What Is Off-Page SEO? Links, Mentions and AI Citations

Off-page SEO is everything other websites, platforms and people say about your brand. Google reads those signals to decide which pages deserve to rank. AI assistants like ChatGPT, Perplexity, Gemini and Google's AI Mode read many of the same signals to decide which brands to recommend.

That second audience changed the job. Ten years ago, off-page SEO meant link building. In 2026 it also means getting your brand onto the roundups, reviews and community threads that AI answers pull from. G2's 2026 survey of 1,076 B2B software buyers found that 51% now start their research in an AI chatbot instead of a search engine. A buyer who asks ChatGPT for the best tool in your category gets an answer built from third-party pages, so your presence on those pages decides whether you make the shortlist.

At Flying Cat we run off-page work for B2B SaaS teams as part of their SEO and AI search programs. This guide covers what counts as off-page SEO today, how Google and AI assistants weigh each signal, which techniques we use with clients and which tactics we refuse to touch.

What is off-page SEO?

Off-page SEO is the set of actions you take outside your own website to build its authority, trust and visibility in search engines and AI answers. The core signals are backlinks, brand mentions, placements on third-party lists, reviews, digital PR coverage, community discussion and, for businesses with a physical location, your Google Business Profile.

On-page SEO tells search engines what a page covers. Technical SEO makes sure they can crawl and render it. Off-page SEO tells them whether anyone else vouches for it.

Why off-page signals drive rankings and AI recommendations

Links remain one of the clearest signals Google has. Semrush's ranking factor research lists domain Authority Score, a link-based metric, among the stronger predictors of a page's position, with referring domains and backlinks close behind (Semrush).

The newer research looks at AI visibility. It points the same way with one twist, because brand mentions count for more than links.

• Brand mentions correlate most strongly with AI Overview visibility. Ahrefs studied 75,000 brands and found branded web mentions had a 0.664 correlation with how often a brand appears in Google's AI Overviews. Backlink count came in at 0.218 (Ahrefs, 2025).
• Authority still counts in AI answers. Semrush analyzed 1,000 domains across ChatGPT, Gemini, AI Overviews and Perplexity and found a 0.65 Pearson correlation between Authority Score and AI mentions (Semrush, 2025).
• "Best X" lists dominate ChatGPT's sources. In Ahrefs' study of 750 top-of-funnel prompts, "best X" blog lists made up 43.8% of the page types ChatGPT cited (Ahrefs, 2025).

The study authors present these as correlations. Our own client data lines up with them. When we earn a client a spot on a list that AI assistants already cite, the client's share of AI answers usually moves within weeks. A new link from an unrelated blog rarely moves anything.

How Google and AI assistants weigh off-page signals

Backlinks

A backlink is a link from another site to yours. Google treats each one as a vote, then weights that vote by where it comes from and how it sits on the page. These factors make the biggest difference.

• Relevance of the linking site and page. A link from a site and article on your topic carries more weight than one from an unrelated niche.
• Authority of the linking page. A page that itself has strong links passes more value than an orphaned page on the same domain.
• Placement on the page. Links inside the body of an article carry more weight than links in footers, sidebars and author boxes. Google's reasonable surfer patent describes weighting links by how likely a reader is to click them.
• Anchor text. The clickable words tell Google what the target page covers. Descriptive anchors help, but a profile full of exact-match keyword anchors looks manufactured.
• Unique referring domains. Ten links from ten relevant sites usually beat a hundred links from one site.

Link attributes and nofollow

Site owners can label links with rel attributes. Google asks for rel="sponsored" on paid placements and ads. It asks for rel="ugc" on user-generated links such as forum posts and comments, with rel="nofollow" for links where neither applies (Google Search Central).

Many older guides say nofollow links pass zero value. Google retired that rule in 2019, when it announced that nofollow, sponsored and ugc would all be treated as hints about which links to count (Search Engine Land). For AI visibility, Semrush found nofollow and followed links had almost identical correlations with AI mentions (Semrush, 2025). A nofollow mention on a roundup that ChatGPT cites can be worth more to your pipeline than a followed link nobody reads.

Brand mentions

A brand mention is any reference to your company on another site, with or without a link. Google has long had patents describing mentions as implied links. AI assistants go further. A language model learns which brands belong to a category from the text it reads and retrieves, so the number of relevant pages that name you next to your category has a direct effect on whether the model names you.

This is why we track mentions as their own line in client reports, separate from links. The two do different jobs.

Third-party lists that AI assistants cite

When a buyer asks "what is the best procurement software" or "top LMS for nonprofits", AI assistants retrieve a small set of pages and summarize them. Many of those pages are listicles and roundups on review sites, industry publications and vendor blogs. If your brand is missing from those specific URLs, it is usually missing from the answer.

Citation patterns also shift fast. Semrush tracked more than 100 million AI citations over 13 weeks in 2025 and saw Reddit's share of ChatGPT responses fall from close to 60% to about 10% in roughly six weeks (Semrush, 2025). The pages you target should come from current citation data for your prompts, refreshed monthly. A list of high-authority domains from last year will mislead you.

Reviews

Reviews on G2, Capterra, TrustRadius and Gartner Peer Insights do three jobs for a B2B SaaS company. They influence buyers directly. They feed the comparison pages that AI assistants cite. They give Google and language models a body of independent text describing what your product does and who it suits. For local businesses, Whitespark's 2026 Local Search Ranking Factors survey reports that review signals gained importance for map pack rankings while traditional citations kept declining (Whitespark).

Off-page SEO techniques for B2B SaaS

Build links to your commercial pages

Link budgets often go to blog posts that will never produce a demo. We point links at product, comparison and solutions pages instead, because those pages sit closest to a signup.

Timing helps too. Trevor.io linked every new page from older pages that already ranked and were crawled often, then built external links only once a page started to rank. Pages reached page one in about a month. Traffic beat the forecast by 402% (Trevor.io case study). At Leapsome, the team moved part of its content budget into link building for its priority pages. We built around 50 targeted links to those pages. Referring domains grew about 30% in twelve months (Leapsome case study).

For more tactics, see our guide to link building strategies that work.

Get placed on the listicles AI assistants cite

This is the off-page technique with the fastest effect on AI visibility we have measured. The process has four steps.

1. Track the prompts your buyers ask in ChatGPT, Perplexity, Gemini, Copilot and Google's AI answers.
2. Pull the URLs those assistants cite for each prompt. Flag the ones that name competitors and leave you out.
3. Pitch the authors of those exact pages, starting with existing listicles, then new pages on the same domain, then new domains.
4. Measure how often AI recommends you on those prompts after the placement goes live.

D2L shifted its outreach from general link building to the third-party roundups that appear as sources in AI answers. Its share of tracked AI answers recommending D2L grew from 12% to 33% in nine months. D2L also moved from 11th to 7th among learning platforms (D2L case study). An enterprise procurement platform we work with earned placements on the roundups and industry publications that AI assistants use as sources. ChatGPT answers citing the company's own site rose from 25% to 41% (enterprise procurement platform case study).

Our AEO guide chapter on off-site visibility goes deeper on finding and pitching these pages.

Run digital PR around original data

Journalists and list authors need fresh numbers. A survey, a benchmark from your product data or an annual report gives them a reason to cite you. Each citation becomes both a link and a mention. Original data also feeds AI answers, since assistants favor pages with specific, attributable figures. Our roundup of digital PR agencies covers when to bring in a specialist.

Build review volume on the platforms buyers check

Pick the two or three review sites your buyers and AI assistants use for your category. Ask happy customers at natural moments, such as after onboarding or a renewal. Reply to every review, including the critical ones. Then check whether the comparison pages on those sites appear in AI citations for your prompts, because many of them do.

Take part in communities

Reddit threads, LinkedIn posts, Slack groups and niche forums shape how buyers and models talk about your category. Links from these platforms usually carry rel="ugc" or nofollow, so the value comes from mentions, referral visits and the discussion itself. Show up as a named expert who answers questions well. Brands that post promotional replies get downvoted and ignored.

Use podcasts, events and partnerships

Guest spots on industry podcasts, speaking slots and integration partner pages all create mentions on relevant sites. Partner directories and integration marketplaces are often cited by AI assistants for "tools that integrate with X" prompts, so keep your listings complete and current.

Keep your Google Business Profile current if you serve local customers

For a B2B SaaS company that sells nationally or globally, a Google Business Profile does little beyond branded searches. If you have offices buyers visit, or you serve a local market, keep the profile complete, collect reviews and keep your name, address and phone consistent across directories.

Plan for markets where link building is off the table

Some companies cannot do link outreach at all. At an ecommerce marketing platform's France program, company policy ruled out link building and legal blocked comparison articles naming competitors. The client still tripled its top-3 rankings in France through native-written content and a topic-cluster focus (ecommerce marketing platform case study). Strong on-page work can carry a program when off-page options are limited, though it takes longer in competitive categories.

How we score a placement before paying for it

Link building is mostly pay to play. Publishers charge fees for inclusion. The fee alone tells you nothing about whether the placement will move rankings or AI answers. We score every brand mention target against citation data before a client spends anything.

For an AI customer service platform, we evaluated a roundup that AI assistants cited in 33 responses across six buying prompts. It appeared in 45% of Copilot answers to "best AI customer service platform" questions, where the client showed up in 7% of answers at the time. We estimated the client could reach roughly 3 to 4 times its visibility on those prompts if it was picked up in a third to half of the answers citing that page. We also expressed the placement fee as a share of revenue, which came to about 1% of one average organic deal. That framing gave the client's leadership a clear yes or no.

The scoring rules we use on every account follow this pattern.

• Citation evidence first. Brand mention targets come from the URLs AI assistants cite for the client's tracked prompts.
• Quality and relevance check. Every target gets a first pass for topic fit, traffic and spam signals. Maeva approves each one before outreach starts.
• Separate tracks. SEO links and AI mentions are tracked and reported separately, because they have different goals.
• Measured after it goes live. We compare the client's mention rate on the relevant prompts before and after each placement.

In the AI tax research category, this kind of targeted work showed up quickly. In the first two months of the engagement, an AI tax research platform we work with saw its citation rate in AI answers triple from 7.8% to 23.9%. Its Perplexity presence rose from 15.1% to 42.6%.

Off-page tactics to avoid

Google's spam policies define link spam as links created mainly to manipulate rankings (Google Search Central). These tactics fall into that category and put your site at risk.

• Paid links that pass ranking credit. Paying for a link is fine when it carries rel="sponsored" or rel="nofollow". Paying for a followed link with a keyword anchor violates Google's policies.
• Excessive link exchanges. "Link to me and I'll link to you" deals and partner pages built only for cross-linking count as spam.
• Private blog networks and link farms. Google devalues sites that exist only to sell links. Links from them waste budget and can trigger a manual action. In one backlink review, we traced 74% of a client's lost links to spam domains.
• Automated links. Tools that generate forum, comment or directory links at scale create exactly the pattern Google's spam systems look for.
• Guest posts and press releases with keyword anchors. Google lists sponsored articles and press releases that pass ranking credit as link spam.
• Fake reviews. Review platforms remove them. In the US, the FTC's 2024 rule on fake reviews allows civil penalties for violations (Davis Wright Tremaine).

How to measure off-page SEO

Domain Rating and Domain Authority scores tell your leadership very little. Measure off-page work by what it changes.

AI ranking position is unreliable from one run to the next, so we report visibility as a percentage across many prompts. Our guides on measuring link building and measuring AI search cover the setup in detail.

How Flying Cat runs off-page SEO

We run link building inside our full-service SEO and GEO and AEO programs. We do not sell it on its own, because links only pay off when the pages they point to are worth ranking.

Each client's budget splits between two tracks. Traditional SEO links carry anchor text to commercial pages and get measured by rankings and demos. Brand mentions target the listicles, roundups and reviews that AI assistants cite and get measured by AI mention rate. Publisher fees pass through at cost. We score each placement before the client pays for it. Maeva leads strategy on every account, with freelance specialists who know the client's vertical. We run programs in English, French, German, Italian and Spanish.

If you want to see where your brand is missing from the sources AI assistants cite, book a call. We will show you which pages to target first and what we would do about them.

Frequently asked questions about off-page SEO

What is the difference between on-page and off-page SEO?

On-page SEO covers what you control on your own pages, such as content, title tags, headings and internal links. Off-page SEO covers signals from other sites and platforms, such as backlinks, brand mentions, reviews and list placements. Search engines use on-page signals to judge relevance and off-page signals to judge authority.

Is off-page SEO the same as link building?

Link building is the largest part of off-page SEO, but the discipline also includes brand mentions, reviews, digital PR, community participation and Google Business Profile management. For AI visibility, unlinked brand mentions on cited pages often do more than links.

Do nofollow links help SEO?

Google treats nofollow as a hint and can still count the link. Nofollow links also drive referral traffic and brand mentions. Semrush found nofollow and followed links correlate almost equally with AI mentions.

How long does off-page SEO take to work?

New links typically take weeks to a few months to affect rankings, depending on competition. AI visibility can move faster when a placement lands on a page assistants already cite. Our clients usually start to see movement within 60 days.

Does off-page SEO affect AI search?

Off-page signals shape AI search directly. AI assistants retrieve and cite third-party pages when they answer category questions. Brands named on those pages get recommended more often. Ahrefs found branded web mentions were the factor most strongly correlated with visibility in Google's AI Overviews.