Why is communicating value half the battle in SEO?
Because showing clients and managers what you are doing and how it moves the needle is critical, and most SEOs confuse providing data with providing a report.
A report is a story, and the higher up you go, the more it needs to be one so it can be understood and relayed to marketing teams, investors, or shareholders. SER trackers also make position reports less reliable because rankings are more fluid and differ per user, so SEOs should connect their work to real outcomes like leads and visibility, identify leading or lagging indicators, and plan measurement tools at the start of the engagement.
As traffic becomes harder to attribute, especially for established businesses, teams need new metrics and must identify keywords that trigger AI overviews and keywords that avoid them.
What is the position 17 rule and how should it be used?
It is a method for finding keywords in Google Search Console where you rank just beyond position 17, indicating a realistic opportunity to reach page one by writing focused content for those phrases. The number itself is not fixed; it could be 13 or 21.
Keywords ranking at position 78 or 90 are usually too competitive for the site to enter, so targeting the middle ground is more efficient. The technique works like a reverse topical authority hack: if a page ranks for one thing but also sits at position 17 for another, focusing on that second topic will likely improve ranking because the algorithm itself cannot be used directly, but Search Console data can reverse engineer it.
How does Google actually relate content, and why did some sites lose rankings?
Google does not conceptually group topics like schools, nonprofits, and communities the way a human editor would; it relates pages largely because they share the same words in their search phrases. Content that is conceptually related to a human may not be related to a search engine.
An article about WordPress for schools and one about SEO for WordPress for schools are related because both phrases contain WordPress and school, not because of deeper semantic understanding. When a site drifts across such conceptually grouped topics, it loses clicks, authority, and the semantic leeway Google grants it; returning to cornerstone content and choosing keywords that share the same words is an easy fix.
What is the difference between discovered not indexed and crawled not indexed?
Discovered not indexed means the page was never actually retrieved from the internet, so Google has not read it at all, and changing the slug and republishing often resolves it. Crawled not indexed is almost always an authority problem: Google fetched the page but does not believe the site has the topical authority to index it.
Googlebot only fetches files, opens them, follows links, and adds pages to indexing queues; it does not evaluate quality. Claims that discovered not indexed means Google read and disliked the page are wrong, and the claim that Google refuses to index duplicate content contradicts the Google SEO starter guide, which is why advice from LLMs quoting support forum posts can be unreliable.
For crawled not indexed, document names signal relevance: a plugin review titled only by the plugin name may be ignored unless placed in context, such as naming it as a WordPress plugin, or the page can be linked from a page with authority so PageRank flows to it.
Is scaled content still viable under Google's policies?
Programmatic SEO has always existed and remains acceptable as a fractional increase relative to an existing site, but building thousands of pages from scratch to enter a new market now draws a target that will not work.
Adding an SEO jobs category that generates 500 pages on an existing 10,000 page site is a fractional increase and fine; launching a 5,000 page site from scratch for that market is not. The backlash exists mainly because people avoid calling the practice scaled content, which is what it actually is.
How many internal links should a page have?
There is no fixed rule or Google penalty; as a rule of thumb, link a page to two or three other pages, though a page with a lot of authority can carry 20 links. More internal links reduce the likelihood of authority transferring between pages.
The doctrine that more internal links help the spider understand the site is simply wrong, and hidden, footer, or discounted links transfer little authority; a single link on a page with traffic beats a footer link. Beyond raw counts, link flow is a click-probability question: if Google does not believe a link will get clicks, authority will not flow through it, and PageRank is recalculated continuously so links can be added and removed freely.
For a new site built from a single origin page, a better strategy is to get one page ranking at a time and expand from there, rather than linking one authoritative page to 20 others at once; if more links are genuinely needed, importing authority through a go-to-market link partnership is an option.
What should a page stuck on page two be diagnosed as?
A page that stays on page two or briefly touches page one is a topical authority problem: the keyword was out of bounds for the site.
Adding internal links or backlinks only imports authority from elsewhere to cover the gap. The two real options are finding another authority source to pipe in via links, or rolling the page back to a longer-tail or lower-difficulty variant. Content that never earns clicks should be addressed before new content is created, and writers should check whether previous posts actually ranked.
Will Google purge AI-generated content?
Google's guidelines do not ban AI content, and a Google engineer liking a post about AI content does not signal a coming purge; the actual rule is that anything done to manipulate search is against the terms of service, whether written by AI or not.
Sites hit by the Helpful Content Update had not violated the specific penalties like keyword stuffing or link spam but were judged to be manipulating search for AdSense revenue, and mass-producing AI blog posts purely to manipulate search falls under the same principle. A detection-based purge would start a chase between detection tools and decentralized LLM builds, and unlike images or video, text has little room for artifacts, so enforcement would create a new problem for Google.
Quality itself is subjective: an excellent technical thesis that omits the Oxford comma is not bad content, just subject to someone's stylistic preference, so using AI to articulate ideas, with visible markers, is not inherently a problem.
How should marketers respond to AI overviews rerouting traffic?
AI overviews redirecting traffic is a reality of the world and cannot be changed, so the focus should shift from rank position to depth and breadth of coverage, aiming to appear in the relevant AI overviews more than once rather than just once.
Even 48 new number-one positions may not translate into traffic when AI overviews absorb the query, which means value communication has to change too; sharing that win with a CEO would set up a disappointment the following week. The difference between SEO and GEO may be that GEO requires owning first, second, and third place in AI answers rather than just first place in classic results.
Visitors arriving through large language models tend to be more educated and further along in their research by the time they reach a site, a pattern consistent with Reddit and G2 research finding that B2B buyers already know what they want by the time they ask to talk to sales.
Will Google expand AI overviews into linkless AI mode results?
The expectation is that Google will not pursue fully expanded, link-free AI overviews; the observed cases were likely a test or an anomaly where the model served a pre-cached result instead of synthesizing one with citations.
For frequently asked questions that do not require real-time synthesis, the system may skip independent links, which explains why some expanded overviews appeared without any links at all.
