Your page export contains three pages. The download folder contains two PDFs. A navigation check reveals a contact page that is missing from the export. Start your website content inventory by keeping those sources together, so a neat page list does not hide the files and pages it never covered.
Record where each item was found before deciding what should happen to it. The worked example combines eight supplied source entries into six distinct paths. It produces a source-backed inventory and an explicit coverage note, not recommendations to remove content or a claim that the whole site has been checked.
Open the complete example · Download the full HTML source
Choose a boundary you can describe
Write the site or section, the date of your collection and the types of content you intend to list. Include pages and relevant downloadable assets if both belong to the task. The US Department of Energy's inventory guidance describes a document or spreadsheet listing website content, including page names, navigation labels and URLs.
For a small site, gather the existing page export, asset-library listing and navigation notes you are authorized to inspect. Preserve each original list with a source name and collection date. Record what the source includes and what you still need to check. An export of one content type is not evidence that every other type is absent.
Our original fictional example uses three sources: CMS-v1 has three page entries, Files-v1 has two document entries and Nav-v1 has three navigation entries. These are supplied teaching records, not exports from a running site. Actual collection dates, owners and response checks are unfilled.
Reconcile repeated addresses without losing the evidence
The CMS list contains /classes, /visit and /news/winter. The file list contains /files/workshop-guide.pdf and /files/access-map.pdf. The navigation notes contain /classes, /visit and /contact.
That is eight source entries but six distinct paths. The classes and visit paths occur in two lists. Keep one inventory row for each exact path and retain both source references. Eight findings do not become eight different pages. Nor does removing two repeated entries prove that every page has been found.
The example's six rows contain four pages and two files. The contact page is recorded from navigation only. Its absence from the supplied CMS list is a discrepancy to investigate, not proof that the page is broken or unpublished. There is no HTTP response evidence in these source lists.
Compare exact addresses before treating them as duplicates. Do not silently discard query strings or change letter case to make two URLs match. If your site has aliases or redirects, record the observed relationship separately before deciding which address should represent the item.
Keep the list useful for the next person
Give each row an ID, exact path or URL, content type, supplied title, source references, owner and last check. Leave an unknown owner or check date blank. A title can help someone identify an item, but it does not establish that its wording or publication status is current.
The Province of British Columbia's inventory guidance includes pages and online assets, with information such as URLs, file types, relationships and ownership. It distinguishes that inventory from an audit that evaluates accuracy, duplication and relevance. Use the fields your handoff needs rather than copying a large institutional template unchanged.
In the reading example, filter by source to see which records each list contributed. Filter by type to separate pages from files. Files-v1 plus Pages correctly produces no records. The empty view describes that selection; it does not mean the website has no pages.
Write the coverage note beside the count
For this fixture, the note is: “Six distinct paths recorded from three supplied lists. Search-only routes, other sections and unlisted assets have not been checked. Total site size is unknown.” Do not turn six recorded paths into a coverage percentage without a supported total.
Keep a follow-up for /contact: find why it appears in navigation but not in CMS-v1, then add the actual explanation and evidence. Preserve Nav-v1 as the source that first identified it. The browser reference only filters prepared rows; it does not crawl, connect to a CMS, check URLs or save changes.
Build the inventory from your actual lists
Create a website content inventory as CSV from these supplied lists.
Site/section and intended scope: [fill in]
Source names, collection dates and known limits: [paste]
Page export, asset listing and navigation observations: [paste]
Keep the original source lists unchanged. Return one inventory row per
exact supplied address with ID, path, kind, supplied title, source_refs,
owner and last_checked. Preserve all source references for repeated
addresses. Leave unknown owners and check dates blank.
Do not invent HTTP status, canonical targets, publication state or
missing pages. Do not normalize different addresses into one without
supplied evidence. Return source-entry and distinct-address counts
separately, plus source discrepancies and unchecked scope.
Make no content-removal, redirect or completeness claim.
Use the downloadable source entries to reproduce the six-row example, or start your own collection with the blank CSV header. Keep the maintained inventory beside its source lists. Once you have an inspectable set, the content audit guide covers deciding what each item needs next. The sitemap guide addresses a later, separate question: which observed URLs belong in a crawler sitemap.
Before handing over the list, ask a teammate to trace /contact back to its source and explain what remains unchecked. If they can do both, the inventory carries its evidence and its boundary into the next review.
Download the original source-entry CSV · Download the reconciled inventory CSV · Download the blank inventory CSV header