You Fixed Everything. Nothing Changed.

Topical Authority Audit; Googlebot Makeup; AI Search Metrics; ChatGPT Citation Influences; News Publications in AI Citations; Amazon Wins; and Much More!

Author

Yoyao
March 14, 2026 • Est. Learning Time: 19 minutes

In partnership with

FIRST …

Here's what a normal SEO audit tells you: where rankings are moving, what technical issues exist, where traffic is falling, what backlinks you're missing.

Here's what it usually does not tell you: whether you're building authority in the topics that actually matter to your brand, your audience and your offer.

You can have a technically clean site, solid on-page SEO and a reporting dashboard full of movement, while your content strategy is drifting further away from the territory you actually need to own. Your SEO tool is listing competitors that aren't even in your industry.

In other words, you can build a very polished website for no one in particular.

Most SEO audits start from site diagnostics. A better model starts somewhere else: your brand, your audience and your topical map. From there, the question changes. It is no longer just "what is broken?" It becomes "are we building the right authority?"

That is what the Topical Authority Audit is built to answer. It measures three pillars:

  • Content Authority - is your foundation right?

  • Market Authority - are you winning real ground against competitors?

  • AI Authority - can AI search actually find, extract and cite you?

I built a 21-question scorecard that walks you through all three pillars. It takes about 15 minutes and no tools.

And the pattern that keeps showing up is this: most sites are not weak across all three. They are strong in two and broken in the third. That is where authority quietly stalls, and it is the one pattern standard audits never show you.

AD

How Jennifer Anniston’s LolaVie brand grew sales 40% with CTV ads

For its first CTV campaign, Jennifer Aniston’s DTC haircare brand LolaVie had a few non-negotiables. The campaign had to be simple. It had to demonstrate measurable impact. And it had to be full-funnel.

LolaVie used Roku Ads Manager to test and optimize creatives — reaching millions of potential customers at all stages of their purchase journeys. Roku Ads Manager helped the brand convey LolaVie’s playful voice while helping drive omnichannel sales across both ecommerce and retail touchpoints.

The campaign included an Action Ad overlay that let viewers shop directly from their TVs by clicking OK on their Roku remote. This guided them to the website to buy LolaVie products.

Discover how Roku Ads Manager helped LolaVie drive big sales and customer growth with self-serve TV ads.

The DTC beauty category is crowded. To break through, Jennifer Anniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

SEO + GEO

Martin Splitt talks with Gary Illyes in Google’s Search Off the Record episode and explains that Googlebot is not one crawler, but a client name layered on top of Google’s shared crawling system. Google’s crawler infrastructure separates fetchers from crawlers, with fetchers handling single-URL requests and crawlers processing URL streams in the background. Read the transcript.

Dan Taylor suggests that Google’s Universal Commerce Protocol could force ecommerce SEO to optimize for purchases inside Google, not just clicks back to a site. He frames Merchant Center data as the operational layer that now shapes whether products can move directly from discovery to checkout across Search and Gemini. Marketers need to prepare policy data, product attributes, exclusions, and post-purchase workflows before developers wire up the technical side. That makes feed quality, eligibility logic, and account linking part of the SEO conversation much earlier than most teams are used to.

Francine Monahan discusses how iPullRank is moving AI search measurement beyond rankings by tracking which content traits correlate with citations across generative engines. The study reframes SEO reporting around new channel metrics, while still treating traffic, conversions and revenue as the bottom-line layer.

  • They analyzed more than 79,000 URL-query pairs across ChatGPT, Claude, Perplexity and Google.

  • 292% lift in citation probability on mid-tail queries when entity density was optimized.

  • Cited URLs averaged 1,800 words, while non-cited URLs averaged 1,200 words.

  • Pages ranking outside the top 10 saw a stark drop-off in AI citation likelihood.

  • Long-tail pages should align about 90% of on-page entities with primary user intent to avoid diluting relevance.

  • Among the seven tracked metrics, Content-Keyword Cosine and Strategic Entity Richness showed the strongest relationship with AI citations.

Oshen Davidson shares that ChatGPT citation visibility depends on more than retrieval, because the system expands prompts through internal fan-out searches and cites only a small share of what it finds. The report makes the case that SEO teams need to track where citation opportunity is created, and lost, across both Google rankings and the follow-up queries ChatGPT generates.

  • AirOps analyzed 548,534 retrieved pages across 15,000 original prompts and 43,233 total original-plus-fan-out queries.

  • Only 15% of retrieved pages were cited, meaning 85% were discovered but never made it into the final answer.

  • 32.9% of cited pages that appeared in a top-20 SERP were found only through a fan-out query, not the original prompt.

  • 95% of ChatGPT fan-out queries had zero monthly search volume by traditional metrics.

  • Pages ranking number one in Google were cited 43.2% of the time, which the report says is 3.5 times higher than pages outside the top 20 results.

  • Pages with 50% or greater title-query overlap had a 20.1% citation rate, versus 9.3% for pages with less than 10% overlap.

  • Sites with domain authority between 20 and 80 accounted for 63.6% of all citations, while DA 80 to 100 sites underperformed after retrieval with a 15.0% citation rate.

Jason Barnard says technical SEO still hides too many distinct failure modes inside the lazy shorthand of crawl and index. The piece breaks infrastructure into five sequential gates, discovery, selection, crawling, rendering fidelity and conversion fidelity, and says the earliest broken gate sets the ceiling for everything downstream. One of the more useful ideas is rendering fidelity, which reframes the problem around how much of a page a bot actually sees after building it, not whether the page seems fine in a browser.

Vince Nero analyzed 4 million citations across 3,600 prompts and found that news publications account for a meaningful but secondary slice of AI visibility. His data shows news citations cluster around bottom-funnel comparison prompts, while syndication, newswires, and local news barely register.

  • News publications accounted for 14.09% of all citations in the study.

  • Evaluative prompts drove news citations 18% of the time, compared with 7.26% for brand awareness prompts.

  • True news and editorial content made up 81% of news citations.

  • Affiliate, review, and list-style content reached 39.07% of evaluative prompt citations, 8.67% of brand awareness citations, and 1.52% of informational citations.

  • Syndicated news represented 6.2% of news citations and 0.9% of the full dataset.

  • Newswires accounted for 0.21% of the full analysis.

  • Internal press releases and newsroom content made up 2.98% of citations overall, but 18.15% of ChatGPT citations.

  • Energy drove 31% of news publication citations, while entertainment contributed 17.57%.

Shahzad Abbas says publishers are getting hit hard by AI Overviews overall, but breaking news is proving to be a rare exception. Using Define Media Group portfolio data, he says that search traffic losses are being partially offset by strong breaking-news performance across Top Stories and especially Discover, which is starting to behave like a standalone growth channel rather than a side effect of web search.

Matt G. Southern reports that post-update Discover data is pointing to a more regionalized feed, where local publishers may be losing national distribution rather than disappearing altogether. The most useful detail is the state-level breakdown, which suggests some publishers held steady in their home markets while getting pushed out of out-of-state feeds after Google’s Discover core update. He also notes that third-party tracking tools disagree on some winners and losers, which makes the findings directional rather than final.

SEO + GEO Ripples

  • Google may be hitting self-promotional LLM listicles again, with Mordy Oberstein pointing to fresh industry chatter around exact-match domains and enterprise sites that scaled thin, self-serving “best” pages for AEO-style visibility. The pattern suggests these tactics can produce short-term gains, but may still run into Google’s organic quality systems once the footprint gets obvious.

  • Search Console now splits branded and non-branded queries with a built-in filter in the Performance report and a matching clicks breakdown in Insights. Google says the classification is handled by an internal AI-assisted system, includes brand variants and product names, and is limited to eligible top-level properties with enough query volume.

AI

Samanyou Garg analyzed 119 ChatGPT conversations and found that GPT-5.4 and GPT-5.3 behave like two different search systems, not just two versions of the same model. The study’s core finding is that GPT-5.4 is far more brand-direct, while GPT-5.3 leans heavily on third-party review and media sites, which changes how we need to think about AI visibility.

  • GPT-5.4 sent 56% of its citations to brand websites, while GPT-5.3 sent only 8%.

  • The two models shared just 7% of cited sources across the tested prompts.

  • GPT-5.4 averaged 8.5 fan-out queries per prompt, compared with 1.0 for GPT-5.3.

  • GPT-5.4 cited 138 pricing pages, versus just 4 for GPT-5.3.

  • GPT-5.4 averaged 14.8 cited sources per prompt, while GPT-5.3 averaged 5.8.

  • GPT-5.3 sent 92% of citation traffic to third-party sites such as Forbes, TechRadar, Tom’s Guide and Reddit.

  • The study says 75% of GPT-5.4’s cited domains did not appear in Google or Bing results for the same raw prompt

My Take: This is the clearest proof yet that AI visibility is fragmenting at the model level. If one version sends users to review ecosystems and another goes straight to pricing and product pages, brands need broader coverage and tighter first-party assets.

Matt G. Southern reports that a federal judge has granted Amazon a preliminary injunction blocking Perplexity from using Comet or other AI agents to access password-protected parts of Amazon’s platform. The ruling matters because the court treated user permission and platform authorization as separate requirements, which cuts against the idea that an agent can inherit full rights from the user alone. Amazon also won an order requiring Perplexity to destroy Amazon data collected through Comet, while the broader case continues. The bigger impact is that agentic commerce may need explicit approval from both the user and the platform owner, not just a clever browser wrapper around user credentials.

My Take: This looks like one of the first real legal stress tests for agentic browsing. If this reasoning holds, a lot of AI product teams will have to stop assuming user consent is enough to unlock any logged-in workflow on the web. The technical challenge is only half the story now. Platform permission is becoming part of the product design surface.

AI Ripples

  • Google introduced Ask Maps and Immersive Navigation to make Maps more conversational and more spatially aware, with Ask Maps rolling out in the U.S. and India on Android and iOS, and Immersive Navigation starting in the U.S. Google says Ask Maps draws on information from more than 300 million places and over 500 million contributors, while navigation now uses Gemini to surface richer 3D guidance, route tradeoffs and arrival details.

  • Google launched Gemini Embedding 2 in public preview as its first natively multimodal embedding model, mapping text, images, video, audio and documents into one shared embedding space. The model supports more than 100 languages, up to 8,192 text tokens, six images, 120 seconds of video, native audio input and PDFs up to six pages long.

  • OpenAI introduces IH-Challenge as a training dataset for getting models to follow higher-trust instructions more reliably when the hierarchy of system, developer, user and tool messages conflict. The claim is that better instruction hierarchy improves both safety steerability and prompt injection resistance. If the model cannot consistently tell whose instruction matters most, every workflow built on top of it stays fragile. The long-term win here is not just safer agents. It is more dependable software behavior.

  • Microsoft Incident Response lays out a playbook for spotting prompt abuse before it turns into a business problem, with prompt injection framed as one of the most important AI application risks to monitor. The clearest example is an indirect attack where hidden text inside a URL fragment gets pulled into an AI summarizer’s prompt and quietly biases the output without the user doing anything obviously unsafe. The post breaks prompt abuse into three buckets, direct prompt override, extractive abuse against sensitive inputs and hidden-instruction attacks, then maps each stage of detection and response to actual controls.

  • OpenAI explains how real prompt injection now looks less like a clever string attack and more like social engineering aimed at an AI agent working through untrusted content. They say the right defense is not just better input filtering, but system design that limits what an agent can silently send, click, or trigger when manipulated. OpenAI frames this through source-sink analysis, where the risk appears when external content can steer an action such as transmitting data to a third party or following a link. Their suggestion is to add controls like Safe Url, sandboxed execution, and user confirmation before sensitive information or risky actions leave the system.

MARKETING

Paul Bannister says programmatic advertising is starting 2026 on softer footing, with cautious spending in the first half of the year and a stronger rebound expected later on. He ties that pressure to big event-driven budget shifts, macro uncertainty and the rise of retail media networks, while arguing that publishers need to adjust to a market where performance outcomes matter more than impression volume.

SOCIAL MEDIA

Facebook is tightening its push toward original content by giving creators clearer originality rules, more distribution upside, and stronger protection against copycats. The update pairs policy clarification with fresh performance data, showing how aggressively Meta is linking reach and monetization to whether a creator is adding real value. Meta says it removed more than 20 million accounts impersonating large content creators in 2025. Impersonation reports related to large content creators fell by 33% in 2025. The updated rules say unoriginal content can be deprioritized in both Feed and Reels, and repeat offenders can become non-recommendable and demonetized.

Meta will start passing location-based digital service taxes and similar regulatory fees directly onto advertisers when impressions are served in certain countries. These charges sit outside campaign budgets and show up as separate billing line items, which means the final invoice can exceed the spend cap marketers thought they were working within. Meta lists location fees of 5% in Austria, 3% in France, 3% in Italy, 3% in Spain, 5% in Türkiye, and 2% in the United Kingdom. The fee is based on where impressions are delivered, not where the advertiser is based.

Amanda Silberling reports that Meta has acquired Moltbook, the Reddit-like social network where AI agents could interact through an always-on directory, and is folding the team into Meta Superintelligence Labs. The deal stands out less for scale than for what it signals about Meta’s interest in agent-to-agent coordination as a product category. She also notes that Moltbook’s viral moment was fueled by human users impersonating agents because the system was not secure, which turned a novelty project into a public trust story.

WAYS WE CAN WORK TOGETHER

Floyi - The only AI-powered tool that builds 4-level topical maps. Don’t just plan your content strategy - make it unstoppable.

TopicalMap.com Service - Let us do the heavy lifting. We handle the research, structure, and strategy. You get a custom topical map designed to boost authority and dominate your niche and industry.

Topical Maps Unlocked 2.0 - Unlock the blueprint to ranking success. Master the art of structuring content that search engines (and your audience) love - and watch your rankings soar.

AD

Turn AI into Your Income Engine

Ready to transform artificial intelligence from a buzzword into your personal revenue generator?

HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.

Inside you'll discover:

  • A curated collection of 200+ profitable opportunities spanning content creation, e-commerce, gaming, and emerging digital markets—each vetted for real-world potential

  • Step-by-step implementation guides designed for beginners, making AI accessible regardless of your technical background

  • Cutting-edge strategies aligned with current market trends, ensuring your ventures stay ahead of the curve

Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.

LIKE DIGITAL SURFER?

Find me and others in the Digital Surfer Discord community.

I’d also love to know what you think and if you have any ideas for the newsletter. Reply or email me at [email protected].

I’d also appreciate it if you shared it with fellow digital surfers.

You currently have 0 referrals, only 3 away from receiving LinkedIn Shout-out.

And don’t forget to drag the Digital Surfer emails to your Primary Inbox 🌊 

Reply

or to participate.