Refresh, Merge or Retire, the Annual Content Pruning Decision for Law Firms

Every law firm blog that has been running for more than three years carries dead weight. Not bad work, usually, just work that has aged out. A 2019 post about a statute that changed in 2023. Four separate 400 word answers to nearly the same client question, written by four different people, quietly competing with each other. A practice area page for a service the firm stopped offering. None of it is doing harm you can feel, which is exactly why nobody deals with it.

The year end content pass is where you deal with it. Done properly it takes a marketing lead one focused week, and the output is a list where every URL on the site has been assigned to one of four buckets. Leave it alone, refresh it in place, merge it into something stronger, or retire it. The decision should come from three numbers you can pull in an afternoon, not from a partner scrolling the blog and saying that one feels old.

Pull the Three Numbers Before You Read a Single Post

Reading posts first is the classic mistake. You start at the top of the archive, get absorbed in the writing quality of the fourth article, and four hours later you have opinions about eleven URLs out of six hundred. Data first, reading second, and you only read the pages the data flags.

Open Google Search Console, go to the Performance report, set the date range to the last 12 months, and switch to the Pages tab. Export it. This gives you clicks, impressions, average CTR and average position per URL. Twelve months matters here because personal injury and family law both have seasonal swings, and a 90 day window will make a perfectly healthy page look dead.

Second, get an internal link count per URL. Screaming Frog will crawl a typical firm site of 400 to 900 pages in under ten minutes on a desktop license, and the Inlinks column in the export is what you want. Free alternatives exist, and Ahrefs or Semrush site audits give you the same column if you already pay for one. What you are looking for is not the absolute number but the shape of the distribution. On most firm sites, roughly 60 to 70 percent of blog posts have only navigational inlinks, meaning the archive and a category page, and nothing else. Those are orphaned in every sense that matters.

Third, get last meaningful update date. Not the WordPress modified timestamp, which a plugin update can bump without a word changing, but the date the substance was last touched. If nobody tracked it, use published date and accept the noise.

Join those three into one sheet. URL, clicks, impressions, average position, inlinks, age. That sheet is the whole exercise. If you have already built a content inventory, this is a much shorter job, and the mapping work in content inventory and topic mapping for law firms is the layer that tells you which posts were supposed to be about the same thing in the first place.

The Decision Tree

Run every URL through these gates in order. Stop at the first one that fires.

Gate One, Is It Earning

If a page generated meaningful clicks in the last 12 months, leave it alone. For most firm blogs meaningful means 50 or more clicks a year, though a page pulling 12 clicks on a term like wrongful death attorney in a mid size market may be worth more than a page pulling 900 clicks on a general legal curiosity question. Judge by intent, not volume. Pages that clear this gate get no changes beyond factual accuracy fixes. The most common self inflicted wound in a content pass is rewriting a page that was already working.

Gate Two, High Impressions and Weak Position

This is the refresh bucket, and it is where the return actually lives. Filter for pages with 1,000 or more impressions in 12 months and an average position between roughly 6 and 20. Google is showing the page to real searchers and they are not clicking, or the page is sitting just below the fold of page one. That is not a content quality verdict, it is a proximity signal. These pages are close.

Refresh means specific things. Pull the query list for that individual URL in Search Console and look at what people actually typed, because on older posts it is common to find the page ranking for a question it never directly answers. Add the section that answers it. Update statutes, filing deadlines, damage caps and court procedures to current law, and say in the copy what changed. Rewrite the title tag against the real query set rather than the one you guessed at three years ago. Add two or three internal links from stronger pages on the site pointing in.

Expect five to fifteen refreshes to produce clear movement out of every twenty you do, with most of it landing six to twelve weeks after publication. Some will not move at all, and a few will slip. Anyone who tells you a refresh reliably lifts a page is selling something. What is defensible is that refreshing twenty near miss pages beats writing twenty new ones from zero, because those pages already have history, links and demonstrated impressions.

Gate Three, Thin, Overlapping and Cannibalizing

This is the merge bucket. Sort your sheet by target topic and look for clusters of three or more URLs chasing the same intent, each with impressions in the low hundreds and none of them ranking well. The classic law firm version is five posts that all essentially answer how long do I have to file a claim, written across four years, each 500 words, each slightly different, all cannibalizing each other.

Merging means picking one canonical URL and making it genuinely comprehensive. Pick the URL with the most inlinks and the strongest history, not the newest one and not the one with the prettiest slug. Fold the unique material from the others into it, which usually means you are keeping about 40 percent of each donor and discarding overlap. Then 301 redirect every donor URL to the survivor, and go update the internal links that pointed at the donors so they point at the survivor directly. Redirects should be a safety net for external links and old bookmarks, not the primary path your own site uses. Moz’s explanation of how redirection works and when each type applies is the clearest short reference to hand a developer who is going to implement the rules.

One caution. Merging genuinely distinct practice area pages because they look similar to you is how firms lose rankings they had. Two pages targeting car accident lawyer and truck accident lawyer serve different searchers with different case values. Similar vocabulary is not the same as duplicate intent.

Gate Four, Zero Impressions and Zero Links

Now you are in retire territory, and this is where discipline matters most. A page qualifies for consideration only if it has near zero impressions over 12 months, near zero organic clicks, no inlinks beyond navigation, no backlinks, and no strategic role. If a page has even a handful of referring domains, it is not a retire candidate no matter how weak the content is, because that equity is worth folding into something else.

The default action here is still a 301 redirect to the most topically relevant surviving page. Not the homepage. A redirect to the homepage from a specific topical page is treated as a soft 404 more often than not, and you gain nothing.

The Narrow Cases Where Removal Actually Beats Redirection

Removal, meaning a 410 Gone or a genuine 404, is right in a small number of situations. Being honest about how small that number is separates a content pass from vandalism.

• There is no relevant destination. An old post about a bar association fundraiser has no topical successor. Redirecting it to a personal injury page is an irrelevant redirect and Google will treat it as a soft 404 anyway. Serve a 410 and stop paying attention to it.
• The content creates legal or ethical exposure. A post that quotes an outdated damages cap, describes a service the firm no longer provides, or reads as a guarantee of outcome under your state bar advertising rules should be gone, not preserved behind a redirect. Redirected pages still get retrieved through archives and screenshots.
• The page was never yours in spirit. Syndicated filler, an old vendor’s boilerplate, scraped or spun content from a previous agency. Redirecting it just passes a signal you do not want.
• Crawl budget on a genuinely large site. This applies to a firm with tens of thousands of URLs, usually from a directory or an unmanaged tag and archive explosion. It almost never applies to a 500 page firm site, and it gets cited far too often as a reason to delete things. If your crawl stats in Search Console show Google fetching your important pages regularly, crawl budget is not your problem.

Before you decide anything is dead weight, check that Google ever saw it as a real page. Plenty of pages with zero impressions were never indexed, and the reason is often technical rather than editorial. Work through when Google will not index your law firm pages before you retire a batch, because a page excluded by a stray noindex tag or a canonical pointing somewhere else is a fixable page, not a failed one.

How to Execute Without Breaking the Site

Work in batches. Twenty to forty URLs at a time, not four hundred in a weekend. Batching means that when something moves in the wrong direction you can tell which change caused it.

Before touching anything, save a full crawl export and a Search Console export as your baseline. Screenshot or export the current rankings for the terms that matter to the firm commercially. If a merge goes wrong two months from now, you will want to know exactly what the before looked like.

Implement redirects at the server or through a maintained redirect manager, and audit the file afterward for chains. A redirect that hops through two intermediate URLs before landing is a slow signal loss, and after three or four annual passes most firm sites have accumulated them. Flatten every chain so each old URL points at its final destination in one hop.

Resubmit the updated sitemap and let it settle. Merged and retired URLs will linger in Search Console coverage reports for weeks, and that is normal processing, not a failure. Measure at the cluster level rather than the URL level, because the whole point of a merge is that traffic consolidated, and looking at the survivor alone will overstate or understate what happened.

Do this once a year, ideally in the quiet stretch between the holidays and the new year push. Keep it annual and it stays a one week job. Skip it for three years and it becomes a project. The other half of keeping it small is publishing at a rate you can actually sustain and maintain, which is the argument in blog cadence for law firms, because most pruning backlogs are really publishing volume decisions catching up with the firm.

What Good Looks Like When You Are Done

A finished pass on a typical firm blog of 250 posts usually lands somewhere near this shape. Around 60 percent untouched because they were earning or harmless. Twenty five to forty pages refreshed in place. Six to ten merge clusters collapsing perhaps thirty URLs into ten. And a genuinely small removal list, often under fifteen pages, of which only three or four get a 410 rather than a redirect. If your removal list is 40 percent of the site, you have built a deletion project and called it a content pass, and you will spend the following quarter explaining a traffic decline.

The deliverable is the sheet, with a decision, a reason and a destination URL in every row, plus the redirect map. That artifact is what makes the pass auditable next year, and it is what stops the next marketing lead from re examining the same six hundred URLs from scratch.

Rubiks runs this pass for law firm clients as a scheduled annual engagement, decision sheet and redirect map included, and we will tell you honestly when a site is too young or too small to need one. Talk to Rubiks and we will show you which of your pages are near misses worth refreshing before you spend another dollar on new posts.