Discussion
  • Content SEO

Since the helpful content system was folded into core updates, do you treat quality as a sitewide signal or a page-level one? What evidence shaped that view?

Published 1 reply

Discussion starter: a content SEO question, posted with an answer from Niraj Raut to open the thread. If you have dealt with this on a site, reply with what you saw, especially where it differs.

Disclosure: Niraj Raut, who posted this answer, runs the SEO consultancy linked at the end of it.

1 reply

  • admin
    Admin
    Original poster

    I treat quality as both, but in a specific way: individual pages are ranked on their own merits, while a site-level assessment acts like a ceiling or a headwind that affects how far those pages can rise. That is roughly what Google’s own documentation says. After the helpful content system was folded into its core ranking systems in March 2024, Google described those systems as primarily working at the page level, while acknowledging that some site-wide signals are also considered. The practical consequence is that you diagnose with segmentation. When losses spread across most templates and topics, a site-level assessment is the more likely explanation. When they concentrate in a few sections, page and section quality are the more likely levers.

    What the evidence actually supports

    It helps to separate the sources, because they carry different weight.

    • Google’s documentation says core systems are mainly page-level with some site-wide signals. It does not say which signals, how much weight they carry, or how quickly they update.
    • The 2023 helpful content update period produced many cases of sites losing traffic across almost every page at once, including pages that were individually strong. That pattern is hard to explain with page-level assessment alone, and it is why practitioners came to think of the helpful content classifier as sitewide.
    • Recovery patterns after that period were slow and partial. Practitioner tracking of hit sites, such as Glenn Gabe’s, showed some sites rebounding together at particular update dates, which again points to a site-level component being re-evaluated periodically rather than page by page.
    • Google’s August 2026 update to the site reputation abuse policy says that in the EEA an affected section may be separated in Google’s systems so it ranks independently of the rest of the site. That is a strong signal that Google can and does reason about sections of a site as units distinct from the whole.

    Together this supports a model with page-level ranking plus site-level and section-level signals. It does not support the idea that a single low-quality page drags down a whole site, or that a sitewide score overrides everything.

    A mental model that works in practice

    Think of a restaurant guide’s reviewers. Each dish is judged on its own, but the restaurant’s reputation shapes how much benefit of the doubt the next dish gets, and a reviewer may treat the bar downstairs as a separate place if it clearly is one. The analogy is loose, but it captures the diagnostic point: most of your gains come from the dishes, and a poor reputation makes every dish work harder to be noticed.

    How to tell which one you are dealing with

    Pattern after a core update More likely explanation Where to focus
    Most directories and templates lose a similar share of clicks, including strong pages Site-level assessment What the site as a whole is for, the ratio of useful to thin content, reputation and trust signals
    Losses concentrated in one section, such as a coupon, affiliate or programmatic area Section-level assessment That section’s purpose, quality and relationship to the rest of the site
    Losses on specific queries where competitors with different page types gained Intent or source-type shift Whether your page type still fits the query; this was a major theme in analyses of the May 2026 core update
    Scattered losses on individual pages, gains on others Page-level reassessment The individual pages: freshness, depth, originality, match to intent

    To see the pattern, export Search Console data for the weeks before and after the update, group pages by directory and template, and compare the change in clicks and impressions per group. Compare page-level distributions too: if the median page in almost every group lost a similar share, that is the sitewide signature.

    What changes in your actions

    If the evidence points to a site-level problem:

    • Review what share of the site’s indexed pages you would be proud to show a sceptical expert. Large volumes of thin, derivative or search-first content are the usual cause of a sitewide headwind.
    • Improve, consolidate or remove the weak content. Google’s helpful content guidance in the 2022 to 2023 period explicitly said that removing unhelpful content could help the rankings of other content. That guidance predates the March 2024 change, so treat it as indicative rather than current documentation.
    • Expect recovery, if it comes, to be tied to later core updates rather than to arrive as soon as you finish the work.

    If the evidence points to page or section problems:

    • Work on the affected pages directly, and compare them with the pages that now rank. Often the difference is page type or intent fit rather than quality in the abstract.
    • If a whole section is the problem, consider whether it belongs on the site at all, whether it needs a different editorial standard, or whether it should be separated.

    Watch for this:

    Mass deletion is the most common overreaction to a sitewide drop. Removing pages that earn traffic, links or conversions because they are “old” or “short” can cut real value without touching the underlying problem. Judge each page on usefulness and demand, not on age or word count, and keep anything that serves users well.

    Hypothetical example:

    A home improvement publisher loses about 30% of organic clicks in a core update. Segmenting by directory shows every section down by a similar amount, including its best-performing, expert-written guides. That points to a site-level assessment. An audit finds that about half the indexed pages are short, AI-assisted articles published in bulk two years earlier. The publisher consolidates or removes most of them, improves the rest, and waits. Its strong guides regain some ground two updates later. Had the losses been confined to the bulk articles, the right move would have been to fix or remove those pages without assuming the whole site had a problem.

    Rank pages, protect the site. Improve individual pages to win queries, and keep the overall mix of content good enough that the site-level signal is a tailwind rather than a ceiling. Let segmentation, not intuition, tell you which problem you have.

    Need help with this on your own site?

    Niraj Raut works with businesses in Nepal, Australia, the UK, Europe and the US on technical SEO, ecommerce SEO, local SEO and AI search.

    See SEO services

Add a reply

Replies are open to approved members. Log in or apply to join.