Trending SEO FAQs

Backlink count alone tells you very little. What matters is the relevance of the linking domain, the topical authority it holds in your niche, and where on the page the link sits. A link from a low-authority blog buried in a footer carries almost no weight compared to a single contextual link from a site Google already trusts for that topic. I’ve seen pages with 40 referring domains get outranked by a page with 12, simply because those 12 came from sites Google considers topically adjacent. Anchor text diversity plays a role too. If every link uses the exact same commercial anchor, that pattern reads as manipulative rather than organic, and Google’s link spam systems have gotten much better at spotting it since the 2024 core updates.

Start in Search Console, not in a keyword tool. Pull the query report for both URLs over a 90-day window and look for overlapping impressions on the same search terms, even if the pages were written for different intents. Cannibalization often hides in long-tail variations rather than the head term. A page built for “how to calculate EMI” and another for “EMI formula explained” look distinct on the surface but frequently split rankings for dozens of shared long-tail queries neither page owner tracked. Once you spot the overlap, check which page Google is alternating between in the SERP over time. If it keeps swapping which URL ranks, that’s your confirmation. The fix is usually consolidation into one stronger page with a 301 from the weaker one, or a clearer intent split with internal links pointing users to the right page for each stage of the funnel.

For most sites that size, no, not in the way people worry about it. Crawl budget becomes a real constraint when you’re dealing with faceted navigation, infinite scroll pagination, or auto-generated pages in the hundreds of thousands. Below 10,000 URLs, Googlebot will typically get through your whole site within days regardless of your server response time. Where I do see smaller sites lose crawl efficiency is through parameter bloat, things like session IDs, sort filters, or tracking parameters generating thousands of near-duplicate URLs that never should have been indexable in the first place. That’s not a budget problem, it’s a technical hygiene problem, and it’s solved with canonical tags and robots.txt disallow rules rather than server upgrades.

The biggest practical shift is that you’re no longer just optimizing for a ranking position, you’re optimizing to be the source an LLM pulls from when it synthesizes an answer. That changes how you structure content. A traditional SEO page might bury the direct answer three paragraphs in after some scene-setting intro. An AEO-optimized page puts the answer in the first two sentences under the heading, because that’s the chunk most likely to get extracted cleanly. Schema markup matters more too, specifically FAQ and HowTo schema, because it gives language models a structured signal about what question your content answers. I’ve also noticed citation-worthy content tends to include specific numbers, named studies, or original data rather than generic statements, because LLMs seem to favor sources that offer something concrete to attribute.

This trips people up constantly. If you rewrite a page and change too much at once, on-page structure, headings, URL, internal link anchor text, Google sometimes treats it as close enough to a new page that it has to re-earn trust signals, even though nothing technically reset. I’ve watched refreshes tank rankings for two to four weeks before recovering stronger than before, purely because the crawler needed to re-evaluate relevance signals from scratch. The safer approach for pages already ranking on page one is incremental updates: refresh data points and add missing subtopics without touching the URL, core heading structure, or existing internal link equity. Save full rewrites for pages stuck on page two or three where you have less to protect.

Domain authority is a third-party metric, mostly from Moz or Ahrefs, and Google doesn’t use it directly at all. Topical authority is Google’s own internal assessment of whether your site is a credible source on a specific subject, built from a combination of your content depth on that topic, the entities you consistently mention and link between, and how other authoritative sites in that space reference you. A site can have a modest domain authority score and still outrank much bigger domains within a narrow topic if it has published deeply and consistently on that subject. This is why niche sites often beat massive generalist publishers for specific long-tail queries. If you’re resource-constrained, building out topical depth in one content cluster beats chasing generic backlinks every time.

The first thing I check is whether Googlebot’s rendered HTML actually matches what a user sees, using the URL Inspection tool’s rendered screenshot alongside a view of the rendered DOM. A lot of JS-heavy sites assume Google renders everything perfectly, but there’s often a delay between crawling and rendering, sometimes days, which means fresh content can sit unindexed longer than expected. For anything commercially important, server-side rendering or at minimum dynamic rendering for bots removes that uncertainty entirely. I’ve also seen cases where critical content loads behind a user interaction, like a tab click or an accordion expand, and Googlebot never triggers that interaction, so the content effectively doesn’t exist for indexing purposes even though it’s visible to real visitors.

Auto-generated related links are almost always based on category or recency, not actual relevance or link equity distribution, so yes, manual work still matters. The highest-impact internal links are the ones from your highest-authority pages pointing down to pages you’re trying to push up the rankings. Most sites do the opposite by default, their homepage and top posts link out to other top posts, while newer or underperforming pages get scraps. I generally map internal links by first identifying which pages hold the most equity using something like Ahrefs’ internal link report or a simple crawl with Screaming Frog, then deliberately route links from those pages toward priority targets, choosing anchor text that reflects the target page’s actual topic rather than generic phrases like “read more.”

Because Google’s language models have moved well past exact match matching toward semantic and entity-based understanding. A page can rank for “best budget laptops for students” without ever using that literal phrase, as long as it comprehensively covers the entities and subtopics a person searching that term would expect: price ranges, battery life comparisons, specific model names, use-case scenarios. Keyword stuffing exact phrases now reads as a weaker signal than genuine topical comprehensiveness. That said, this doesn’t mean keywords are irrelevant, it means the target has shifted from phrase matching to concept coverage, and your content briefs should be built around subtopics and related entities rather than keyword density targets.

Content relevance and quality will always outweigh Core Web Vitals as a ranking factor, but Vitals functions more like a tiebreaker and a user retention lever than a direct ranking booster. Where I’ve seen it matter most is on competitive queries where multiple pages are already well-optimized for content, in that scenario a genuinely poor Largest Contentful Paint or high Cumulative Layout Shift can be the difference that keeps a page stuck below competitors with similar content depth. It also affects rankings indirectly through behavior signals, since slow pages see higher bounce rates, and Google does appear to weigh engagement patterns over time. I wouldn’t chase a perfect Lighthouse score before fixing thin content, but I also wouldn’t ignore a site loading in six seconds while debating whether an H2 needs one more sentence.

Proactive disavowing is largely unnecessary for most sites now. Google’s algorithms are generally good at ignoring low-quality links rather than penalizing sites for having them, which is part of why negative SEO through spammy link building rarely works anymore. The exception is if you’ve received a manual action notice in Search Console specifically citing unnatural links, or if you inherited a domain with a clear history of purchased link schemes. In those cases, disavowing the worst offenders alongside a reconsideration request makes sense. Outside of that, spending hours auditing and disavowing links each month is usually time better spent on content or technical fixes, since the downside risk of ignoring toxic links is much smaller than most SEOs assume.

This is genuinely one of the harder measurement problems right now since there’s no native reporting dashboard. The workaround most practitioners use is manual or semi-automated prompt testing, running a consistent set of queries related to your target topics through ChatGPT, Perplexity, and Google’s AI Overviews on a recurring schedule, then logging whether and how your brand or content gets cited. It’s manual, it doesn’t scale well past a few dozen prompts without tooling, and results can shift between sessions even for the same prompt due to model non-determinism. Referral traffic segmentation in GA4 filtered by AI platform referrers gives a partial signal too, though a lot of AI-driven traffic arrives with no referrer at all, so it undercounts. Until platforms build proper visibility reporting, triangulating between prompt testing and referral data is the most reliable approach available.