if they affect sales, support, compliance, or organic search.
| Group | Column | What to record |
|---|---|---|
| Identity | Inventory ID | Stable ID such as YWT-0001 |
| Identity | Current URL | Full live URL |
| Identity | Page type | Home, service, product, location, article, case study, utility, legal, or file |
| Identity | Page title | Browser and search result title |
| Identity | H1 | Main visible heading |
| Identity | Parent section | Site section or navigation branch |
| Technical | HTTP status | 200, 301, 404, 410, or other response |
| Technical | Index status | Indexed, excluded, blocked, noindex, or unknown |
| Technical | Canonical URL | Declared canonical and Google’s selected canonical if different |
| Technical | In sitemap | Yes or no |
| Technical | Internal links in | Count of internal links pointing to the URL |
| Technical | Redirect target | Destination if the page moves or is removed |
| Performance | Users | Organic and total users for the selected period |
| Performance | Search clicks | Google organic clicks for the selected period |
| Performance | Search impressions | Google organic impressions for the selected period |
| Performance | Conversions | Leads, sales, bookings, downloads, or another business event |
| Performance | Conversion rate | Conversions divided by eligible sessions |
| Performance | Backlinks | Referring domains and important inbound links |
| Quality | Primary audience | The specific buyer, customer, candidate, partner, or employee served |
| Quality | User job | What the visitor needs to decide, learn, compare, or complete |
| Quality | Business purpose | Revenue, support, trust, compliance, recruiting, or navigation |
| Quality | Accuracy | Accurate, needs verification, or incorrect |
| Quality | Original value | Evidence, examples, tools, data, or experience not found everywhere |
| Quality | Readability | Clear, needs editing, or confusing |
| Quality | Accessibility | Pass, issues found, or not tested |
| Governance | Content owner | Person accountable for accuracy |
| Governance | Subject expert | Person who can approve technical claims |
| Governance | Last meaningful update | Date the main content last changed |
| Governance | Review frequency | Monthly, quarterly, twice yearly, yearly, or event-driven |
| Decision | Score | Weighted score from the model below |
| Decision | Action | Keep, improve, merge, redirect, remove, or investigate |
| Decision | Notes and deadline | Reason, dependencies, assignee, and target date |
Do not omit the owner, purpose, and redirect-target columns. Those are the fields that turn a spreadsheet into an operating tool. A long list of URLs without accountability becomes stale almost immediately.
Where to get the data
No single export contains everything you need. Build the inventory in layers and preserve the original source in separate tabs so somebody can verify it later.
1. Crawl the live website
Use a crawler to collect URLs, status codes, titles, headings, canonicals, index directives, sitemap inclusion, and internal link counts. HTTP response codes have defined meanings: the current HTTP standard identifies 301 as Moved Permanently and 404 as Not Found. Record the actual response instead of relying on what the browser appears to show.
Start the crawl from the homepage, then compare its URL list with the XML sitemap. A crawl mainly finds linked pages. A sitemap can reveal pages that are live but poorly connected to the rest of the site.
Google says a sitemap’s lastmod value should represent the date of the last significant page update, such as a change to main content, structured data, or links. Treat it as a lead, not unquestionable evidence, because publishing systems do not always maintain it correctly.
2. Export search data
Export page-level clicks and impressions from Google Search Console for the longest useful comparison period. Twelve months smooths out seasonality for many businesses. Compare year over year when products, tracking, and market conditions are similar enough to make the comparison meaningful.
For important or confusing URLs, use Search Console’s URL Inspection tool. Google’s documentation says it reports information about the indexed version of a page and allows a live URL test. The Page Indexing report also explains why URLs may be excluded, but Google cautions that not every non-indexed URL represents a problem.
Record both the declared canonical and Google’s selected canonical when they disagree. Google describes redirects and rel="canonical" as strong signals, while sitemap inclusion is a weaker canonicalization signal. A disagreement can expose duplicate pages, inconsistent internal links, or a migration mistake.
3. Add analytics and conversion data
Export landing-page users, sessions, and meaningful business events from your analytics platform. Avoid treating every event as equal. A pricing request, completed purchase, qualified call, and newsletter signup do not carry the same business value.
GA4 defines an engaged session as one that lasts longer than 10 seconds, has a key event, or includes at least two page or screen views, and calculates engagement rate as engaged sessions divided by sessions. That can help with diagnosis, but it should not determine a page’s fate by itself. A phone-number page can work in seconds. A long article can hold attention without producing revenue.
4. Add backlinks and off-site dependencies
Add referring domains from the backlink tool you already use, then manually flag links from important customers, trade publications, universities, government sites, partners, and industry directories. A page with little traffic may still carry authority or support a valuable relationship.
Also check QR codes, paid ads, email automations, sales documents, printed brochures, customer portals, and software integrations. A URL can have no visible internal links and still be in active use outside the website.
5. Interview the people who use the content
Meet with sales, support, operations, legal, HR, and product owners. Ask which pages they send, which claims cause confusion, which files customers request, and what has changed in the business. This step finds value that analytics misses.
The current Content Marketing Institute B2B research was based on 1,229 global responses collected in 2025. Its recurring operational theme is that content performance depends on strategy, systems, and cross-team processes, not publication volume alone. Your inventory should reflect those same operating realities.
Score each page on five dimensions
A score creates consistency, but it should guide discussion rather than automate deletion. Rate each dimension from 0 to 4.
| Dimension | Weight | 0 means | 4 means |
|---|---|---|---|
| Business value | 30% | No defined purpose | Directly supports revenue, retention, compliance, or a critical task |
| User value | 25% | No clear audience need | Answers a real need clearly and completely |
| Performance | 20% | No useful traffic, action, or links | Strong qualified traffic, actions, or authoritative links |
| Accuracy | 15% | Wrong or risky | Current, verified, and approved |
| Findability and accessibility | 10% | Blocked, confusing, or inaccessible | Easy to find, understand, navigate, and use |
Use this formula:
Weighted score = (Business × 0.30) + (User × 0.25) + (Performance × 0.20) + (Accuracy × 0.15) + (Findability × 0.10)
The result ranges from 0 to 4. Add three override fields beside it: legal hold, active campaign, and known external dependency. If any override is yes, the page requires review regardless of score.
Accessibility belongs in the score because content quality includes whether people can use the page. WCAG 2.2 requires pages to have titles that describe their topic or purpose, provides criteria for headings and labels that describe topic or purpose, and requires text alternatives for non-text content under Success Criterion 1.1.1. These checks belong in the page review, not in a separate spreadsheet nobody sees.
Turn the evidence into six actions
Every reviewed row should end with one of six decisions and a plain-language reason.
- Keep: The page is accurate, useful, and performing its job. Assign an owner and next review date.
- Improve: The purpose is valid, but the page needs better evidence, clearer copy, updated facts, stronger accessibility, or a better conversion path.
- Merge: Two or more pages serve the same audience and job. Consolidate the useful material into one destination, update internal links, and redirect retired URLs.
- Redirect: The old URL has links, traffic, bookmarks, or known off-site use, but another page now satisfies the need.
- Remove: The content has no continuing purpose, value, dependency, or suitable replacement. Return the technically appropriate response and remove internal references.
- Investigate: Ownership, legal requirements, performance data, or dependencies remain unclear. Name the person who must resolve it and set a deadline.
Suggested score bands are 3.2 to 4.0 for keep, 2.2 to 3.19 for improve or merge, 1.2 to 2.19 for redirect or investigate, and below 1.2 for removal review. These bands are management rules, not industry benchmarks. Adjust them after scoring 20 representative pages so they match the risk and economics of your website.
Never redirect every removed page to the homepage. A redirect should send the visitor to a relevant replacement. When there is no replacement, a clear 404 or 410 response is more honest than an unrelated destination. The HTTP specification defines 410 Gone for cases where the resource is intentionally unavailable and likely permanently so.
A practical audit workflow for 100, 1,000, or 10,000 pages
For a site under 100 pages, review every row. The time spent sampling and designing automation will not save much.
For 100 to 1,000 pages, group URLs by page type and section, then review high-value pages first. Start with conversion pages, pages receiving organic clicks, pages with backlinks, paid campaign destinations, and legally sensitive content. After that, work through the remaining groups in batches.
For more than 1,000 pages, stratify the sample. Pull examples from every template, traffic band, age band, language, product line, and index state. Use the sample to discover patterns, then apply rules to matching groups. Manually inspect exceptions and high-risk pages before publishing bulk changes.
A clean sequence looks like this:
- Freeze the column definitions and measurement period before exports begin. Combine crawler, sitemap, Search Console, analytics, backlink, CMS, and stakeholder data using the normalized URL as the key.
- Score a calibration set with at least two reviewers. Resolve disagreements, tighten the rubric, then assign sections to named owners.
- Approve actions in batches. Build a redirect map, update links and navigation, make content edits, run quality checks, launch, and monitor indexing, traffic, conversions, and errors.
Keep the raw exports. If a number looks wrong later, you need to know whether the source changed, the join failed, or somebody overwrote the cell.
Quality checks before any deletion or migration
Before approving a destructive action, check whether the URL has conversions, backlinks, organic queries, internal links, campaign use, legal obligations, customer documentation, translated versions, structured data, or a close replacement. Review desktop and mobile layouts, forms, downloadable files, video captions, image alternatives, page titles, headings, and link wording.
Then test the planned destination as a visitor would. Does it answer the same need? Can the visitor complete the next step? Is the replacement accessible? Is the claim still true?
After implementation, crawl both the old and new URL sets. Confirm intended redirects reach a final 200 page without chains. Check canonicals, index directives, sitemaps, internal links, forms, analytics events, and Search Console. Monitoring should continue long enough to cover the site’s normal crawl and sales cycle.
Content inventory example
Imagine a commercial HVAC company has three pages:
| URL | Evidence | Score | Action |
|---|---|---|---|
/rooftop-unit-repair/ | Generates quote requests, accurate, strong local links | 3.7 | Keep; review every six months |
/commercial-ac-repair/ | Same intent as rooftop page, weaker copy, some backlinks | 2.4 | Merge useful sections, then redirect |
/2020-reopening-policy/ | No traffic, obsolete instructions, no links or dependencies | 0.6 | Remove after legal and operations approval |
The middle page is why simple traffic rules fail. Deleting it would waste its backlinks. Keeping both pages would preserve duplication. Merging and redirecting retains the useful material and gives visitors one stronger answer.
Who owns the inventory after the audit?
The website owner should maintain the master file, but accuracy belongs to named subject experts. Sales may own offer and qualification claims. Operations may own service availability. HR may own job and benefit content. Legal may own policy language. Marketing can coordinate reviews without pretending to be the expert on every statement.
Set review dates based on risk. Pricing, product availability, staff, locations, deadlines, and legal terms may need event-driven or quarterly review. Evergreen educational pages may need annual review. A visible “last updated” date is useful only when it reflects a meaningful review, the same principle Google applies to sitemap lastmod dates.
The inventory should also become part of launch and change management. New pages need an owner, purpose, measurement plan, and review date before publication. Retired products need a content action as part of the retirement checklist. Staff departures should trigger reassignment of the pages they owned.
Frequently asked questions
How often should a website content inventory be updated?
Update technical and performance fields on a regular schedule, often monthly or quarterly, and review content according to its risk. Trigger an immediate update after a redesign, migration, acquisition, product change, rebrand, location change, or major regulatory update.
Should low-traffic pages be deleted?
No. Low traffic is one signal. A page may serve a small but valuable audience, convert offline, answer a support question, satisfy a legal need, or hold backlinks. Delete only after checking business purpose, user value, dependencies, and replacement options.
What is the difference between a content inventory and a sitemap?
A sitemap helps search engines discover preferred URLs. A content inventory is an internal management record containing technical, performance, quality, ownership, and decision data. The sitemap is one input to the inventory.
Can AI complete a content audit?
AI can classify pages, summarize copy, flag similarities, and propose actions. It cannot reliably know undocumented contracts, sales usage, regulatory duties, or political ownership inside the company. Use it to reduce sorting work, then require accountable human approval for changes.
What should we do first if the spreadsheet is overwhelming?
Start with pages tied to revenue, paid campaigns, organic traffic, backlinks, compliance, and customer support. Assign every one an owner and action. That first pass reduces the largest business risks while the rest of the inventory is still being built.
Make the inventory pay for itself
The spreadsheet is not the deliverable. The deliverable is a website with fewer dead ends, clearer ownership, safer updates, and better pages for the people deciding whether to call, buy, apply, or trust you.
If your inventory reveals a website that has outgrown its structure, talk with Your Web Team about the next step. We can help turn the evidence into a practical redesign or improvement plan.
