Content Pruning for SEO: How I’m Rebuilding a Site After a 95% Traffic Drop and You Can Too

I own a website I’ve had for a while in the tech software space that totally lost 95% of its traffic in December 2025.

It wasn’t producing much as is and I kind of ignored it as we got busy with agency clients. I had only updated a few pages a few times in 2023-2025. I was actually shocked it made it so long in such a competitive space.

The traffic loss coincided with Google’s December 2025 core update. Timing alone does not prove that the update caused the decline, but it gave me a clear starting point for investigating what changed. After comparing the pages that fell with the pages that retained visibility, my working theories are:

• I had some old content from 2017-2018 that was poorly written by outsourced writers (Textbroker if you know what I mean) – this algorithm finally caught up
• Some of my title tags and dates made pages appear current even though the underlying content had not been meaningfully updated.
• Outdated information, including incorrect company and product names
• Not enough information gain
• Too much commodity content

There are some good takeaways and ideas from this that I’ll share later, but first, lets get into the content pruning process for SEO and why we may or may not want to do it.

What is Content Pruning for SEO Purposes?

Content pruning for SEO is the process of evaluating underperforming, outdated, overlapping, or unhelpful pages and deciding what should happen to them to best achieve your goals. Pruning does not always mean deleting content. A page may be kept, rewritten, consolidated with another page, redirected, removed, or excluded from search results. Many pages have a purpose outside SEO.

The goal is not to remove an arbitrary percentage of the website. It is to improve the overall usefulness of the site while protecting pages that still earn traffic, attract links, generate conversions, or support other important content.

What does Google say about pruning?

On this page, Google itself says:

“If you’re considering deleting entire sections of your site, that’s likely a sign those sections were created for search engines first, and not people. If that’s the case for your site, then deleting the unhelpful content can help the good content on your site perform better.“

Google brings up a good point. Lots of successful sites don’t necessarily do pruning batches unless they are overly SEO-focused and saw a drop. This supports the concept of doing it right the first time. Sometimes cowboys like to push the limits and test new approaches for geometric results, but this doesn’t always work in SEO. Or it may work for a short period and then stop.

Why is Content Pruning More Important in the Age of AI?

Content pruning is even more important in the age of AI because of the level of competition and content flooding the internet from AI writing tools. It’s easier than ever to create B to B+ content. So outsourced content from 2017 that is now C- content is even worse than AI, and that’s a huge shift.

Google now explicitly recommends creating what it calls non-commodity content for its AI search experiences. In plain English, that means publishing something more valuable than a summary of what every other ranking page already says. Firsthand experience, original research, testing, expert analysis, and a genuinely useful point of view are increasingly important differentiators.

In the world of SEO, the term “information gain” is used. But that’s a nerdy way to say: create something real, unique, and helpful that hasn’t been rehashed 100 times.

You Don’t Have to Always Delete Content for Pruning

It is important to understand that content pruning is not simply a nicer term for deleting blog posts. Removing a page is only one of several possible outcomes.

“Content pruning doesn’t always lead to deleting content. Some of your low-traffic pages could have SEO, conversion, or engagement potential.” Sydney Go, Semrush

That is the framework I am using for this project. Each page will be assigned one of four actions: keep it, improve it, consolidate it with a stronger page, or remove it. Traffic is part of that decision, but I will also consider backlinks, conversions, search impressions, topical relevance, accuracy, and whether the page still serves a useful purpose.

Anecdotal Experience: One Competitor Has Just 25 Pages

One of the most interesting competitors in this niche has only about 25 substantive pages. That does not mean 25 is a magic number, and it certainly does not mean that Google automatically rewards small websites. Page count by itself is not a useful measure of quality.

What it does show is that a relatively small, coherent collection of useful pages can compete with much larger websites. Most of this competitor’s pages appear to have a clear purpose, address a distinct search intent, and fit naturally within the subject of the website.

My site took a different path. It accumulated articles over several years, including pages that were no longer accurate, pages that overlapped with one another, and pages that were never especially good to begin with. The website became larger without necessarily becoming more valuable.

I have watched the SEO industry treat topical focus and authority like opposite ends of a pendulum. One year, everyone recommends publishing hundreds of tightly related articles. Later, sites with thin or repetitive topic clusters lose visibility and the advice shifts back toward links, brands, and authority.

In reality, focus and authority are not substitutes for quality. A tightly focused website can still be shallow. A powerful domain can still publish an irrelevant page. My objective is not to chase whichever theory is currently popular. It is to make the purpose and value of every remaining page easier for both users and search engines to understand.

How I Am Auditing the Site Before Pruning It

A content-pruning project should not begin with the delete button in WordPress. Before changing anything, I need an inventory of the site’s pages and enough information to make a defensible decision about each one.

One note I’ll say up front – I’m guilty of doing too much analysis before making moves. So you want to balance shipping changes that are impactful with doing deep and sitewide analysis.

1. Build a Complete Content Inventory

The first step is collecting every indexable URL into a spreadsheet. This can come from the XML sitemap, a website crawler, Google Search Console, analytics data, and the WordPress post list. Using multiple sources matters because an old or orphaned page may not appear in the current navigation or sitemap.

For each URL, I want to record:

• The page title and URL
• The original publication date and last meaningful update
• The page type and primary topic
• Its current indexation and canonical status
• Organic clicks and impressions
• Keywords and average ranking positions
• Organic sessions and conversions
• Internal links and external backlinks
• Automated or manual labeling and categorization – this may be mulitple columns
• The recommended action and reason for it
• Expected outcomes
• Final score
• Final priority

This turns a slow, editorial cleanup project into a documented content audit. Instead of saying that a page “feels bad,” I will be able to explain what purpose it serves, what evidence it has accumulated, and why I am keeping or changing it. Again, you should always align this with your team and company’s other goals around marketing and strategy.

2. Review More Than Traffic

When I was a young SEO, I disrespected any page that didn’t have organic traffic. Now, although I understood some pages were for paid campaigns or other purposes, I had a naive view that it was all about SEO, SEO, SEO. Over time I realized a website serves many, many purposes and channels, and many pages supported important partnerships, initiatives and events that brought in big clients and customers.

So of course, a page with no clicks is not automatically worthless. It may be new, seasonal, poorly optimized, hidden by a technical problem, or ranking just outside the positions that receive meaningful traffic. A page with very little traffic may also generate leads, attract backlinks, answer an important customer question, or provide internal-link support to a valuable commercial page.

I plan to review at least 12 months of Google Search Console and analytics data whenever possible. I will pay particular attention to the period before and after the December 2025 traffic loss, while remembering that timing alone does not establish causation. I need to also predict the 2026 and 2027 algorithms coming.

High impressions with few clicks may indicate a weak title, poor search-intent alignment, or increased competition from SERP features. A falling average position may indicate content decay. A page with no impressions at all may have a technical, indexation, demand, or quality problem. Those are different diagnoses and should not receive the same treatment.

3. Check Backlinks, Internal Links, and Business Value

Before removing a URL, I will check whether other websites link to it. An outdated page with strong referring domains may be far more valuable than its current traffic suggests. It may be better to update that URL or consolidate it into a closely related resource than to discard the authority it has earned.

Internal links matter too. Some pages act as useful bridges between major sections of a website even if they are not significant organic landing pages themselves. Removing those pages without updating the links can create broken journeys, orphan important content, and make the site harder to navigate.

Finally, I will consider business value. A page that generates a small number of highly qualified leads should not be treated the same as a page that attracts irrelevant visitors and produces nothing. It goes without saying that you should check in with stakeholders on your team that are in charge of different sections of the site.

4. Perform a Manual Quality Review

Metrics can identify suspicious pages, but they cannot determine whether an article is genuinely helpful. Every serious pruning candidate still needs to be reviewed by a person, ideally a subject matter expert.

For each page, I will ask:

• Is the information accurate and current?
• Does the page satisfy a real and relevant search intent?
• Does it contain firsthand experience or information that competing pages lack?
• Is another page on the site already doing the same job better?
• Would I be comfortable attaching my name and professional reputation to it?
• Would the website be less useful if this page disappeared?
• What is the purpose of this page, and is it fulfilling that?
• What are the goals for this page, and how is it tracking towards them?

That final few questions are particularly useful. If removing a page would make no meaningful difference to a visitor, a customer, or the rest of the website, it is probably a legitimate pruning candidate. Every page has a purpose and that should be labeled and tracked. Finally, we’re creating all of these with a goal in mind. How are you tracking to goals?

Efficiency hack: look at your site as a collection of sections and templates rather than individual pages. You will see strong patterns across sections, and after noticing these you can make batch decisions at scale so you don’t have to actually look at every single page.

Keep, Update, Consolidate, Remove, or Noindex?

After collecting the data and reviewing the content, I will assign every page one of five actions.

Keep

A page should be kept when it is accurate, useful, relevant to the website, and performing an identifiable job. That job may be earning traffic, generating conversions, attracting links, supporting another page, or answering a question that matters to the intended audience.

Not every page needs to be a traffic leader. The important point is that it has a defensible reason to exist.

Update

Updating is the best option when the underlying topic is still valuable but the execution has become outdated or inadequate. Pages receiving impressions without many clicks may be particularly promising because Google is already testing them for relevant searches.

An update should be meaningful. Changing the publication date, rewriting the introduction, or adding a few generic paragraphs is not enough. (That got me). A real update may require new product information, original screenshots, revised comparisons, clearer explanations, firsthand testing, stronger sourcing, and the removal of claims that are no longer accurate.

This is also where I can replace the old outsourced writing with material based on what I have actually learned from working in the industry.

Growth hack: updating can actually do wonders in some scenarios. I saw a site come completely back from the dead after a full website refresh. This could be the missing link for a lot of you.

Consolidate and Redirect

Consolidation makes sense when two or more pages serve substantially the same search intent. Rather than maintaining several mediocre articles that compete with one another, I can combine their strongest material into one more complete resource.

The weaker URLs should then receive a 301 redirect to the surviving page, provided that the destination is genuinely relevant. I also need to update internal links so they point directly to the final URL instead of passing through unnecessary redirects.

I will not redirect every deleted page to the homepage. An unrelated redirect does not help the user and may be treated like a soft 404. If there is no relevant destination, allowing the removed URL to return a proper 404 or 410 response may be cleaner.

Warning: I’ve seen in the SEO world a lot of attempts to do internal link redirects to help pages rank higher in search. It may work sometimes but I wouldn’t count on it. Instead combine a few smaller clusters and write/rollup really good content.

Remove

Removal is appropriate when a page is inaccurate, unhelpful, irrelevant, and has no realistic path to improvement. The strongest candidates will have no meaningful traffic, impressions, conversions, backlinks, internal purpose, or unique information worth preserving.

Before deleting anything, I will record the URL, the reason for removal, the date of the change, and any replacement or redirect destination. This change log will make it possible to reverse mistakes and evaluate the results later.

Great example: for the current site I’m working on, some of these companies are totally dead or acquired, so it’s better to just move on from these pages or redirect them to a “dead companies” page.

For companies that no longer exist, I will determine whether the page still has historical value. I will redirect it only when a closely relevant successor or consolidated resource exists. If there is no useful replacement, removing it with a proper 404 or 410 response may be the cleaner option.

Noindex

Some pages may be useful to existing visitors without belonging in search results. Examples can include thin tag archives, internal utility pages, account-related content, or near-duplicate filtered views. Other examples of course are pages for events, paid search, or other marketing campaigns that are outside of search and SEO areas.

A noindex directive may be appropriate for those pages. That decision is different from deleting the content because users can still access it directly. The page also needs to remain crawlable long enough for Google to see the directive.

Content I Will Not Delete Just Because It Has Low Traffic

The most dangerous version of content pruning is sorting URLs by traffic and deleting everything below an arbitrary threshold. Before removing a low-traffic page, I will look for reasons it may deserve protection.

• It has valuable backlinks or referring domains.
• It produces leads, sales, signups, or assisted conversions.
• It answers an important customer or support question.
• It is seasonal and is being reviewed during its quiet period.
• It was published too recently to have reliable performance data.
• It supports a larger topic cluster through useful internal links.
• It ranks for relevant long-tail searches that aggregate reports may hide.
• It contains original research, quotes, screenshots, or historical information worth preserving.

Some of these pages may still need improvement, but low traffic by itself is not a sufficient reason to destroy them.

The Biggest Risks of Content Pruning

Pruning is not without its risks.

Content pruning can go wrong when it is too aggressive or based on incomplete information. Deleting a page can eliminate rankings, links, referral traffic, and conversions that are difficult to rebuild.

Consolidation also creates risk when pages only appear similar. Two articles can share vocabulary while satisfying different intentions. Combining them may cause the surviving page to become less focused and rank for neither topic particularly well. This is why it’s important to bring in subject-matter experts.

For that reason, I like to release changes in documented batches. I can begin with the clearest cases, such as inaccurate pages with no traffic, links, conversions, or useful purpose for other channels. I will then monitor the results before making harder decisions.

This will not perfectly isolate the effect of pruning. Rankings can change because of competitors, seasonality, search-result features, technical changes, and Google updates. A staged rollout still gives me a better chance of spotting a mistake than deleting half the website in a weekend.

How I Will Measure Whether the Pruning Worked

Success should not be defined only by whether total traffic immediately increases. My baseline will include both sitewide performance and the performance of the pages or topic groups affected by each batch.

I plan to monitor:

• Total and non-branded organic clicks
• Organic impressions and the number of ranking queries
• Average positions for the site’s important topics
• Conversions and assisted conversions from organic visitors
• The number of valid indexed pages
• Traffic and rankings for consolidated destination pages
• Broken internal links, redirect chains, and orphaned pages

I will annotate the implementation dates and compare equivalent periods rather than reacting to a few volatile days. Early technical effects may appear quickly, but broader quality reassessment can take considerably longer. Google itself warns that improvements following a core update may take days or several months to be reflected.

Most importantly, I am not assuming that pruning alone will recover the lost 95%. The site may also need stronger individual pages, better links, clearer authorship, technical repairs, improved user experience, and more original evidence. Pruning is one part of the recovery plan.

What I Will Add Instead of More Commodity Content

The pages I keep cannot simply be longer versions of the same generic articles. They need information that is difficult to reproduce without genuine experience.

For a software-focused site, that can include original screenshots, hands-on testing, pricing and feature changes, documented methodology, comparisons based on real use cases, limitations that vendors do not advertise, and explanations of who should not buy a product.

That is the difference between expanding a page and improving it. Additional word count is not automatically additional value. The objective is to give the reader information that changes or strengthens a decision.

What I Expect From This Project

I do not expect to delete a group of posts and watch the site’s old traffic return the following morning. My first goal is more modest: create a smaller, clearer, and more maintainable collection of content that I am willing to stand behind.

If the process works, the remaining pages should be easier to update, easier to navigate, and less likely to compete with one another. Search engines should also receive clearer signals about what the website covers and which pages are its strongest resources.

I will continue documenting the number of pages kept, updated, consolidated, and removed, along with what happens to rankings and traffic afterward. That will turn this from a theoretical SEO article into a real content-pruning case study, whether the final result is a recovery, a partial improvement, or a lesson in what not to do.

Frequently Asked Questions About Content Pruning for SEO

Does deleting old content improve SEO?

Sometimes, but age alone is not a reason to delete a page. An older page may still earn traffic, links, or conversions. Deletion is most appropriate when the page is inaccurate, irrelevant, unhelpful, and has no valuable signals or realistic path to improvement.

How often should a website perform content pruning?

The appropriate frequency depends on how quickly the website publishes and how rapidly its subject matter changes. A large publisher may review content in regular quarterly batches, while a smaller website may only need a thorough review once or twice per year.

What is the difference between content decay and content pruning?

Content decay describes a page gradually losing accuracy, rankings, traffic, or usefulness. Content pruning is the process used to address that decline through updating, consolidation, removal, redirection, or another deliberate action.

What are the risks of content pruning?

The primary risks are deleting pages that still earn valuable traffic or links, redirecting URLs to irrelevant destinations, removing content that supports other pages, and changing too many pages at once to understand the results.

Final Takeaway

Content pruning is not about making a website smaller for the sake of being smaller. It is about making deliberate decisions regarding what deserves to remain available in search.

For my neglected software site, that means confronting years of content debt instead of publishing another layer of articles on top of it. Some pages will be rewritten, some will be combined, and some will disappear. The deciding factor will not be their age or word count. It will be whether they still provide a useful, distinct, and trustworthy reason to exist.