Skip to main content
859.396.8606 Get a quote

SEO & AEO glossary, defined and scored.

122 terms that decide whether you rank in Google and get cited by ChatGPT, Claude, Perplexity, Gemini and Google's AI answers. Each one in plain English, with an example and a score from 0 to 10 for how much it really moves results.

  • 122terms defined
  • 9sections
  • 18top priorities for every site

Scores as of September 30, 2026.

Video still: of 122 variables, 83 score 5 and up and move the needle, while 39 score 4 or below and are over-hyped or risky

Hype or helpful?All 122 variables, sorted · 3:51

How to read the scores

One score. Two kinds of search.

Each of the 122 variables gets one score from 0 to 10 for how much it moves real outcomes: ranking well in Google organic results and being cited or recommended by LLMs (ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews and AI Mode).

Three rules shape the numbers. First, a score reflects impact for a typical business site; a 5 for hreflang becomes a 9 for a site selling in ten countries, and those context swings are noted. Second, risky tactics are scored on expected value after penalty risk, not on their best-case upside. Third, many technical items are gatekeepers: they rarely lift rankings on their own, but failing them blocks rankings entirely, so they score high.

A useful mental model: Google and LLMs are both trying to answer one question, "which source is the most trustworthy, relevant, and useful answer here?" Technical SEO decides whether you are even in the room. Content and keyword work decide whether you are relevant. Links, mentions, and entity signals decide whether you are believed.

Tap a score range to show only those terms.

All 122, by score
83score 5 and up
39score 4 or below

Crawlability

10/10

Crawlability is whether a bot (Googlebot, Bingbot, GPTBot, ClaudeBot, PerplexityBot) can reach your pages by following links and requesting URLs without being blocked. Blockers include robots.txt disallows, login walls, links hidden inside JavaScript click events, infinite redirect loops, and firewalls or bot-protection services that challenge crawlers.

Analogy: your site is a library and crawlers are inspectors. If the doors are locked or the hallways dead-end, the best books inside are never catalogued. A common real-world failure is a Cloudflare or WAF rule that blocks AI crawlers by default, which silently removes a brand from ChatGPT and Perplexity answers.

Why 10: nothing else matters if bots cannot reach the page. This applies doubly to LLMs, since many AI crawlers do not render JavaScript and give up faster than Googlebot.

#

Indexability

10/10

Indexability is whether a crawled page is allowed into, and accepted by, the search index. Pages can be crawlable but not indexable because of a noindex meta tag, an X-Robots-Tag header, a canonical pointing elsewhere, thin or duplicate content, or a quality judgment by Google ("Crawled, currently not indexed" in Search Console).

Example: a staging-site noindex tag accidentally pushed to production can deindex an entire site within days. That single line of code is one of the most common causes of catastrophic traffic loss.

Why 10: a page that is not indexed cannot rank in Google, and AI Overviews and Gemini draw from Google's index. Bing's index similarly feeds ChatGPT search and Copilot.

#

XML sitemaps

5/10

An XML sitemap is a file (usually /sitemap.xml) listing the URLs you want indexed, often with last-modified dates. It is a discovery aid, not a ranking factor. Think of it as a table of contents handed to the crawler, useful when the internal link structure alone might miss pages.

Example: a news site with 50,000 articles uses sitemaps segmented by type and date so new articles are found within minutes. Sitemaps also give clean per-section index coverage reports in Search Console.

Why 5: essential for large, new, or poorly linked sites, but a small site with solid internal linking gains little. Accurate lastmod dates do help fresh content get recrawled, which matters for AI search freshness.

#

Robots.txt

7/10

Robots.txt is a plain-text file at your domain root that tells crawlers which paths they may or may not request. It controls crawling, not indexing: a disallowed URL can still appear in results if other sites link to it. In the AI era it also governs which AI bots (GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot, Google-Extended) can access your content.

Analogy: it is the "Staff Only" sign on doors. Polite visitors obey it, and it is the main lever for deciding whether AI systems can read and cite you. Blocking OAI-SearchBot, for example, removes you from ChatGPT search citations.

Why 7: misconfiguration is catastrophic and the AI-bot decisions are strategic. Once set correctly, though, it needs little ongoing effort.

#

Canonical tags

7/10

A canonical tag (<link rel="canonical" href="...">) tells search engines which URL is the master version when several URLs show the same or very similar content. Common sources of duplicates: tracking parameters, sort orders, print versions, HTTP vs HTTPS, and trailing slashes.

Example: /shoes?color=red&sort=price and /shoes?sort=price&color=red are the same page. A canonical pointing both to /shoes?color=red consolidates link signals onto one URL instead of splitting them. Note that canonicals are hints; Google may override them if other signals disagree.

Why 7: central for e-commerce and large CMS sites, where parameter duplication can dilute rankings badly. Lower stakes for a 30-page brochure site.

#

Site architecture / URL structure

8/10

Site architecture is how pages are organized and linked: the hierarchy of categories, subcategories, and detail pages, and how many clicks each page sits from the homepage. URL structure is the visible reflection of that hierarchy (/services/roofing/metal-roof-repair/).

Analogy: a well-organized department store puts related items in clearly labeled sections a short walk from the entrance. A flat, logical structure (important pages within three clicks) concentrates authority on key pages and tells search engines and LLMs what your site is about.

Why 8: architecture shapes how authority flows, how topics are understood, and how efficiently bots crawl. It is hard to fix later, so getting it right early compounds.

#

HTTPS / SSL

6/10

HTTPS encrypts traffic between browser and server using an SSL/TLS certificate. Google confirmed it as a lightweight ranking signal in 2014, and browsers label HTTP sites "Not secure."

Example: an HTTP checkout page triggers browser warnings that kill conversions and trust. Mixed content (HTTPS pages loading HTTP images or scripts) can also break features.

Why 6: today it is table stakes. Nearly every competitor already has it, so it gives no edge, but lacking it is a visible trust and ranking penalty.

#

Core Web Vitals (LCP, INP, CLS)

5/10

Core Web Vitals are Google's user-experience metrics measured from real Chrome users. LCP (Largest Contentful Paint) is how fast the main content appears; good is under 2.5 seconds. INP (Interaction to Next Paint, which replaced FID in March 2024) is how fast the page responds to clicks and taps; good is under 200 ms. CLS (Cumulative Layout Shift) is how much the layout jumps while loading; good is under 0.1.

Analogy: a restaurant where food arrives fast (LCP), the waiter responds when you wave (INP), and the table does not move while you eat (CLS). Example: an ad slot that loads late and pushes the article down causes high CLS and accidental clicks.

Why 5: Google calls page experience a tie-breaker. Relevance and authority beat speed, but in competitive SERPs where content is comparable, good vitals help, and they lift conversions regardless.

#

Page speed optimization

6/10

Page speed optimization is the broader engineering work behind fast pages: reducing server response time (TTFB), minifying and deferring JavaScript and CSS, caching, compressing assets, and cutting third-party scripts. Core Web Vitals measure the outcome; page speed work is how you get there.

Example: removing three unused marketing tags and deferring a chat widget can cut load time by a second or more. Speed also affects crawl efficiency; a faster server lets Googlebot fetch more pages per visit, and AI fetchers that retrieve pages in real time may time out on slow ones.

Why 6: slightly above CWV because it also affects crawl rate, AI retrieval reliability, bounce rate, and conversions.

#

Mobile-friendliness / responsive design

8/10

Responsive design means one site adapts its layout to any screen size. Google uses mobile-first indexing, meaning it crawls and ranks based on the mobile version of your pages. If content, links, or structured data are missing on mobile, Google effectively does not see them.

Example: a site that hides its FAQ section and half its internal links behind a desktop-only menu loses those signals entirely under mobile-first indexing.

Why 8: most searches happen on mobile and Google ranks the mobile view. Nearly all modern themes are responsive, so the main risk is content parity, not layout.

#

Structured data / schema markup (JSON-LD)

7/10

Structured data is code (usually JSON-LD in the page head) that labels your content with a shared vocabulary from Schema.org: this is a Product with a price of $49, this is an Organization with these social profiles, this Person wrote this Article. It powers rich results (review stars, product prices, breadcrumbs, events) and helps machines understand entities and relationships.

Analogy: your page is a letter; schema is the clearly printed address label. The postal worker could read the letter to figure out where it goes, but the label removes the guesswork.

Why 7: it is not a direct ranking factor, but it unlocks rich results that lift click-through rate and clarifies entities for Google's Knowledge Graph. The LLM benefit is real but indirect. Note that Google retired FAQ rich results on May 7, 2026, after retiring HowTo in 2023, so markup choices should follow what Google still displays.

#

Hreflang / international targeting

4/10

Hreflang attributes tell Google which language and regional version of a page to show to which users (for example en-us, en-gb, es-mx). They prevent the UK version of a page from ranking in the US and stop near-identical English versions from being treated as duplicates.

Example: a SaaS company with /us/pricing in dollars and /uk/pricing in pounds uses hreflang so British searchers see pound prices. Implementations must be reciprocal (each page references all others), which is where most errors happen.

Why 4: it scores 0 for single-country sites and 9 for multinational ones. The average is low because most sites do not need it.

#

Pagination handling

3/10

Pagination handling is how you structure content split across multiple pages (/blog/page/2, /category?page=3). Google dropped support for rel=next/prev in 2019, so best practice now is plain crawlable links to each page, self-referencing canonicals on each page, and not noindexing deep pages that hold unique products or articles.

Example: a category page with 400 products across 20 pages; if pages 2 to 20 are canonicalized to page 1, the products listed only there may never get discovered.

Why 3: matters mainly for large e-commerce, forums, and blogs. A secondary concern for most sites.

#

JavaScript rendering / SSR vs CSR

8/10

Client-side rendering (CSR) builds the page in the visitor's browser with JavaScript; the raw HTML is nearly empty. Server-side rendering (SSR) or static generation sends fully formed HTML. Googlebot can render JavaScript, but in a deferred second wave, and most AI crawlers (GPTBot, ClaudeBot, PerplexityBot) do not execute JavaScript at all.

Analogy: CSR sends a flat-pack box and assembly instructions; SSR sends the finished chair. Google will assemble it eventually. Most AI bots look at the box, see no chair, and leave.

Example: a React single-page app with product descriptions loaded via API calls may be invisible to ChatGPT and Perplexity even while ranking fine in Google.

Why 8: for JavaScript-heavy sites, this is the difference between existing and not existing in AI search. It is the most under-appreciated technical issue for LLM visibility.

#

Duplicate content management

6/10

Duplicate content is substantially identical text appearing at more than one URL, on your site or across sites. Google does not penalize ordinary duplication, but it picks one version to show and splits link signals among the copies. Management tools include canonicals, 301 redirects, parameter handling, noindex, and writing unique content.

Example: a franchise with 40 location pages that differ only in the city name creates near-duplicates; Google may index a few and ignore the rest.

Why 6: rarely dramatic on small sites, but scaled duplication (location pages, product variants, syndicated articles) wastes crawl budget and dilutes rankings.

#

Redirect management (301, 302, chains)

6/10

Redirects send users and bots from one URL to another. A 301 is permanent and passes ranking signals to the destination; a 302 is temporary and tells Google to keep the original indexed. Chains (A to B to C) and loops slow crawling and can leak signal.

Analogy: a forwarding address after you move. One forward works fine; a chain of five forwarding addresses means some mail gets lost.

Example: during a site migration, failing to 301 old URLs to their new equivalents is the most common cause of a 30 to 70 percent traffic drop.

Why 6: low day-to-day, but critical during migrations, redesigns, and URL changes, when mistakes can erase years of accumulated authority.

#

HTTP status codes / error handling (404, 5xx)

6/10

Status codes are the server's response type. 200 means OK, 301/302 mean redirect, 404 means not found, 410 means gone permanently, 5xx means server error. "Soft 404s" are empty or error pages that wrongly return 200.

Example: 404s on deleted pages are normal and harmless. Recurring 5xx errors are not; if Googlebot repeatedly hits server errors it slows crawling and can drop pages from the index.

Why 6: occasional 404s do not hurt rankings. Persistent 5xx errors, broken internal links, and 404s on pages with valuable backlinks do.

#

Log file analysis

3/10

Log file analysis means reading your server's raw access logs to see exactly which bots requested which URLs, how often, and what status codes they got. It is the only ground truth for crawler behavior, and it now shows which AI crawlers visit and which pages they fetch.

Example: logs reveal Googlebot spends 60 percent of its visits on filter URLs with no search value, while new product pages are crawled once a month.

Why 3: a diagnostic, not a lever. Invaluable on large sites, overkill for small ones.

#

Crawl budget optimization

3/10

Crawl budget is the number of URLs a search engine will crawl on your site in a given period, shaped by server capacity and how much Google wants your content. Optimization means cutting waste (duplicate, parameter, and low-value URLs) so crawlers spend time on pages that matter.

Analogy: an inspector with eight hours a day. If half the day goes to identical storage closets, the new wing never gets inspected.

Why 3: Google says crawl budget matters mainly for sites with more than roughly a million pages, or tens of thousands that change daily. Most sites never hit the limit.

#

Faceted navigation control

4/10

Faceted navigation is the filter system on category pages (size, color, brand, price). Each combination can create a unique URL, producing millions of near-duplicate pages. Control means deciding which filter pages deserve indexing ("red running shoes" has search demand) and blocking or canonicalizing the rest.

Example: a site with 10 filters of 10 values each can theoretically generate billions of URL combinations, swamping crawlers.

Why 4: a site-killer for large e-commerce if ignored, irrelevant elsewhere. The e-commerce section scores its opportunity side separately.

#

CDN / hosting performance

5/10

A CDN (content delivery network) stores copies of your site on servers around the world so users load from a nearby location. Hosting quality determines server response time, uptime, and capacity to handle crawler and user traffic.

Example: a site on overloaded shared hosting with 1.5-second server response will struggle with Core Web Vitals no matter how optimized the front end is. CDN bot-protection settings can also accidentally block AI crawlers, which ties back to crawlability.

Why 5: uptime and speed are foundations, and cheap hosting creates a ceiling. Beyond "fast and reliable," further investment shows diminishing returns.

#

Image optimization (compression, WebP, lazy loading)

4/10

Image optimization means serving images at the right dimensions, compressed in modern formats (WebP, AVIF), with lazy loading for images below the fold and eager loading for the main hero image. Images are usually the heaviest part of a page.

Example: a 4 MB hero photo converted to a 180 KB WebP can cut LCP by over a second. A common mistake is lazy-loading the hero image, which delays LCP.

Why 4: mostly a page-speed contributor. It also enables Google Images traffic, which matters a lot for e-commerce, recipes, and design niches.

#

Orphan page detection

4/10

An orphan page has no internal links pointing to it. Bots can only find it via sitemaps or external links, and it receives no internal authority. Detection means comparing your crawl against sitemaps, analytics, and logs to find pages that exist but are unlinked.

Analogy: a room in your house with no door from any hallway. It exists, but nobody wanders into it, and visitors assume it is unimportant.

Example: landing pages built for paid campaigns, or old blog posts dropped from pagination, often become orphans.

Why 4: fixing orphans is quick, cheap, and reliably helps the affected pages. But it rarely moves sitewide performance unless many important pages are orphaned.

#

Title tags

8/10

The title tag (<title>) is the page's official name, shown as the blue clickable headline in search results and in browser tabs. It is still one of the strongest on-page relevance signals and the biggest driver of click-through rate from the SERP.

Example: "Metal Roof Repair in Lexington, KY | Same-Week Service | Acme Roofing" beats "Services | Acme." Google rewrites roughly 60 percent of titles it considers unhelpful, so the title must accurately describe the page, ideally under about 60 characters.

Why 8: a strong relevance signal and the main CTR lever, at almost no cost. Small title changes often produce measurable ranking and click gains within weeks.

#

Meta descriptions

3/10

The meta description is a short summary (around 150 to 160 characters) that may appear under the title in search results. It is not a ranking factor, and Google rewrites it most of the time, preferring a snippet from the page that matches the query.

Analogy: the blurb on the back of a book. It does not change where the book is shelved, but it can persuade someone to pick it up.

Why 3: zero direct ranking value and frequently ignored. A good one modestly improves CTR on pages where Google uses it.

#

Header hierarchy (H1 to H6)

5/10

Headers structure a page into an outline: one H1 for the main topic, H2s for major sections, H3s for subsections. They help users scan, help Google understand topical subsections, and help LLMs and AI Overviews extract self-contained passages.

Example: a guide titled "How to Choose a Metal Roof" with H2s for cost, durability, styles, and installation lets Google rank a specific section (passage ranking) for "how long do metal roofs last."

Why 5: a meaningful structure and extraction aid, especially for AI, but Google states that skipping levels or using multiple H1s does not hurt rankings.

#

Keyword placement / density

3/10

Keyword placement means putting the target term in high-signal spots: title, H1, the first paragraph, a subheading, the URL. Keyword density is the percentage of words that are the keyword. Placement still matters a little; density is an obsolete concept from the early 2000s.

Example: a page about "commercial roof coatings" should say that phrase plainly near the top. Repeating it 40 times ("keyword stuffing") is a spam signal that can hurt rankings.

Why 3: basic placement is worth a few minutes per page. Modern semantic search understands synonyms and topics, so density targets are wasted effort.

#

Content optimization

9/10

Content optimization is the process of making a page the best possible answer for its target query: matching search intent, covering the subtopics searchers expect, answering likely follow-up questions, adding unique expertise, and presenting it clearly.

Analogy: two contractors bid on the same job. One sends a one-line quote; the other sends a clear scope, timeline, photos of past work, and answers your questions before you ask. Both are "relevant"; only one wins.

Example: a page targeting "best CRM for small business" that compares options in a table, states pricing, includes hands-on testing notes, and recommends picks by use case will outrank a generic 500-word overview.

Why 9: content is what Google ranks and what LLMs quote. Helpfulness and intent match are the core of every modern ranking system.

#

Internal linking

8/10

Internal links are links from one page on your site to another. They help crawlers discover pages, distribute authority (PageRank) from strong pages to weaker ones, and use anchor text to tell Google what the destination is about.

Analogy: internal links are the hallway signs and the recommendations of a trusted employee. "If you liked this, see our guide to X" passes both the visitor and credibility along.

Example: linking from a high-traffic blog post about roof leaks to the "emergency roof repair" service page can lift that service page several positions with no new backlinks.

Why 8: one of the highest-ROI, fully controllable levers in SEO. Most sites underuse it.

#

Anchor text optimization

5/10

Anchor text is the clickable text of a link. For internal links, descriptive anchors ("metal roof repair pricing") are a relevance signal. For backlinks, anchor text is more sensitive: too many exact-match keyword anchors from other sites looks manipulative.

Example: "click here" tells Google nothing; "compare our roof warranty options" tells it what the destination covers.

Why 5: reliably useful and free for internal links. The off-page version is covered under anchor text distribution.

#

Image alt text

3/10

Alt text is a short written description of an image in the HTML (alt="..."). It exists for screen readers and accessibility, and it helps Google Images understand the picture.

Example: alt="standing seam metal roof on a two-story brick house" describes the image; alt="roof roofing roofer best roof" is stuffing.

Why 3: small effect on web rankings, real effect on image search and accessibility compliance. Do it for users first.

#

URL slugs

3/10

The slug is the readable end of a URL: /metal-roof-repair/ rather than /page?id=4827. Good slugs are short, lowercase, hyphenated, and descriptive.

Example: a user deciding between two results trusts acme.com/metal-roof-repair more than acme.com/p/4827-x. Google has said words in URLs are a very lightweight signal.

Why 3: set them well on new pages, but never change existing slugs just for SEO; the redirect risk outweighs the tiny gain.

#

Content freshness / updating

7/10

Freshness is how recently content was meaningfully updated. Google's "query deserves freshness" systems favor recent content for time-sensitive topics (prices, laws, product reviews, "best X in 2026"), and LLM search tools strongly prefer recent sources when retrieving live results.

Example: "Best Project Management Software" updated in September 2026 with current pricing outranks the same article last touched in 2023. Changing the date without changing content is detectable and does not work.

Why 7: high for time-sensitive topics and AI retrieval, low for evergreen topics like "how photosynthesis works." Regular refreshes are one of the cheapest ways to recover declining pages.

#

Semantic HTML

4/10

Semantic HTML uses elements that describe meaning rather than appearance: <article>, <nav>, <main>, <table>, <ol>, <header>, <footer>. It helps machines separate main content from navigation and boilerplate.

Analogy: labeling the boxes when you move ("kitchen," "books") versus writing nothing. The movers can still unpack, but labels speed everything up.

Example: a comparison presented as a real HTML <table> is far easier for Google and LLMs to extract than the same data built from styled <div> blocks or embedded in an image.

Why 4: modest direct impact, but it improves extraction for AI and featured snippets, and it improves accessibility.

#

Outbound linking strategy

3/10

Outbound links point from your site to other sites. Linking to authoritative, relevant sources (studies, official documentation, original data) supports claims and signals that your content is well-researched. Paid or sponsored outbound links must use rel="sponsored" or nofollow.

Example: a health article citing the CDC and a peer-reviewed study reads as more trustworthy to users and quality raters than one making unsupported claims.

Why 3: Google says outbound links are not a direct ranking boost. They matter as part of credibility and E-E-A-T, especially in health, finance, and legal topics.

#

Content depth / comprehensiveness

8/10

Depth means covering a topic thoroughly enough that the searcher does not need to go elsewhere: the main question, the related subtopics, edge cases, and practical next steps. It is not the same as word count; a 600-word page can be complete and a 5,000-word page can be padded.

Analogy: a good doctor's visit answers your question, then answers the three follow-ups you had not thought of yet.

Example: a page on "roof replacement cost" that covers cost by material, by roof size, regional variation, hidden costs, financing, and insurance outranks one giving a single national average.

Why 8: comprehensive pages rank for hundreds of long-tail variations and give LLMs more extractable answers. The penalty for padding is real, so depth must be useful depth.

#

Readability

5/10

Readability is how easily people can read and understand the content: short paragraphs, plain language, clear structure, scannable formatting, defined jargon. Readability scores are not ranking factors, but readability drives engagement, which Google observes through its user-interaction systems.

Example: a wall-of-text legal page loses readers in seconds; the same information as short sections with plain-language summaries keeps them.

Why 5: an indirect but real factor. Clear, well-structured writing is also easier for LLMs to quote accurately.

#

Digital PR

9/10

Digital PR is creating newsworthy stories, data, or expert commentary and pitching it to journalists so publications cover you and link to you. It produces the highest-authority links (major news sites, trade publications) and simultaneously creates brand mentions in exactly the sources LLMs trust most.

Example: a moving company analyzes 50,000 of its own moves and publishes "The States Americans Are Leaving in 2026." The study is picked up by regional news outlets and real estate blogs, earning dozens of links and citations.

Why 9: the rare tactic that is top-tier for both Google and LLMs at once. Expensive and skill-intensive, but the most durable form of off-page work.

#

Guest posting

4/10

Guest posting is writing articles for other websites in exchange for exposure and, usually, a link back. Done on genuinely relevant, editorially strict publications, it builds expertise and real referral traffic. Done at scale on "write for us" sites that exist to sell posts, it is a link scheme that Google targets.

Example: a cybersecurity founder writing a byline for a respected industry publication is valuable. Buying 40 posts on generic blogs with exact-match anchors is not.

Why 4: quality guest posts help, but the tactic has been heavily abused and Google discounts most of those links. Its best use today is authority and visibility, not link volume.

#

HARO / journalist outreach

6/10

HARO (Help A Reporter Out) was a service connecting journalists with expert sources; it was shut down and relaunched in different forms, and similar platforms (Qwoted, Featured, Source of Sources, Connectively and its successors) fill the role. The tactic is responding to journalist requests with expert quotes, earning mentions and links in articles.

Example: a CPA answers a reporter's request about 2026 tax deadline changes and gets quoted, with a link, in a personal-finance article on a national outlet.

Why 6: produces high-authority links and expert mentions LLMs pick up. These platforms are now flooded with AI-written pitches, so success requires speed and genuine expertise.

#

Unlinked brand mention reclamation

5/10

This means finding articles that mention your brand, product, or people without linking to you, then politely asking the author to add a link. The author already likes you enough to mention you, so conversion rates are relatively high.

Example: a trade magazine writes "tools like Acme Scheduler have changed field service." A short email asking them to link the name often succeeds.

Why 5: cheap, white-hat, and good conversion. Limited by how often you are mentioned, and the unlinked mention already carries some value for LLMs on its own.

#

Social syndication

3/10

Social syndication is distributing content across social platforms (LinkedIn, X, Facebook, Pinterest, YouTube) and republishing it on platforms like Medium or LinkedIn articles. Social links are nofollow and not a ranking factor, but distribution gets content seen by people who may link to it, and social posts themselves can rank and be cited.

Example: a data study shared on LinkedIn reaches an editor who then covers it with a link.

Why 3: indirect value for Google. Slightly higher for LLMs, since platforms like LinkedIn and YouTube are frequently cited in AI answers.

#

Anchor text distribution

4/10

This is the mix of anchor texts in your backlink profile: branded ("Acme Roofing"), naked URL ("acmeroofing.com"), generic ("this site"), partial match ("roofing tips from Acme"), and exact match ("metal roof repair Lexington"). Natural profiles are mostly branded and generic. Heavy exact-match anchors were the main target of Google's Penguin update.

Example: if 60 percent of your links use "best divorce lawyer Dallas," it looks built rather than earned.

Why 4: mainly a risk metric. Earned links produce a natural distribution automatically.

#

Brand mentions / citations

9/10

Brand mentions are references to your brand name across the web, whether linked or not: news articles, reviews, forums, podcasts, social posts, comparison articles. Google has long used unlinked mentions as a soft signal. For LLMs, mentions are the core signal: models learn and retrieve which brands are associated with which topics from how often and how positively they are mentioned together.

Analogy: reputation in a small town. People recommend the plumber everyone talks about, not the one with the most signs.

Example: when someone asks ChatGPT "best CRM for real estate agents," the brands named are those most frequently and consistently mentioned alongside that phrase in reviews, listicles, Reddit threads, and articles.

Why 9: studies of AI visibility consistently find brand-mention frequency is among the strongest predictors of being cited or recommended by LLMs, ahead of backlinks.

#

Influencer collaboration

4/10

Influencer collaboration means partnering with creators (YouTube, Instagram, TikTok, newsletters, blogs) to feature your product. Benefits for search are indirect: brand searches increase, mentions spread, and YouTube videos and creator posts are cited by AI answers. Paid partnerships must be disclosed and links marked sponsored.

Example: a kitchen-knife brand gifted to 30 cooking YouTubers ends up named in video titles like "Best Chef's Knife Under $100," which both rank in Google and appear in AI answers.

Why 4: more a brand and demand-generation lever than an SEO lever, but it feeds brand mentions and branded search, which both matter.

#

Podcast / interview placements

5/10

Getting your experts interviewed on podcasts and industry shows earns show-notes links, brand mentions, and transcripts that are indexed and crawled. It also builds the person-entity of your experts, a growing E-E-A-T and LLM signal.

Example: a founder appearing on 12 industry podcasts in a year gets 12 relevant links, 12 transcripts that tie their name to their topic, and audiences that search for the brand afterward.

Why 5: reliable, relevant, white-hat, and good for authority building. Individual podcast links are usually modest in strength.

#

Private blog networks (PBNs)

1/10

A PBN is a group of websites, often built on expired domains with leftover authority, that one person or company controls and uses only to link to its money sites. It is the classic black-hat link scheme.

Analogy: writing your own letters of recommendation under fake names.

Example: buying 20 expired domains of defunct local businesses, putting thin articles on each, and linking them all to your site. Google routinely deindexes networks and penalizes beneficiaries, sometimes years later.

Why 1: can produce short-term gains in some niches, which is why it persists among churn-and-burn affiliates. For any real business, expected value is negative.

#

Skyscraper technique

4/10

The skyscraper technique, coined by Brian Dean around 2015, is finding content that already has many backlinks, creating something substantially better (more current, more complete, better designed), then contacting everyone who links to the original and asking them to link to yours.

Example: an outdated 2019 guide to drone regulations has 200 links. You publish a fully updated 2026 guide with an interactive state-by-state map and pitch the linkers.

Why 4: the "make something better" half is excellent and drives content quality. The outreach half has been so overused that response rates are low, unless your asset offers real information gain.

#

Search intent classification

9/10

Search intent is the goal behind a query. Informational queries seek knowledge ("how do heat pumps work"). Navigational queries seek a specific site ("Trane login"). Commercial investigation queries compare options before buying ("best heat pump brands 2026"). Transactional queries are ready to act ("heat pump installation near me," "buy Mitsubishi mini split").

Analogy: a customer walking into a store and asking "what is this?" needs a different response from one asking "where do I pay?" Hand the browser a checkout form and they leave.

Example: a service company trying to rank its sales page for "how do heat pumps work" will fail, because Google shows educational guides for that query. The fix is a guide that links to the sales page.

Why 9: intent mismatch is the most common reason good pages do not rank. The fastest way to diagnose intent is to look at what currently ranks: Google is showing you what it believes satisfies the query.

#

Search volume analysis

6/10

Search volume is the estimated number of monthly searches for a keyword, reported by tools like Semrush, Ahrefs, and Google Keyword Planner. It helps prioritize opportunities by potential traffic.

Example: "roof repair" may show 40,000 monthly searches nationally; "roof repair Lexington KY" may show 90. The second is far more valuable to a local roofer despite lower volume.

Caveats: tool volumes are estimates that can be off by 50 percent or more, they are often rounded averages that hide seasonality, and one page usually ranks for hundreds of variations, so a single keyword's volume understates real traffic.

Why 6: necessary for prioritization, but over-reliance leads teams to chase vanity keywords and ignore valuable low-volume terms.

#

Keyword difficulty scoring

5/10

Keyword difficulty (KD) is a tool's estimate, usually 0 to 100, of how hard it would be to reach page one for a keyword. Most tools calculate it mostly from the backlink strength of the currently ranking pages.

Example: a new site with little authority should target KD 0 to 20 terms first and build up, rather than chasing "life insurance" at KD 95.

Why 5: a useful rough filter. It ignores content quality, intent fit, and your own topical authority, and each tool calculates it differently, so a manual SERP review is more reliable.

#

Long-tail keywords

7/10

Long-tail keywords are longer, more specific queries with lower individual volume: "standing seam metal roof cost per square foot in Kentucky" rather than "metal roof." Collectively they make up the majority of all searches, and they convert better because the searcher's need is clearer.

Analogy: the main aisle of a bookstore holds bestsellers everyone fights over; the long tail is the thousands of niche titles that, added together, sell more.

AI-era note: prompts typed into ChatGPT and Google AI Mode are much longer than classic searches, so long-tail, conversational coverage maps directly onto how people now ask AI.

Why 7: the most realistic path for newer sites, higher conversion, and strong alignment with conversational AI queries.

#

Keyword clustering / topic grouping

7/10

Keyword clustering groups keywords that can be satisfied by the same page, usually by checking whether Google shows the same URLs for them. This decides how many pages to build and which keywords each page targets.

Example: "metal roof cost," "how much is a metal roof," and "metal roofing price per square" share the same top results, so one page covers all three. "Metal roof vs shingles" has different results and needs its own page.

Why 7: prevents both thin duplicate pages and overstuffed single pages. It is the bridge from a keyword list to a sound site architecture.

#

Competitor keyword gap analysis

6/10

Gap analysis compares the keywords your competitors rank for with the keywords you rank for, revealing topics you are missing. Tools like Semrush's Keyword Gap automate this.

Example: three competing law firms rank for "what happens at a deposition" and "how long does a personal injury case take"; your firm ranks for neither, so those become content priorities.

Why 6: a fast, data-backed way to find proven opportunities. It only shows what competitors already do, so pair it with original ideas.

#

Question-based keywords (PAA mining)

7/10

These are the questions people ask about a topic, mined from Google's "People Also Ask" boxes, autocomplete, forums, Reddit, AlsoAsked, and sales or support calls.

Example: PAA for "metal roof" includes "Do metal roofs make your house hotter?" and "Do metal roofs attract lightning?" Answering them directly on the page captures those queries and the objections buyers actually have.

Why 7: questions map almost one-to-one to how people prompt LLMs and how AI Overviews structure answers. Direct question-and-answer coverage is one of the most transferable SEO and AEO skills.

#

Zero-volume keywords

5/10

Zero-volume keywords are terms that tools report as having no measurable search volume, often because the data samples are too small or the query is new. Many still get real searches, especially specific, high-intent B2B or niche queries.

Example: "HIPAA compliant scheduling software for chiropractors" might show 0 volume yet bring in the exact buyers a niche SaaS needs. Customers and sales calls are the best source of these terms.

Why 5: high conversion, low competition, and well-matched to the specific prompts people give LLMs. Easy wins that competitors relying only on tools miss.

#

Keyword cannibalization detection

5/10

Cannibalization happens when several of your pages target the same query and intent, so Google alternates between them or ranks neither well. Detection uses Search Console (multiple URLs receiving impressions for the same query) or rank tracking.

Analogy: two of your own salespeople pitching the same client with different prices. The client trusts neither.

Example: a blog with "Best Running Shoes 2024," "Best Running Shoes 2025," and "Top Running Shoes" competes with itself. Merging into one updated page usually lifts rankings.

Why 5: fixing true cannibalization can produce quick gains, but it is often over-diagnosed; two pages ranking for one query is fine when they serve different intents.

#

SERP feature analysis per keyword

6/10

This is checking which features appear in the results for each target keyword (AI Overview, featured snippet, People Also Ask, local pack, video carousel, shopping results, images) and how far down the first organic result sits.

Example: a keyword where an AI Overview, four ads, and a map pack sit above the first organic result may deliver a fraction of the clicks its volume suggests. A keyword with a video carousel means you need a video, not just an article.

Why 6: directly shapes what content format to build and how much traffic is realistically available, which matters more each year as AI Overviews absorb clicks.

#

Topical authority building

9/10

Topical authority is the perception, by Google and by LLMs, that your site is a go-to expert on a subject. It is built by covering a topic thoroughly and consistently over time (the core concepts, the subtopics, the comparisons, the beginner and advanced questions) rather than publishing scattered posts on unrelated subjects.

Analogy: a general store versus a specialist shop. If you want a rare fishing lure, you trust the tackle shop that stocks 2,000 lures over the supermarket with a small rack.

Example: a site with 80 interlinked, high-quality articles on home solar (costs, incentives by state, battery storage, panel comparisons, installer questions) will often outrank a high-authority general publication for new solar queries.

Why 9: Google's systems increasingly favor focused, demonstrably expert sites, and LLMs associate brands with the topics they cover most consistently.

#

Content hubs / pillar pages / clusters

7/10

This is the structural model for topical authority. A pillar page covers a broad topic comprehensively ("Complete Guide to Home Solar"). Cluster pages go deep on subtopics ("Solar Tax Credit Explained," "Best Solar Batteries"). All clusters link to the pillar and the pillar links to each, forming a hub.

Analogy: a wheel with a hub and spokes. The structure makes the relationship between pages obvious to readers and crawlers alike.

Why 7: an effective way to organize internal linking and coverage. The model is a means, not the goal; the score is lower than topical authority because the structure without the quality does little.

#

Content gap analysis

6/10

Content gap analysis finds topics, questions, and formats your audience wants that your site does not cover. Inputs include competitor gap analysis, PAA questions, customer support tickets, sales calls, forums, and AI prompt testing (asking ChatGPT and Perplexity your buyers' questions and seeing whose content is cited).

Example: a B2B software company discovers prospects constantly ask "how does this integrate with QuickBooks?" but there is no page answering it. Creating one captures searches and AI citations for that exact question.

Why 6: consistently produces high-value topics. Its value depends on executing what it finds.

#

Content pruning / consolidation

6/10

Pruning removes or noindexes content that is outdated, thin, or irrelevant. Consolidation merges several weak, overlapping pages into one strong page with 301 redirects. The idea is that low-quality pages can drag down how Google evaluates your whole site.

Analogy: pruning a fruit tree. Cutting weak branches sends energy to the strong ones.

Example: a blog with 600 posts, 400 of which get zero traffic, merges 120 overlapping posts into 30 thorough guides and deletes 150 obsolete announcements; organic traffic rises over the next quarter.

Why 6: often produces real gains on large, older sites affected by helpful-content and core updates. Must be done carefully, since deleting pages with backlinks or long-tail traffic can backfire.

#

Content refreshing

7/10

Refreshing means substantively updating existing pages: new data, current prices, new sections answering newer questions, improved examples, removed outdated information. It differs from pruning in that the page stays and improves.

Example: an article that slipped from position 3 to position 11 over 18 months is updated with 2026 data and two new sections; within weeks it climbs back to the top five.

Why 7: often the highest-ROI content activity, because the page already has authority and links. AI search tools also strongly prefer recently updated sources.

#

Content formats (guides, listicles, comparisons, tools)

7/10

Format is the shape the content takes: how-to guides, listicles ("10 best..."), comparison pages ("X vs Y"), templates, calculators, interactive tools, original research, case studies, glossaries, and videos. The right format is the one Google and users already reward for that query.

Example: "mortgage calculator" demands a working calculator, not an article about calculators. "HubSpot vs Salesforce" demands a comparison table. "Best CRM" demands a list with clear picks.

Why 7: format mismatch is a common reason pages fail. Comparisons and "best of" lists are also among the content types LLMs cite most often for commercial questions.

#

Programmatic SEO

5/10

Programmatic SEO creates many pages from a template and a dataset: "[service] in [city]," "[tool A] vs [tool B]," "[currency] to [currency] converter," "flights from [city] to [city]." Done well, each page has unique, useful data. Done badly, it is thousands of near-identical pages, which Google's scaled-content-abuse policy targets.

Example: Zillow's property pages, Tripadvisor's "things to do in [city]," and Wise's currency pages are successful programmatic SEO because each page has genuinely unique data. A site generating 5,000 city pages that only swap the city name is not.

Why 5: enormous upside for businesses with real proprietary data, and real penalty risk for those without. The average outcome across all attempts is modest.

#

User-generated content

5/10

User-generated content (UGC) is content created by users: reviews, Q&A, comments, forum posts, community answers, photos. It adds fresh, authentic, long-tail language and first-hand experience that a brand cannot write itself.

Example: an e-commerce product page with 300 customer reviews ranks for queries like "does [product] run small" because customers used those exact words. Reddit's massive visibility gains in Google since 2023 are the most dramatic example of UGC value.

Why 5: valuable where a community exists, but it requires moderation (spam, low-quality posts) and takes time to build.

#

Author bios / entity building

6/10

Author bios show who wrote the content and why they are qualified: credentials, experience, photo, links to their other work and profiles. Entity building goes further, establishing the author (and the company) as a recognized entity that Google and LLMs can connect across the web, through consistent names, Person schema with sameAs links, LinkedIn profiles, speaking, and bylines elsewhere.

Example: a medical article written by "Admin" versus one by a named, board-certified physician with a bio linking to her hospital profile. Quality raters, and increasingly algorithms, treat these very differently.

Why 6: central for health, finance, legal, and other high-stakes ("Your Money or Your Life") topics, lower elsewhere. It supports E-E-A-T and helps LLMs attribute expertise.

#

Information gain / original data

9/10

Information gain is the amount of new information a page adds beyond what already ranks: original research, proprietary data, first-hand testing, unique expert insight, real photos, case studies with numbers. Google holds a patent on scoring information gain, and its helpful-content guidance asks whether content provides original information, reporting, research, or analysis.

Analogy: if ten articles all summarize the same three sources, an eleventh summary adds nothing. The article that ran its own experiment is the one everyone else ends up citing.

Example: instead of "How Long Do Tires Last?" rehashed from manufacturer claims, a tire shop publishes data from 12,000 tire replacements showing actual mileage by brand. Journalists link to it, competitors cite it, and LLMs quote its numbers.

Why 9: this is the antidote to AI-generated sameness. It earns links naturally, earns LLM citations (models need sources for specific facts), and is very hard for competitors to copy.

#

Content briefs / SOP creation

4/10

A content brief is a document given to a writer before writing: target keyword and intent, audience, required subtopics and questions, competing pages, internal links, unique angle, and expert sources. SOPs (standard operating procedures) document the repeatable process for briefs, writing, editing, optimization, and publishing.

Example: a brief that specifies "include our 2025 install-cost data, answer the five PAA questions, and link to the financing page" produces far better first drafts than "write 1,500 words about metal roofs."

Why 4: an enabler of quality and consistency at scale. The impact is real but flows through other variables.

#

Google Business Profile optimization

10/10

Google Business Profile (GBP, formerly Google My Business) is the free listing that appears in Google Maps and the local "map pack." Optimization means choosing the most accurate primary category (the single most important field), adding secondary categories, services, products, accurate hours, high-quality photos, a description, attributes, Q&A, and regular posts, and verifying the listing.

Analogy: GBP is your storefront on Google's main street. Many customers decide based on it alone, without ever visiting your website.

Example: a dentist who changes the primary category from "Dentist" to the more specific "Cosmetic Dentist" (when accurate) may suddenly rank for veneer and whitening searches. Primary category changes are among the fastest-acting local tweaks.

Why 10: for local businesses, the map pack often captures more clicks and calls than the organic results, and the GBP is the asset that ranks there. Gemini and Google AI Mode also lean heavily on GBP data for local answers.

#

NAP consistency (Name, Address, Phone)

5/10

NAP consistency means your business name, address, and phone number appear identically across your website, GBP, directories, social profiles, and data aggregators. Inconsistencies (old addresses, tracking numbers, "St." vs "Street" is fine, but different suite numbers or old phone numbers are not) confuse search engines about which entity is which.

Example: a business that moved in 2021 still listed at the old address on Yelp, Bing Places, and Apple Business Connect may see map rankings split or suppressed, and customers drive to the wrong place.

Why 5: it used to be a major factor. Google is now much better at resolving entities, so consistency is more hygiene than lever. Major errors (wrong address, wrong phone on big platforms) still matter, especially since LLMs pull business details from these sources.

#

Local citations / directories

4/10

Citations are listings of your business on other sites: Yelp, Bing Places, Apple Business Connect, Facebook, BBB, Angi, Healthgrades, Avvo, chamber of commerce, and industry or city directories. They confirm your existence and details.

Example: a plumber listed on the top 30 general and home-services directories has a solid foundation. Being listed on 300 low-quality directories adds nothing.

Why 4: a core set of accurate, high-quality citations matters, especially Apple and Bing (which feed Siri, Apple Maps, ChatGPT, and Copilot). Beyond that core, returns diminish quickly.

#

Review generation & management

9/10

This means systematically asking happy customers to leave reviews (on Google first, then relevant platforms), responding to all reviews professionally, and handling negative reviews constructively. Review count, rating, recency, and the words customers use all influence local rankings and conversions. Buying fake reviews or "review gating" (only asking happy customers through a filter) violates Google and FTC rules.

Analogy: reviews are word-of-mouth made visible to strangers. A 4.8 rating with 400 recent reviews beats a 5.0 with six reviews from 2022.

Example: an auto shop that texts a review link after every service grows from 40 to 350 Google reviews in a year; reviews that mention "brake repair" and "Lexington" help it rank for those searches.

Why 9: a top local ranking factor, the top conversion factor, and a heavy input into how LLMs describe and recommend local businesses ("highly rated for...").

#

Location pages / service area pages

7/10

Location pages are dedicated pages for each physical location (address, hours, staff, directions, local reviews). Service area pages target cities or neighborhoods a business serves without having an office there. Each needs genuinely unique, local content to avoid being thin duplicates.

Example: a roofing company serving Lexington, Georgetown, and Richmond creates three pages, each with local project photos, testimonials from that city, permit information, and weather-specific roofing advice. A template that only swaps the city name rarely ranks.

Why 7: the main way to rank organically in cities beyond your office address, and they give LLMs clear evidence of where you operate.

#

Local pack / map pack ranking

8/10

The local pack is the block of three businesses with a map that appears for local-intent searches ("dentist near me," "roofers Lexington"). Ranking there depends on GBP optimization, reviews, proximity, links, citations, and on-site signals. As a variable, it means actively tracking and optimizing your map visibility across the area you serve, often with grid-based rank trackers that show rankings from different points in the city.

Example: a geo-grid scan shows a business ranks #1 within a mile of its office but drops off the pack three miles away, which guides where to build location content, links, and reviews.

Why 8: for local businesses it is often the single most valuable placement in search, ahead of the organic results below it.

#

Geo-targeted content

5/10

Geo-targeted content is content written for a specific locality: local guides, cost data for your area, local regulations, neighborhood-specific advice, community news, and local case studies.

Example: "Kentucky Roof Permit Requirements by County" or "How Lexington's Freeze-Thaw Cycles Affect Your Roof" builds local relevance and earns local links and citations.

Why 5: strengthens local relevance and creates content national competitors cannot match. A supporting lever rather than a primary one.

#

Local schema markup

4/10

Local schema is LocalBusiness structured data (or a more specific subtype like Dentist, Plumber, or Restaurant) on your website, describing name, address, phone, hours, geo-coordinates, service area, and links to your profiles via sameAs.

Example: a law firm adds LegalService schema on each office page, linking to its GBP, Avvo, and LinkedIn profiles.

Why 4: helps search engines and AI systems confirm business details and connect your site to your GBP entity. Not a strong ranking factor on its own.

#

Proximity / relevance / prominence signals

8/10

Google states that local rankings are based on three factors. Proximity: how close the business is to the searcher or the location in the query. Relevance: how well the business matches what was searched (categories, services, website content, reviews). Prominence: how well-known and trusted the business is (reviews, links, mentions, overall web presence).

Analogy: when you ask a friend for a good pizza place, they weigh how close it is, whether it actually serves the pizza you want, and how much people rave about it.

Example: you cannot move your building, so proximity is mostly fixed. Relevance (categories, service pages) and prominence (reviews, local PR, links) are where effort goes.

Why 8: this is the framework every other local variable feeds. Proximity often dominates the map pack, so prominence and relevance work determines how far your radius extends.

#

Product page optimization

8/10

Product page optimization makes each product detail page (PDP) rank and convert: a descriptive title with brand, model, and key attributes; unique descriptions that answer buyer questions; full specs; multiple high-quality images and video; price and availability; shipping and return details; reviews; and links to related products and the parent category.

Example: "Men's Waterproof Hiking Boot" becomes "Merrell Moab 3 Mid Waterproof Hiking Boot, Men's, Wide Width Available." The page adds a sizing guide, a waterproofing test video, and answers to "are these good for snow?"

Why 8: product pages capture high-intent, long-tail searches, and their structured details are what AI shopping answers and comparison prompts pull from.

#

Collection page optimization

9/10

Collection (category) pages list products in a group: "Women's Trail Running Shoes," "Cast Iron Skillets." They usually target the highest-volume commercial keywords in a store, because shoppers search for categories more than specific SKUs. Optimization includes a keyword-aligned title and H1, a short helpful intro, buying-guide content or FAQs below the product grid, smart internal links to subcategories, and indexable filter combinations with real demand.

Analogy: the product page is a single item on the shelf; the collection page is the aisle sign that brings people into the aisle.

Example: a cookware store's "Cast Iron Skillets" page with a 150-word intro, a size comparison table, and a care guide below the grid outranks a competitor whose collection page is only a grid of products.

Why 9: often the most valuable pages in an online store for organic revenue, and the most commonly under-optimized.

#

Product schema (price, reviews, availability)

8/10

Product schema is Product structured data including Offer (price, currency, availability, shipping, returns) and AggregateRating and Review. It makes products eligible for rich results such as price, stock status, and review stars in organic listings, and feeds Google's Shopping Graph.

Example: a search result showing "$89.99 · In stock · 4.7 stars (1,284)" gets far more clicks than a plain blue link for the same product.

Why 8: a direct CTR lever and a key data source for AI shopping experiences. Required for competitive product SERPs.

#

Faceted navigation SEO

6/10

This is the opportunity side of faceted navigation (the risk side is in Technical SEO): deliberately making high-demand filter combinations into indexable, optimized landing pages with clean URLs, unique titles, and intros, while keeping the millions of low-value combinations out of the index.

Example: "black leather sofa" and "sectional sofa with chaise" have real search demand, so those filter combinations get static, indexable URLs. "Sofa, color: black, price: $1,200 to $1,400, sort: newest" does not.

Why 6: can unlock thousands of long-tail category rankings for large catalogs, but requires careful engineering to avoid index bloat.

#

Out-of-stock page handling

5/10

This is deciding what happens to product pages when items go out of stock or are discontinued. For temporarily out-of-stock items: keep the page live, mark availability correctly in schema, offer back-in-stock alerts and alternatives. For permanently discontinued items: 301 redirect to the replacement or closest category if one exists; otherwise return 404 or 410, or keep the page live with clear alternatives if it still earns traffic and links.

Example: deleting a popular discontinued blender page that has 40 backlinks throws away that authority. Redirecting it to the successor model preserves it.

Why 5: mishandling it at scale (mass deletions, soft 404s, redirecting everything to the homepage) can quietly erode rankings over time.

#

Product review optimization

8/10

This means collecting more, better, and more detailed customer reviews on product pages (post-purchase emails, photo and video reviews, Q&A), displaying them in crawlable HTML, marking them up with schema, and syndicating them to Google through review partners. Reviews add unique, constantly fresh content in customer language.

Example: a review that says "I'm 6'2" and the medium fit perfectly in the shoulders" answers the exact long-tail question "does [shirt] run small for tall guys."

Why 8: improves rankings, CTR (stars), and conversion together. Reviews and review aggregators are also among the most cited sources when people ask AI assistants which product to buy.

#

Merchant Center / Google Shopping feed

8/10

Google Merchant Center hosts your product feed (titles, prices, GTINs, images, availability, shipping). The feed powers free Shopping listings, the Shopping tab, product knowledge panels, and Google's Shopping Graph, which AI Overviews and AI Mode draw on for product answers. Other AI shopping experiences, such as ChatGPT's, increasingly rely on merchant product feeds too.

Example: a store with complete feeds (correct GTINs, detailed titles, shipping info, high-quality images) appears in free product listings and in AI Mode product carousels; a store without one never does, regardless of how good its website is.

Why 8: free, high-intent visibility that a website alone cannot earn, and a growing input into AI shopping answers.

#

Duplicate product description handling

5/10

Many stores use the manufacturer's stock description, which hundreds of other retailers also use. Google then picks one version to rank, usually the manufacturer or the biggest retailer. Handling means rewriting descriptions for priority products with unique detail, adding original content (reviews, Q&A, usage guides, comparison tables), and managing internal duplicates from product variants with canonicals.

Example: a niche outdoor retailer rewrites descriptions for its top 200 products with field-testing notes and fit advice, and those pages begin to outrank larger stores using the boilerplate text.

Why 5: meaningful for competitive products, but rewriting thousands of SKUs is costly, so it should be prioritized by revenue.

#

Internal search optimization

4/10

This covers two things. First, using your site-search data (what customers type into your search box) to discover demand, missing products, and new category pages to create. Second, preventing internal search result pages from being crawled and indexed, since Google treats them as low-value, infinite pages.

Example: site-search logs show 800 monthly searches for "wide calf boots" but no such category exists. Creating that collection page captures both internal and Google demand.

Why 4: a valuable research source and a technical hygiene item, but not a direct ranking driver.

#

Competitor data extraction

3/10

This is collecting competitors' public data at scale: product catalogs, pricing, category structures, titles and descriptions, review counts, and ranking keywords, using SEO tools, price-monitoring services, or scrapers. It informs pricing, assortment, and content decisions.

Example: a scrape shows a competitor's 30 top-ranking collection pages all include a buying guide and a comparison table, which shapes your own template.

Caution: scraping can violate a site's terms of service, copying competitors' descriptions creates duplicate content and potential copyright problems, and aggressive scraping can be legally risky. Commercial SEO and price-intelligence tools are the safer route.

Why 3: useful competitive intelligence, but it informs strategy rather than moving rankings, and its value depends entirely on what you do with it.

#

People Also Ask

5/10

People Also Ask (PAA) is the expandable box of related questions; clicking one reveals a short answer from a web page and loads more questions. Pages earn PAA placements by answering those questions clearly and concisely.

Example: searching "metal roof" shows PAA questions like "What are the disadvantages of a metal roof?" Each answer is a citation opportunity.

Why 5: individually low-traffic, but PAA is the best free map of what searchers and AI systems consider the follow-up questions for a topic, and answering them strengthens topical coverage.

#

Knowledge panels

6/10

A knowledge panel is the information box (usually on the right on desktop) about an entity: a company, person, organization, or place. It pulls from Google's Knowledge Graph, using sources like Wikipedia, Wikidata, official websites, and social profiles. Verified entities can claim their panel and suggest edits.

Example: searching a well-known company name shows a panel with its logo, founders, headquarters, stock price, and social links.

Why 6: a panel signals that Google understands your entity, which correlates strongly with how accurately LLMs describe you. It is mainly a branded-search and trust asset rather than a traffic driver.

#

Image packs

3/10

An image pack is a row of image thumbnails embedded in regular results, linking to Google Images. It appears for visually oriented queries.

Example: "modern farmhouse kitchen ideas" shows an image pack; a design blog with original, well-labeled photos can capture clicks there.

Why 3: high value in visual niches (home decor, fashion, recipes, products) and negligible elsewhere. Earned through original images, descriptive filenames, alt text, and fast image delivery.

#

Video carousels

5/10

A video carousel is a horizontal row of videos in results, dominated by YouTube but also including other platforms and on-site videos with VideoObject schema. They appear for how-to, review, and demonstration queries.

Example: "how to replace a toilet flapper" shows a video carousel above most articles. A plumbing company with a clear three-minute YouTube video can outrank national publishers.

Why 5: a real opportunity in how-to and review niches, and YouTube videos are among the most frequently cited sources in AI Overviews and LLM answers. Requires video production capacity.

#

Review stars

6/10

Review stars (review snippets) are star ratings shown under a result, drawn from Review or AggregateRating schema. Since 2019, Google does not show stars for "self-serving" reviews of your own business on your own site (LocalBusiness and Organization types), but they remain available for products, recipes, software, books, courses, and similar items.

Example: two identical product listings, one showing "4.8 stars (2,341 reviews)" and one plain. Studies consistently find the starred result gets a meaningfully higher click-through rate.

Why 6: a significant CTR boost for eligible content types, especially e-commerce, at low cost once reviews and schema are in place.

#

Top stories

3/10

Top stories is the carousel of recent news articles shown for newsworthy queries. Inclusion favors established news publishers with strong reputations, fast publishing, and original reporting; no specific markup is required, though Article schema and fast pages help.

Example: "Federal Reserve rate decision" shows top stories from major outlets within minutes of an announcement.

Why 3: essential for news publishers (for them it is a 9), largely irrelevant for most businesses except when they are the subject of news.

#

Local pack

8/10

The local pack is the map with three business listings shown for local-intent searches; it is covered in depth in the Local SEO section (map pack ranking and Google Business Profile). As a SERP feature it typically appears above the organic results and captures a large share of clicks and calls for local queries.

Example: "emergency plumber" on a phone shows a map and three plumbers with call buttons before any organic result.

Why 8: for local businesses it is often the most valuable real estate on the page. For non-local businesses it scores 0, which is why this is not a 10.

#
Where to spend effort first

Access. Relevance. Trust.

For any site, the priority order is: make sure bots can reach and index you, build the best and most original content on a focused topic, then earn the links and mentions that make Google and LLMs trust you.

  1. 1Access

    Make sure bots can reach and index you.

    Technical SEO
  2. 2Relevance

    Build the best and most original content on a focused topic.

    On-page, keywords and content
  3. 3Trust

    Earn the links and mentions that make Google and LLMs trust you.

    Off-page and AI search

Local businesses should add Google Business Profile (10), review generation (9), local pack ranking (8), and proximity / relevance / prominence work (8) to the top of this list. Online stores should add collection pages (9), plus product pages, product schema, product reviews, and the Merchant Center feed (all 8).

Your site, scored

Where does your site stand?

We'll check your site against the variables that matter most, for Google and for AI answers, and tell you what to fix first.