Content Pruning Decision Matrix Template

Content Pruning Decision Matrix (fill-in)

When to use: One row per flagged URL with traffic, links, conversions, strategic fit, decision, owner and sign-off.

Content pruning decision matrix

Use this only for pages already flagged as weak in your audit. One row per URL. Fill the evidence first, then pick one decision, then get sign-off.

Per-row detail (fill for every URL)

• Traffic: clicks [n], impressions [n], average position [n] in Search Console for [date range]
• Top queries: [query - clicks], and do they belong to another URL? [yes, [URL] / no]
• Backlinks: referring domains [n], notable links [domain - linked URL]
• Internal links: [n] pages link here, including [nav / footer / product pages / emails]
• Conversions: [leads, sales, sign-ups or assisted conversions] in [same window]
• Strategic fit: [core topic / supporting / off-topic], owner says [reason]
• User or legal need: [policy, support doc, QR code, partner link, required notice / none]
• Overlap: same intent as [URL or none], from the Content Cannibalization Register Template
• Decision: [improve / merge + 301 / keep as is / noindex / remove], because [evidence that met your rule]
• Technical action: status code [200 / 301 to [target] / 404 / 410], robots meta [index / noindex], sitemap [keep / remove]
• Approval: SEO [name], page owner [name], legal or product [name or not needed]

Next action

• SEO lead: [name] - action: fill the evidence and propose a decision by [date]
• Page owner: [name] - action: confirm strategic fit and user or legal need by [date]
• Developer: [name] - action: ship the status code, redirect or noindex, and update the sitemap by [date]
• Analyst: [name] - action: re-check traffic, conversions and redirect targets on [change date + [n] weeks]

Completion check

• Every row has traffic, backlinks, internal links and conversions from the same date range.
• Every decision names the rule it met, using your own thresholds.
• Pages needed for users or legal reasons are marked keep or noindex, never remove.
• Each merge has one relevant target URL, not the home page.
• Internal links and the XML sitemap are updated for every URL that changed status.
• The sign-off column is filled before anything ships.

Decision Rules and Evidence Thresholds

When to use: Write your own threshold for improve, merge + 301, keep, noindex and remove before you judge any URL.

Decision rules and your evidence thresholds

Write the thresholds once, before you look at single URLs. The numbers are yours to set. They depend on your site size, niche and sales cycle.

What Google says about these options

• Deleting is not a default fix. Google's core updates guidance calls deleting content a last resort, for content that can't be salvaged.
• Google treats all 4xx codes except 429 the same. Indexed URLs that return a 4xx are removed from the index.
• A 301 is a signal that the target should become canonical. Google shows the target in results.
• Noindex only works if Google can crawl the page. Do not block it in robots.txt.
• Google advises against sending many old URLs to one irrelevant page. It can be treated as a soft 404.

Not sure the page is weak at all? Run it through the Content Optimizer first. A page that only needs an update belongs in the improve row.

Still choosing which pages to flag? Start with the Content Audit and Refresh Template, then bring the weak ones here.

Risk Check Before Removal or Merge

When to use: Check backlinks, internal links, conversions, sitemap, redirects and legal needs before a row is approved.

Risk check before you merge, noindex or remove

Run these checks for every URL that is not marked improve or keep. One unchecked box means the row is not ready for sign-off.

Backlinks

• Referring domains to this URL: [n]. Run your domain in the Backlink Analyzer and check its Links to column.
• Links you would lose with a 404 or 410: [domain - anchor]
• If there are links worth keeping, the decision is merge + 301 to [closest relevant URL], not remove.

Internal links

• Pages that link here: [n], from your CMS or crawl export. After release, the Website Audit flags broken links and links to redirects.
• Navigation, footer, related posts or product pages that link here: [list]
• Plan: update each link to [new target] or remove it, so no internal link points to a redirect or a 404.

Conversions and revenue

• Conversions or assisted conversions in [window]: [n]
• Used in ads, emails, QR codes, sales decks or partner pages: [yes, where / no]
• If yes, the owner of that channel signs off: [name]

Sitemap, canonical and redirects

• In the XML sitemap now: [yes / no]. Remove it if the URL will redirect, 404, 410 or carry noindex.
• Other URLs that canonicalize to this one: [list or none]
• Existing redirects that point to this URL: [list]. Repoint them to the final target to avoid chains.
• Target URL returns 200 and is indexable: [checked on [date]]

User, legal and policy

• Required by law, contract or policy: [yes, which / no]
• Linked from support docs, apps or printed material: [yes, where / no]
• Third-party content hosted mainly to rank on your site's strength: [yes / no]. Google's spam policies cover this, so review it with care.

Implementation Picker (Status Codes and Noindex)

When to use: Turn each decision into the right status code, robots meta, sitemap and internal link change.

Implementation picker: status codes, noindex and sitemap

Each decision needs the same technical steps every time. Copy the row into the ticket for your developer.

404 or 410?

• Google says all 4xx errors except 429 are treated the same, and it stops using those URLs over time.
• Pick one for your site and write it here: [404 / 410], because [reason, such as CMS default or team rule].
• Show a helpful error page with links to your main sections. The status code must still be 404 or 410.

Redirect rules for merges

1. Redirect to the page that answers the same need, not to the home page.
2. Redirect straight to the final URL. No chains.
3. Use a permanent redirect only when you are sure it will not be reverted.
4. Keep redirects in place for a long time. In its site move guide, Google says generally at least 1 year.
5. Test each old URL after release in the Redirect Checker: [old URL - final URL - status]

Noindex rules

• Use the robots meta tag on HTML pages, or the X-Robots-Tag header for PDFs and other files.
• Do not block the URL in robots.txt, or Google never sees the noindex.
• Google must recrawl the page to see it. That can take a while for pages Google rarely visits.
• Do not use noindex to pick a canonical between duplicates. Merge or use rel canonical instead.

Moving many URLs at once? Map them in the 301 Redirect Mapping Template and track leftovers in the Broken Link and 404 Management Template.

Approval, Release and Rollback Log

When to use: Ship pruning in signed-off batches with a check date and a rollback trigger.

Approval, release and rollback log

Ship pruning in small batches. Each batch has a sign-off, a check date and a rollback trigger you set before release.

Sign-off questions

• Did each row meet the threshold written in the decision rules, not a new one?
• Did the page owner confirm there is no user, legal or sales need?
• Is the redirect target or improved page live and indexable?
• Is there a copy of every removed page, so you can restore it?

After release

• Redirected URLs: compare clicks to the target URL before and after, for [queries the old page got].
• Removed URLs: check the Page indexing report for [Not found (404)] on the expected URLs only.
• Run an on-demand top 10 check for the target pages in the Rank Tracker: [keyword - position before - after]
• Conversions on target pages: [before / after]
• Rollback: if [your trigger] happens, restore [URL] from the copy and remove the redirect.

Want someone to review your pruning batch before it ships? Book a free consultation with Anatolii Ulitovskyi.

Spreadsheet Header (CSV)

When to use: A paste-ready header row and allowed values to run the matrix in Google Sheets or Excel.

Paste-ready spreadsheet header

Paste this row into cell A1 of Google Sheets or Excel, then split text to columns by comma. One row per flagged URL.

row_id,url,flag_reason,date_range,clicks,impressions,avg_position,top_queries,referring_domains,notable_links,internal_links_in,conversions,assisted_conversions,strategic_fit,user_or_legal_need,overlap_url,decision,rule_met,target_url,status_code,robots_meta,sitemap_action,internal_links_action,risk_check_done,seo_owner,page_owner,approved_by,approval_date,batch,release_date,check_date,rollback_trigger,outcome,status

Allowed values

• strategic_fit: core, supporting, off_topic
• decision: improve, merge_301, keep, noindex, remove_404, remove_410
• robots_meta: index, noindex
• sitemap_action: keep, remove, add_target
• risk_check_done: yes, no
• status: proposed, approved, shipped, checked, rolled_back

Sheet tips

• Add a rule that flags rows where decision is remove and referring_domains is above your threshold.
• Flag rows where conversions is above zero and decision is remove or noindex.
• Block status approved while approved_by or risk_check_done is empty.
• Filter by target_url to see every page merged into one URL.

Worked Example (fictional houseplant shop)

When to use: Five filled rows for an invented shop, one for each decision, with risk notes and a check after release.

Worked example: Larkspindle plant care pages

Example only. Larkspindle is an invented houseplant shop. Its URLs, thresholds, dates and numbers are made up to show a filled-in matrix.

Larkspindle's own rules: remove only with 0 conversions and fewer than 2 referring domains. Improve when impressions are above 3,000. Merge when an overlap is confirmed.

Risk check notes

• Snake plant merge: two useful sections moved into the guide. 7 internal links repointed. Old URL removed from the sitemap.
• Winter sale: no links, no emails still pointing to it. Footer link removed. Returns 410 with links to current offers.
• Pet-safe plants: linked from 40 product pages and assists sales. Kept, so the next audit skips it.
• Care card: printed QR codes point to it, so it stays live. Noindex added, not blocked in robots.txt.

Check after 6 weeks (late October 2026)

• The repotting guide now ranks for the old page's main query, and the old URL shows as a redirect.
• The winter sale URL is out of the index. No other 404 or 410 URLs appeared.
• The fiddle leaf page moved from position 14 to 9 after the rewrite.

Why this works

• Only one of five weak pages was removed. The rest kept their links, users or sales.
• Thresholds were set before the review, so no one argued case by case.
• Operations signed the care card row, because the team that prints the QR codes needed a say.