A cassette review can get the recording history right while sending readers to an expired ordering page. Its audio player might be dead, too. Each problem calls for a different response: check the historical claim, remove or date the sales detail, and replace the player or explain what is missing. Rewriting the whole piece could erase what it still gets right.
A content audit records what each page contains, what it is meant to do, what can be verified and what needs attention. For an independent music journal, label or archive, it can also show where a subject has no adequate page. The goal is not to make every old article sound new. It is to tell a dated document from a page that misleads someone reading it now.
Set the scope before opening a spreadsheet
Choose a section of the site you can actually inspect. A journal with years of reviews could start with its most visited articles, pages that discuss current availability, and pieces about artists or releases for which corrections have arrived. A small label might begin with its release catalogue, contact details, shipping notes and artist biographies. Set a review date and decide who can approve factual changes. Without those limits, an audit becomes a plan to improve everything indefinitely.
Keep the inventory simple and consistent. A spreadsheet will do. Give every URL a row, even when two pages appear to be duplicates. Useful columns include:
- URL and page title: record the actual address and visible heading, not a proposed replacement title.
- Page type and purpose: note whether it is a review, interview, release page, guide, archive entry or contact page, and what a visitor should learn or do there.
- Publication date and last substantive update: distinguish a factual correction from a cosmetic edit.
- Evidence and ownership: identify who can check the claims and who has authority to change them.
- Condition: assess factual accuracy, sources, media, accessibility, links and signs that the page no longer serves its purpose.
- Decision, priority and status: record the action, the person responsible and the next step.
An export from the publishing system can supply titles, URLs and dates, but inspect the live pages as well. An export will not catch a caption that contradicts the text beside it or a release page embedding the wrong track. Search results may turn up old pages missing from the site menu. Mark what you have reviewed so you do not mistake an unfound page for a gap.
Judge each page by its job, not its age
A contemporary account of a performance may be valuable because it records what someone heard at the time. A new guide, meanwhile, can fail by leaving a basic term unexplained. Write one sentence describing each page’s purpose, then read it as someone arriving without prior context would. Does it do that job?
Check claims at the level where errors happen
Check names, titles, dates, formats, credits and quotations individually. On a release page, compare the label name, physical format and track listing with documentation the editor can rely on. In a review, separate listening judgments from verifiable release facts. A listener’s different interpretation does not invalidate the writer’s account; an incorrect release year still needs fixing.
Mark anything you cannot confirm as unverified rather than substituting a plausible detail. A vanished shop listing does not establish that a cassette is permanently unavailable. If current stock cannot be checked, remove the availability claim or make clear when it was true. Keep a record of the evidence behind corrections in your working notes, even when the article does not need a source note for every detail.
Read the page as a visitor would
Open it on a phone and at desktop width. Can a reader find the main answer without working through a long preamble? Do the headings describe what follows, and is it clear which recording each paragraph discusses? Check whether the audio embed loads and whether the text gives listeners enough context if it does not. Look at image descriptions, transcripts where relevant, and links whose wording should still make sense out of context. A page can work technically and still make its subject difficult to follow.
Read neighboring pages, too. Two guides might repeat the same explanation while leaving another question unanswered. A review might promise background found in an interview without pointing readers to it. Those are structural problems, not automatic reasons to add more words.
Separate updates from gaps
An update improves a page that already has the right purpose. A gap is a reader need the existing pages do not meet. Confusing the two can bloat an article: a small correction turns into an unrelated history, while the question that history should answer remains hard to find.
For each possible gap, write down the missing question and the pages a visitor would currently need to consult. Say a journal has several reviews of field-recording works but nothing on how location, microphone placement or editing shape what the listener hears. A focused guide might help. Adding the same paragraph about recording practice to every review would not.
Use editorial evidence to identify gaps: recurring reader questions, searches within the site, navigation paths that end without an answer, or a release page missing format information a buyer needs. Search data may suggest how to phrase a page, but low traffic does not make a specialist subject unworthy of coverage. Nor does interest in a broad term oblige a journal to write a thin overview outside its expertise.
A useful gap note names the question, the intended page and the evidence the editor can obtain. “Explain the difference between a documented performance and a later studio edit” is workable if recordings, credits or participant accounts can support it. “Cover the entire underground” is not an editorial assignment.
Choose an action instead of a vague improvement score
Give each page a decision before you finish reviewing it. Otherwise, “needs work” can sit in a spreadsheet forever.
| Decision | Use it when | Immediate task |
|---|---|---|
| Keep | The page is accurate and serves its purpose. | Record the review date and leave it alone. |
| Correct | A discrete fact, link, caption or media item is wrong. | Verify and amend that item. |
| Expand | The page answers the right question but lacks necessary context. | Outline what is missing and how it can be supported. |
| Combine | Two pages substantially duplicate one purpose. | Identify what each contributes before choosing a main page. |
| Retire | The page cannot be maintained or no longer has a defensible purpose. | Check references and decide what should replace access to it. |
| Investigate | An important claim lacks adequate evidence. | Assign verification rather than guessing at a fix. |
After a genuine consolidation, redirecting an old URL may make sense if the destination serves the same need. Sending a specific interview to a broad category page loses context and frustrates anyone citing the original. Archival material may be better left in place. If a page documents an earlier state of a project, keep its original date clear and add a dated note when later information changes how it should be read. The distinction between preserving a record and altering it comes up in the investigation of wasted fjpg, a 1KB JPEG ghost associated with the Japanese noise underground: an uncertain artifact calls for careful description, not a tidier invented account.
Prioritise risk before polish
Fix a wrong artist credit before smoothing an introduction that merely feels slow. Consider four things when setting priorities: harm if the information is wrong, reach or likelihood of discovery, effort required, and confidence in the proposed fix. They are prompts for judgment, not numbers to multiply into a score.
Contact details, purchase instructions, rights and attribution matter even on lightly visited pages. A broken player on a frequently read listening guide may be urgent because it stops the page doing its job. A long essay with few visitors and sound evidence might need nothing at all. Put important claims you cannot verify into an investigation queue; speed is no substitute for a reliable correction.
Make revisions traceable
Before changing a substantial passage, save the old wording in the publishing system’s revision history or your editorial records. Note what changed, when and why. A visible correction note is appropriate if a factual error may have affected readers’ understanding; routine copy edits generally do not need one. Do not quietly recast an earlier review as a present-day assessment if the writer has not returned to the recording.
After editing, read the whole page again. A corrected release date might now conflict with a timeline below it; a replacement image might leave the caption wrong. If you combine posts, check incoming references you control and preserve distinctive reporting rather than discarding the shorter piece by default. Test the result on the live site, not just in the editor preview.
Use a small, repeatable review cycle
Keep the work queue small enough to finish. Review a defined group of pages, assign each an action, and separate quick factual fixes from work that needs research. Track proposed new pages separately, with their reader question and evidence requirements. Record when revisions go live. If you use analytics, watch for signs that visitors still leave before reaching the information promised, but do not treat a traffic change alone as a measure of editorial quality.
Set the next review according to how quickly the material changes. Contact and ordering details need checking more often than a dated listening essay. Recheck affected pages when an embedded player is replaced, a release changes availability or a correction arrives; those events give you better reasons to act than a blanket call to refresh old posts.
For a first pass, take ten pages and finish this sentence for each: “This page is for someone trying to ______.” If filling the blank takes two unrelated needs, mark the page for a closer structural review. If the purpose is clear, test what the page actually says against it before proposing anything new.