Reform needs to happen before the Bank Secrecy Act gets to celebrate its next big milestone, argues Nicholas Anthony, a policy analyst at the Cato Institute.
Marketing, technology, and business leaders today are asking an important question: how do you optimize for large language models (LLMs) like ChatGPT, Gemini, and Claude?
LLM optimization is taking shape as a new discipline focused on how brands surface in AI-generated results and what can be measured today.
For decision makers, the challenge is separating signal from noise – identifying the technologies worth tracking and the efforts that lead to tangible outcomes.
The discussion comes down to two core areas – and the timeline and work required to act on them:
Tracking and monitoring your brand’s presence in LLMs.
Improving visibility and performance within them.
Tracking: The foundation of LLM optimization
Just as SEO evolved through better tracking and measurement, LLM optimization will only mature once visibility becomes measurable.
We’re still in a pre-Semrush/Moz/Ahrefs era for LLMs.
Tracking is the foundation of identifying what truly works and building strategies that drive brand growth.
Without it, everyone is shooting in the dark, hoping great content alone will deliver results.
The core challenges are threefold:
LLMs don’t publish query frequency or “search volume” equivalents.
Their responses vary subtly (or not so subtly) even for identical queries, due to probabilistic decoding and prompt context.
They depend on hidden contextual features (user history, session state, embeddings) that are opaque to external observers.
Why LLM queries are different
Traditional search behavior is repetitive – millions of identical phrases drive stable volume metrics. LLM interactions are conversational and variable.
People rephrase questions in different ways, often within a single session. That makes pattern recognition harder with small datasets but feasible at scale.
These structural differences explain why LLM visibility demands a different measurement model.
This variability requires a different tracking approach than traditional SEO or marketing analytics.
The leading method uses a polling-based model inspired by election forecasting.
The polling-based model for measuring visibility
A representative sample of 250–500 high-intent queries is defined for your brand or category, functioning as your population proxy.
These queries are run daily or weekly to capture repeated samples from the underlying distribution of LLM responses.
Tracking tools record when your brand and competitors appear as citations (linked sources) or mentions (text references), enabling share of voice calculations across all competitors.
Over time, aggregate sampling produces statistically stable estimates of your brand visibility within LLM-generated content.
Early tools providing this capability include:
Profound.
Conductor.
OpenForge.
Consistent sampling at scale transforms apparent randomness into interpretable signals.
Over time, aggregate sampling provides a stable estimate of your brand’s visibility in LLM-generated responses – much like how political polls deliver reliable forecasts despite individual variations.
Building a multi-faceted tracking framework
While share of voice paints a picture of your presence in the LLM landscape, it doesn’t tell the complete story.
Just as keyword rankings show visibility but not clicks, LLM presence doesn’t automatically translate to user engagement.
Brands need to understand how people interact with their content to build a compelling business case.
Because no single tool captures the entire picture, the best current approach layers multiple tracking signals:
Share of voice (SOV) tracking: Measure how often your brand appears as mentions and citations across a consistent set of high-value queries. This provides a benchmark to track over time and compare against competitors.
Referral tracking in GA4: Set up custom dimensions to identify traffic originating from LLMs. While attribution remains limited today, this data helps detect when direct referrals are increasing and signals growing LLM influence.
Branded homepage traffic in Google Search Console: Many users discover brands through LLM responses, then search directly in Google to validate or learn more. This two-step discovery pattern is critical to monitor. When branded homepage traffic increases alongside rising LLM presence, it signals a strong causal connection between LLM visibility and user behavior. This metric captures the downstream impact of your LLM optimization efforts.
Nobody has complete visibility into LLM impact on their business today, but these methods cover all the bases you can currently measure.
Be wary of any vendor or consultant promising complete visibility. That simply isn’t possible yet.
Understanding these limitations is just as important as implementing the tracking itself.
Because no perfect models exist yet, treat current tracking data as directional – useful for decisions, but not definitive.
Measuring LLM impact is one thing. Identifying which queries and topics matter most is another.
Compared to SEO or PPC, marketers have far less visibility. While no direct search volume exists, new tools and methods are beginning to close the gap.
The key shift is moving from tracking individual queries – which vary widely – to analyzing broader themes and topics.
The real question becomes: which areas is your site missing, and where should your content strategy focus?
To approximate relative volume, consider three approaches:
Correlate with SEO search volume
Start with your top-performing SEO keywords.
If a keyword drives organic traffic and has commercial intent, similar questions are likely being asked within LLMs. Use this as your baseline.
Layer in industry adoption of AI
Estimate what percentage of your target audience uses LLMs for research or purchasing decisions:
High AI-adoption industries: Assume 20-25% of users leverage LLMs for decision-making.
Slower-moving industries: Start with 5-10%.
Apply these percentages to your existing SEO keyword volume. For example, a keyword with 25,000 monthly searches could translate to 1,250-6,250 LLM-based queries in your category.
Using emerging inferential tools
New platforms are beginning to track query data through API-level monitoring and machine learning models.
Accuracy isn’t perfect yet, but these tools are improving quickly. Expect major advancements in inferential LLM query modeling within the next year or two.
The technologies that help companies identify what to improve are evolving quickly.
While still imperfect, they’re beginning to form a framework that parallels early SEO development, where better tracking and data gradually turned intuition into science.
Optimization breaks down into two main questions:
What content should you create or update, and should you focus on quality content, entities, schema, FAQs, or something else?
How should you align these insights with broader brand and SEO strategies?
Identify what content to create or update
One of the most effective ways to assess your current position is to take a representative sample of high-intent queries that people might ask an LLM and see how your brand shows up relative to competitors. This is where the Share of Voice tracking tools we discussed earlier become invaluable.
These same tools can help answer your optimization questions:
Track who is being cited or mentioned for each query, revealing competitive positioning.
Identify which queries your competitors appear for that you don’t, highlighting content gaps.
Show which of your own queries you appear for and which specific assets are being cited, pinpointing what’s working.
From this data, several key insights emerge:
Thematic visibility gaps: By analyzing trends across many queries, you can identify where your brand underperforms in LLM responses. This paints a clear picture of areas needing attention. For example, you’re strong in SEO but not in PPC content.
Third-party resource mapping: These tools also reveal which external resources LLMs reference most frequently. This helps you build a list of high-value third-party sites that contribute to visibility, guiding outreach or brand mention strategies.
Blind spot identification: When cross-referenced with SEO performance, these insights highlight blind spots; topics or sources where your brand’s credibility and representation could improve.
Understand the overlap between SEO and LLM optimization
LLMs may be reshaping discovery, but SEO remains the foundation of digital visibility.
Across five competitive categories, brands ranking on Google’s first page appeared in ChatGPT answers 62% of the time – a clear but incomplete overlap between search and AI results.
That correlation isn’t accidental.
Many retrieval-augmented generation (RAG) systems pull data from search results and expand it with additional context.
The more often your content appears in those results, the more likely it is to be cited by LLMs.
Brands with the strongest share of voice in LLM responses are typically those that invested in SEO first.
Strong technical health, structured data, and authority signals remain the bedrock for AI visibility.
What this means for marketers:
Don’t over-focus on LLMs at the expense of SEO. AI systems still rely on clean, crawlable content and strong E-E-A-T signals.
Keep growing organic visibility through high-authority backlinks and consistent, high-quality content.
Use LLM tracking as a complementary lens to understand new research behaviors, not a replacement for SEO fundamentals.
Redefine on-page and off-page strategies for LLMs
Just as SEO has both on-page and off-page elements, LLM optimization follows the same logic – but with different tactics and priorities.
Off-page: The new link building
Most industries show a consistent pattern in the types of resources LLMs cite:
Wikipedia is a frequent reference point, making a verified presence there valuable.
Reddit often appears as a trusted source of user discussion.
Review websites and “best-of” guides are commonly used to inform LLM outputs.
Citation patterns across ChatGPT, Gemini, Perplexity, and Google’s AI Overviews show consistent trends, though each engine favors different sources.
This means that traditional link acquisition strategies, guest posts, PR placements, or brand mentions in review content will likely evolve.
Instead of chasing links anywhere, brands should increasingly target:
Pages already being cited by LLMs in their category.
Reviews or guides that evaluate their product category.
Articles where branded mentions reinforce entity associations.
The core principle holds: brands gain the most visibility by appearing in sources LLMs already trust – and identifying those sources requires consistent tracking.
On-page: What your own content reveals
The same technologies that analyze third-party mentions can also reveal which first-party assets, content on your own website, are being cited by LLMs.
This provides valuable insight into what type of content performs well in your space.
For example, these tools can identify:
What types of competitor content are being cited (case studies, FAQs, research articles, etc.).
Where your competitors show up but you don’t.
Which of your own pages exist but are not being cited.
From there, three key opportunities emerge:
Missing content: Competitors are cited because they cover topics you haven’t addressed. This represents a content gap to fill.
Underperforming content: You have relevant content, but it isn’t being referenced. Optimization – improving structure, clarity, or authority – may be needed.
Content enhancement opportunities: Some pages only require inserting specific Q&A sections or adding better-formatted information rather than full rewrites.
Leverage emerging technologies to turn insights into action
The next major evolution in LLM optimization will likely come from tools that connect insight to action.
Early solutions already use vector embeddings of your website content to compare it against LLM queries and responses. This allows you to:
Detect where your coverage is weak.
See how well your content semantically aligns with real LLM answers.
Identify where small adjustments could yield large visibility gains.
Current tools mostly generate outlines or recommendations.
The next frontier is automation – systems that turn data into actionable content aligned with business goals.
Timeline and expected results
While comprehensive LLM visibility typically builds over 6-12 months, early results can emerge faster than traditional SEO.
The advantage: LLMs can incorporate new content within days rather than waiting months for Google’s crawl and ranking cycles.
However, the fundamentals remain unchanged.
Quality content creation, securing third-party mentions, and building authority still require sustained effort and resources.
Think of LLM optimization as having a faster feedback loop than SEO, but requiring the same strategic commitment to content excellence and relationship building that has always driven digital visibility.
From SEO foundations to LLM visibility
LLM traffic remains small compared to traditional search, but it’s growing fast.
A major shift in resources would be premature, but ignoring LLMs would be shortsighted.
The smartest path is balance: maintain focus on SEO while layering in LLM strategies that address new ranking mechanisms.
Like early SEO, LLM optimization is still imperfect and experimental – but full of opportunity.
Brands that begin tracking citations, analyzing third-party mentions, and aligning SEO with LLM visibility now will gain a measurable advantage as these systems mature.
In short:
Identify the third-party sources most often cited in your niche and analyze patterns across AI engines.
Map competitor visibility for key LLM queries using tracking tools.
Audit which of your own pages are cited (or not) – high Google rankings don’t guarantee LLM inclusion.
Continue strong SEO practices while expanding into LLM tracking – the two work best as complementary layers.
Approach LLM optimization as both research and brand-building.
Don’t abandon proven SEO fundamentals. Rather, extend them to how AI systems discover, interpret, and cite information.
AI tools can help teams move faster than ever – but speed alone isn’t a strategy.
As more marketers rely on LLMs to help create and optimize content, credibility becomes the true differentiator.
And as AI systems decide which information to trust, quality signals like accuracy, expertise, and authority matter more than ever.
It’s not just what you write but how you structure it. AI-driven search rewards clear answers, strong organization, and content it can easily interpret.
This article highlights key strategies for smarter AI workflows – from governance and training to editorial oversight – so your content remains accurate, authoritative, and unmistakably human.
Your organization will benefit from clear boundaries and expectations. Creating policies for AI use ensures consistency and accountability.
Only 7% of companies using genAI in marketing have a full-blown governance framework, according to SAS.
However, 63% invest in creating policies that govern how generative AI is used across the organization.
Source- “Marketers and GenAI- Diving Into the Shallow End,” SAS
Even a simple, one-page policy can prevent major mistakes and unify efforts across teams that may be doing things differently.
As Cathy McPhillips, chief growth officer at the Marketing Artificial Intelligence Institute, puts it:
“If one team uses ChatGPT while others work with Jasper or Writer, for instance, governance decisions can become very fragmented and challenging to manage. You’d need to keep track of who’s using which tools, what data they’re inputting, and what guidance they’ll need to follow to protect your brand’s intellectual property.”
So drafting an internal policy sets expectations for AI use in the organization (or at least the creative teams).
When creating a policy, consider the following guidelines:
What the review process for AI-created content looks like.
When and how to disclose AI involvement in content creation.
How to protect proprietary information (not uploading confidential or client information into AI tools).
Which AI tools are approved for use, and how to request access to new ones.
How to log or report problems.
Logically, the policy will evolve as the technology and regulations change.
Keep content anchored in people-first principles
It can be easy to fall into the trap of believing AI-generated content is good because it reads well.
LLMs are great at predicting the next best sentence and making it sound convincing.
But reviewing each sentence, paragraph, and the overall structure with a critical eye is absolutely necessary.
Think: Would an expert say it like that? Would you normally write like that? Does it offer the depth of human experience that it should?
“People-first content,” as Google puts it, is really just thinking about the end user and whether what you are putting into the world is adding value.
Any LLM can create mediocre content, and any marketer can publish it. And that’s the problem.
People-first content aligns with Google’s E-E-A-T framework, which outlines the characteristics of high-quality, trustworthy content.
E-E-A-T isn’t a novel idea, but it’s increasingly relevant in a world where AI systems need to determine if your content is good enough to be included in search.
According to evidence in U.S. v. Google LLC, we see quality remains central to ranking:
“RankEmbed and its later iteration RankEmbedBERT are ranking models that rely on two main sources of data: [redacted]% of 70 days of search logs plus scores generated by human raters and used by Google to measure the quality of organic search results.”
Source: U.S. v. Google LLC court documentation
It suggests that the same quality factors reflected in E-E-A-T likely influence how AI systems assess which pages are trustworthy enough to ground their answers.
So what does E-E-A-T look like practically when working with AI content? You can:
Review Google’s list of questions related to quality content: Keep these in mind before and after content creation.
Demonstrate firsthand experience through personal insights, examples, and practical guidance: Weave these insights into AI output to add a human touch.
Use reliable sources and data to substantiate claims: If you’re using LLMs for research, fact-check in real time to ensure the best sources.
Insert authoritative quotes either from internal stakeholders or external subject matter experts: Quoting internal folks builds brand credibility while external sources lend authority to the piece.
Create detailed author bios: Include:
Relevant qualifications, certifications, awards, and experience.
Links to social media, academic papers (if relevant), or other authoritative works.
Add schema markup to articles to clarify the content further: Schema can clarify content in a way that AI-powered search can better understand.
Become the go-to resource on the topic: Create a depth and breadth of material on the website that’s organized in a search-friendly, user-friendly manner. You can learn more in my article on organizing content for AI search.
Source: Creating helpful, reliable, people-first content,” Google Search Central
The do’s and don’ts of phrases and language to use.
Formatting rules such as SEO-friendly headers, sentence length, paragraph length, bulleted list guidelines, etc.
You can refresh this as needed and use it to further train the model over time.
Build a prompt kit
Put together a packet of instructions that prompts the LLM. Here are some ideas to start with:
The style guide
This covers everything from the audience personas to the voice style and formatting.
If you’re training a custom GPT, you don’t need to do this every time, but it may need tweaking over time.
A content brief template
This can be an editable document that’s filled in for each content project and includes things like:
The goal of the content.
The specific audience.
The style of the content (news, listicle, feature article, how-to).
The role (who the LLM is writing as).
The desired action or outcome.
Content examples
Upload a handful of the best content examples you have to train the LLM. This can be past articles, marketing materials, transcripts from videos, and more.
If you create a custom GPT, you’ll do this at the outset, but additional examples of content may be uploaded, depending on the topic.
Sources
Train the model on the preferred third-party sources of information you want it to pull from, in addition to its own research.
For example, if you want it to source certain publications in your industry, compile a list and upload it to the prompt.
As an additional layer, prompt the model to automatically include any third-party sources after every paragraph to make fact-checking easier on the fly.
SEO prompts
Consider building SEO into the structure of the content from the outset.
Early observations of Google’s AI Mode suggest that clearly structured, well-sourced content is more likely to be referenced in AI-generated results.
With that in mind, you can put together a prompt checklist that includes:
Crafting a direct answer in the first one to two sentences, then expanding with context.
Covering the main question, but also potential subquestions (“fan-out” queries) that the system may generate (for example, questions related to comparisons, pros/cons, alternatives, etc.).
Chunking content into many subsections, with each subsection answering a potential fan-out query to completion.
Being an expert source of information in each individual section of the page, meaning it’s a passage that can stand on its own.
Provide clear citations and semantic richness (synonyms, related entities) throughout.
A custom GPT is a personalized version of ChatGPT that’s trained on your materials so it can better create in your brand voice and follow brand rules.
It mostly remembers tone and format, but that doesn’t guarantee the accuracy of output beyond what’s uploaded.
Some companies are exploring RAG (retrieval-augmented generation) to further train LLMs on the company’s own knowledge base.
RAG connects an LLM to a private knowledge base, retrieving relevant documents at query time so the model can ground its responses in approved information.
While custom GPTs are easy, no-code setups, RAG implementation is more technical – but there are companies/technologies out there that can make it easier to implement.
That’s why GPTs tend to work best for small or medium-scale projects or for non-technical teams focused on maintaining brand consistency.
Create a custom GPT in ChatGPT
RAG, on the other hand, is an option for enterprise-level content generation in industries where accuracy is critical and information changes frequently.
Run an automated self-review
Create parameters so the model can self-assess the content before further editorial review. You can create a checklist of things to prompt it.
For example:
“Is the advice helpful, original, people-first?” (Perhaps using Google’s list of questions from its helpful content guidance.)
“Is the tone and voice completely aligned with the style guide?”
Have an established editing process
Even the best AI workflow still depends on trained editors and fact-checkers. This human layer of quality assurance protects accuracy, tone, and credibility.
Writers and editors need to continue to upskill in the coming year, and, according to the Microsoft 2025 annual Work Trend Index, AI skilling is the top priority.
Source: 2025 Microsoft Work Trend Index Annual Report
Professional training creates baseline knowledge so your team gets up to speed faster and can confidently handle outputs consistently.
This includes training on how to effectively use LLMs and how to best create and edit AI content.
In addition, training content teams on SEO helps them build best practices into prompts and drafts.
Editorial procedures
Ground your AI-assisted content creation in editorial best practices to ensure the highest quality.
This might include:
Identifying the parts of the content creation workflow that are best suited for LLM assistance.
Conducting an editorial meeting to sign off on topics and outlines.
Drafting the content.
Performing the structural edit for clarity and flow, then copyediting for grammar and punctuation.
Getting sign-off from stakeholders.
AI editorial process
The AI editing checklist
Build a checklist to use during the review process for quality assurance. Here are some ideas to get you started:
Every claim, statistic, quote, or date is accompanied by a citation for fact-checking accuracy.
All facts are traceable to credible, approved sources.
Outdated statistics (more than two years) are replaced with fresh insights.
Draft meets the style guide’s voice guidelines and tone definitions.
Content adds valuable, expert insights rather than being vague or generic.
For thought leadership, ensure the author’s perspective is woven throughout.
Draft is run through the AI detector, aiming for a conservative percentage of 5% or less AI.
Draft aligns with brand values and meets internal publication standards.
Final draft includes explicit disclosure of AI involvement when required (client-facing/regulatory).
Grounding AI content in trust and intent
AI is transforming how we create, but it doesn’t change why we create.
Every policy, workflow, and prompt should ultimately support one mission: to deliver accurate, helpful, and human-centered content that strengthens your brand’s authority and improves your visibility in search.
Many PPC advertisers obsess over click-through rates, using them as a quick measure of ad performance.
But CTR alone doesn’t tell the whole story – what matters most is what happens after the click. That’s where many campaigns go wrong.
The problem with chasing high CTRs
Most advertisers think the ad with the highest CTR is often the best. It should have a high Quality Score and attract lots of clicks.
However, in most cases, lower CTR ads usually outperform higher CTR ads in terms of total conversions and revenue.
If all I cared about was CTR, then I could write an ad:
“Free money.”
“Claim your free money today.”
“No strings attached.”
That ad would get an impressive CTR for many keywords, and I’d go out of business pretty quickly, giving away free money.
When creating ads, we must consider:
Type of searchers we want to attract.
Ensure the users are qualified.
Set expectations for the landing page.
I can take my free money ad and refine it:
“Claim your free money.”
“Explore college scholarships.”
“Download your free guide.”
I’ve now:
Told searchers they can get free money for college through scholarships if they download a guide.
Narrowed down my audience to people who are willing to apply for scholarships and willing to download a guide, presumably in exchange for some information.
If you focus solely on CTR and don’t consider attracting the right audience, your advertising will suffer.
While this sentiment applies to both B2C and B2B companies, B2B companies must be exceptionally aware of how their ads appear to consumers versus business searchers.
B2B companies must pre-qualify searchers
If you are advertising for a B2B company, you’ll often notice that CTR and conversion rates have an inverse relationship. As CTR increases, conversion rates decrease.
The most common reason for this phenomenon is that consumers and businesses can search for many B2B keywords.
B2B companies must try to show that their products are for businesses, not consumers.
For instance, “safety gates”is a common search term.
The majority of people looking to buy a safety gate are consumers who want to keep pets or babies out of rooms or away from stairs.
However, safety gates and railings are important for businesses with factories, plants, or industrial sites.
These two ads are both for companies that sell safety gates. The first ad’s headlines for Uline could be for a consumer or a business.
It’s not until you look at the description that you realize this is for mezzanines and catwalks, which is something consumers don’t have in their homes.
As many searchers do not read descriptions, this ad will attract both B2B and B2C searchers.
The second ad mentions Industrial in the headline and follows that up with a mention of OSHA compliance in the description and the sitelinks.
While both ads promote similar products, the second one will achieve a better conversion rate because it speaks to a single audience.
We have a client who specializes in factory parts, and when we graph their conversion rates by Quality Score, we can see that as their Quality Score increases, their conversion rates decrease.
They will review their keywords and ads whenever they have a 5+ Quality Score on any B2B or B2C terms.
This same logic does not apply to B2B search terms.
Those terms often contain more jargon or qualifying statements when looking for B2B services and products.
B2B advertisers don’t have to use characters to weed out B2C consumers and can focus their ads only on B2B searchers.
How to balance CTR and conversion rates
As you are testing various ads to find your best pre-qualifying statements, it can be tricky to examine the metrics. Which one of these would be your best ad?
15% CTR, 3% conversion rate.
10% CT, 7% conversion rate.
5% CTR, 11% conversion rate.
When examining mixed metrics, CTR and conversion rates, we can use additional metrics to define our best ads. My favorite two are:
Conversion per impression (CPI): This is a simple formula dividing your conversion by the number of impressions (conversions/impressions).
Revenue per impression (RPI): If you have variable checkout amounts, you can instead use your revenue metrics to decide your best ads by dividing your revenue by your impressions (revenue/impressions).
You can also multiply the results by 1,000 to make the numbers easier to digest instead of working with many decimal points. So, we might write:
CPI = (conversions/impressions) x 1,000
By using impression metrics, you can find the opportunity for a given set of impressions.
CTR
Conversion rate
Impressions
Clicks
Conversions
CPI
15%
3%
5,000
750
22.5
4.5
10%
7%
4,000
400
28
7
5%
11%
4,500
225
24.75
5.5
By doing some simple math, we can see that option 2, with a 10% CTR and a 7% conversion rate, gives us the most total conversions.
How do you dissuade users who don’t fit your audience from clicking on your ads?
How do you attract your qualified audience?
Are your ads setting proper landing page expectations?
By considering each of these questions as you create ads, you can find ads that speak to the type of users you want to attract to your site.
These ads are rarely your best CTRs. These ads balance the appeal of high CTRs with pre-qualifying statements that ensure the clicks you receive have the potential to turn into your next customer.
The web’s purpose is shifting. Once a link graph – a network of pages for users and crawlers to navigate – it’s rapidly becoming a queryable knowledge graph.
For technical SEOs, that means the goal has evolved from optimizing for clicks to optimizing for visibility and even direct machine interaction.
Enter NLWeb – Microsoft’s open-source bridge to the agentic web
At the forefront of this evolution is NLWeb (Natural Language Web), an open-source project developed by Microsoft.
NLWeb simplifies the creation of natural language interfaces for any website, allowing publishers to transform existing sites into AI-powered applications where users and intelligent agents can query content conversationally – much like interacting with an AI assistant.
Developers suggest NLWeb could play a role similar to HTML in the emerging agentic web.
Its open-source, standards-based design makes it technology-agnostic, ensuring compatibility across vendors and large language models (LLMs).
This positions NLWeb as a foundational framework for long-term digital visibility.
Schema.org is your knowledge API: Why data quality is the NLWeb foundation
NLWeb proves that structured data isn’t just an SEO best practice for rich results – it’s the foundation of AI readiness.
Its architecture is designed to convert a site’s existing structured data into a semantic, actionable interface for AI systems.
In the age of NLWeb, a website is no longer just a destination. It’s a source of information that AI agents can query programmatically.
The NLWeb data pipeline
The technical requirements confirm that a high-quality schema.org implementation is the primary key to entry.
Data ingestion and format
The NLWeb toolkit begins by crawling the site and extracting the schema markup.
The schema.org JSON-LD format is the preferred and most effective input for the system.
This means the protocol consumes every detail, relationship, and property defined in your schema, from product types to organization entities.
For any data not in JSON-LD, such as RSS feeds, NLWeb is engineered to convert it into schema.org types for effective use.
Semantic storage
Once collected, this structured data is stored in a vector database. This element is critical because it moves the interaction beyond traditional keyword matching.
Vector databases represent text as mathematical vectors, allowing the AI to search based on semantic similarity and meaning.
For example, the system can understand that a query using the term “structured data” is conceptually the same as content marked up with “schema markup.”
This capacity for conceptual understanding is absolutely essential for enabling authentic conversational functionality.
Every NLWeb instance operates as an MCP server, an emerging standard for packaging and consistently exchanging data between various AI systems and agents.
MCP is currently the most promising path forward for ensuring interoperability in the highly fragmented AI ecosystem.
The ultimate test of schema quality
Since NLWeb relies entirely on crawling and extracting schema markup, the precision, completeness, and interconnectedness of your site’s content knowledge graph determine success.
The key challenge for SEO teams is addressing technical debt.
Custom, in-house solutions to manage AI ingestion are often high-cost, slow to adopt, and create systems that are difficult to scale or incompatible with future standards like MCP.
NLWeb addresses the protocol’s complexity, but it cannot fix faulty data.
If your structured data is poorly maintained, inaccurate, or missing critical entity relationships, the resulting vector database will store flawed semantic information.
This leads inevitably to suboptimal outputs, potentially resulting in inaccurate conversational responses or “hallucinations” by the AI interface.
Robust, entity-first schema optimization is no longer just a way to win a rich result; it is the fundamental barrier to entry for the agentic web.
By leveraging the structured data you already have, NLWeb allows you to unlock new value without starting from scratch, thereby future-proofing your digital strategy.
NLWeb vs. llms.txt: Protocol for action vs. static guidance
The need for AI crawlers to process web content efficiently has led to multiple proposed standards.
A comparison between NLWeb and the proposed llms.txt file illustrates a clear divergence between dynamic interaction and passive guidance.
The llms.txt file is a proposed static standard designed to improve the efficiency of AI crawlers by:
Providing a curated, prioritized list of a website’s most important content – typically formatted in markdown.
Attempting to solve the legitimate technical problems of complex, JavaScript-loaded websites and the inherent limitations of an LLM’s context window.
In sharp contrast, NLWeb is a dynamic protocol that establishes a conversational API endpoint.
Its purpose is not just to point to content, but to actively receive natural language queries, process the site’s knowledge graph, and return structured JSON responses using schema.org.
NLWeb fundamentally changes the relationship from “AI reads the site” to “AI queries the site.”
Attribute
NLWeb
llms.txt
Primary goal
Enables dynamic, conversational interaction and structured data output
Improves crawler efficiency and guides static content ingestion
Operational model
API/Protocol (active endpoint)
Static Text File (passive guidance)
Data format used
Schema.org JSON-LD
Markdown
Adoption status
Open project; connectors available for major LLMs, including Gemini, OpenAI, and Anthropic
Proposed standard; not adopted by Google, OpenAI, or other major LLMs
Strategic advantage
Unlocks existing schema investment for transactional AI uses, future-proofing content
Reduces computational cost for LLM training/crawling
The market’s preference for dynamic utility is clear. Despite addressing a real technical challenge for crawlers, llms.txt has failed to gain traction so far.
NLWeb’s functional superiority stems from its ability to enable richer, transactional AI interactions.
It allows AI agents to dynamically reason about and execute complex data queries using structured schema output.
The strategic imperative: Mandating a high-quality schema audit
While NLWeb is still an emerging open standard, its value is clear.
It maximizes the utility and discoverability of specialized content that often sits deep in archives or databases.
This value is realized through operational efficiency and stronger brand authority, rather than immediate traffic metrics.
Several organizations are already exploring how NLWeb could let users ask complex questions and receive intelligent answers that synthesize information from multiple resources – something traditional search struggles to deliver.
The ROI comes from reducing user friction and reinforcing the brand as an authoritative, queryable knowledge source.
For website owners and digital marketing professionals, the path forward is undeniable: mandate an entity-first schema audit.
Because NLWeb depends on schema markup, technical SEO teams must prioritize auditing existing JSON-LD for integrity, completeness, and interconnectedness.
Publishers should ensure their schema accurately reflects the relationships among all entities, products, services, locations, and personnel to provide the context necessary for precise semantic querying.
The transition to the agentic web is already underway, and NLWeb offers the most viable open-source path to long-term visibility and utility.
It’s a strategic necessity to ensure your organization can communicate effectively as AI agents and LLMs begin integrating conversational protocols for third-party content interaction.
With the yen freely convertible and backed by Japan’s deep government bond market, JPYC’s launch stands apart from the region’s onshore-only experiments in Korea, Taiwan, and beyond.
Samsung is reportedly preparing for One UI 8.5, which could debut alongside the Galaxy S26 series early next year. However, recent reports suggest the company might be running late with the Galaxy S26 launch, possibly pushing the event beyond January 2026.
The delay appears to be connected to Samsung’s change in its phone lineup. Earlier rumors said the regular Galaxy S26 might be called “Pro” and a slim “Edge” model would replace the “Plus”.
Now, those plans are reportedly canceled. Samsung is going back to the familiar lineup – Galaxy S26, Galaxy S26 Plus, and Galaxy S26 Ultra. The Plus model is back, while the Edge and Pro names are gone.
This could also affect One UI 8.5. Since the Galaxy S26 Plus development is running late, the release of One UI 8.5 Beta may also be delayed. If the Galaxy Unpacked event is postponed to late February or early March, users will also have to wait longer to get another major update.
Phones in picture – Galaxy S25 Ultra, Plus and vanilla
However, the One UI 8.5 Beta program might start in late November, which gives users an early look at new features. But if the phone launch is postponed, the beta could run for several months before the official release, which may feel long for eager users. Or Samsung might delay One UI 8.5 Beta Program.
Despite these delays, the changes could be beneficial. Samsung seems focused on improving hardware and software, with upgrades expected in performance and camera capabilities with the next series. Going back to a simple naming system also makes it easier for people to understand the lineup.
While fans might be disappointed by the delay, it could mean a more polished experience when new phones and software finally launch. Samsung has not confirmed any dates yet, so users will have to wait for official announcements.
The return of Galaxy S26 Plus and the lineup reshuffle may push back the One UI 8.5 beta, but it could result in better phones and a smoother software update for users in 2026. Stay tuned.
If you’re paying attention, you’ll notice the warning signs: click-through rate (CTR) slips, engagement falls, and cost-per-click (CPC) creeps up.
If you’re not, one day your former top performer is suddenly costing you money.
Creative fatigue – the decline in ad performance caused by overexposure or audience saturation – is often the culprit.
It’s been around as long as advertising itself, but in an era where platforms control targeting, bidding, and even creative testing, it’s become one of the few variables marketers can still influence.
This article explains how to spot early signs of fatigue across PPC platforms before your ROI turns sour, and how to manually refresh your creative in the age of AI-driven optimization.
We’ll look at four key factors:
Ad quality.
Creative lifecycle.
Audience saturation.
Platform dynamics.
1. Ad quality
Low-quality ads burn out much faster than high-quality ones.
To stand the test of time, your creative needs to be both relevant and resonant – it has to connect with the viewer.
But it’s important to remember that creative fatigue isn’t the same as bad creative. Even a brilliant ad will wear out if it’s shown too often or for too long.
Think of it like a joke – no matter how good it is, it stops landing once the audience has heard it a dozen times.
The data behind ad quality
To track ad quality, monitor how your key metrics trend over time – especially CTR, CPC, and conversion rate (CVR).
A high initial CTR followed by a gradual decline usually signals a strong performer reaching the end of its natural run.
Because every campaign operates in a different context, it’s best to compare an ad’s results against your own historical benchmarks rather than rigid KPI targets.
Factor in elements like seasonality and placement to avoid overgeneralizing performance trends.
And to read the data accurately, make sure you’re analyzing results by creative ID, not just by campaign or ad set.
Every ad has a natural lifespan – and every platform its own life expectancy.
No matter how timely or novel your ad was at launch, your audience will eventually acclimate to its visuals or message.
Keeping your creative fresh helps reset the clock on fatigue.
Refreshing doesn’t have to mean reinventing.
Sometimes a new headline, a different opening shot, or an updated call to action is enough to restore performance. (See the table below for rule-of-thumb refresh guidelines by platform.)
The data behind creative lifecycle
To distinguish a normal lifecycle from an accelerated one that signals deeper issues, track declining performance metrics like CTR and frequency – how many times a user sees your ad.
A high-performing ad typically follows a predictable curve.
Engagement drops about 20-30% week over week as it nears the end of its run. Any faster, and something else needs fixing.
Your refresh rate should also match your spend. Bigger budgets drive higher frequency, which naturally shortens a creative’s lifespan.
You’ve got your “cool ad” – engaging visuals, a catchy hook, and a refresh cadence all mapped out.
You put a big budget behind it, only to watch performance drop like a stone after a single day. Ouch.
You’re likely running into the third factor of creative fatigue: audience saturation – when the same people see your ad again and again, driving performance steadily downward.
Failing to balance budget and audience size leads even the strongest creative to overexposure and a shorter lifespan.
The data behind audience saturation
To spot early signs of saturation, track frequency, and reach together.
Frequency measures how many times each person sees your ad, while reach counts the number of unique people who’ve seen it.
When frequency rises but reach plateaus, your ad hits the same people repeatedly instead of expanding to new audiences.
Ideally, both numbers should climb in tandem.
Some platforms – including Google, Microsoft, LinkedIn, and DSP providers – offer frequency caps to control exposure.
These days, algorithms don’t just reflect performance – they shape it.
Once an ad starts to underperform, a feedback loop kicks in.
Automated systems reduce delivery, which further hurts performance, which leads to even less delivery.
How each platform evaluates creative health – and how quickly you respond before your ad is demoted – is the fourth and final factor in understanding creative fatigue.
The data behind platform dynamics
Every platform has its own system for grading creative performance, but the clearest sign of algorithmic demotion is declining impressions or spend despite stable budgets and targeting.
The tricky part is that this kind of underdelivery can look a lot like normal lifecycle decline or audience saturation. In reality, it’s often a machine-level penalty.
To spot it, monitor impression share and spend velocity week over week, at the creative level (not by campaign or ad set).
What to do when the algorithm punishes you
When impressions or spend drop despite a stable budget and consistent targeting, your ad has likely been demoted by the platform.
That doesn’t necessarily mean it’s poor quality.
This usually means the algorithm has lost “confidence” in its ability to achieve your chosen goal, such as engagement or conversions.
Here’s how to recover:
Check your performance metrics: Sharp declines in CTR, engagement, or conversions can trigger a penalty. Compare the trend line to earlier in the campaign.
Assess audience saturation: If frequency exceeds 3 for prospecting or 5 for retargeting, your audience may be too small for the budget. Broaden targeting or reduce spend.
Refresh the creative: Launch new or updated versions under new ad IDs so the system re-enters its learning phase.
Don’t make drastic edits: Frequent budget, bid, or targeting changes reset learning and slow recovery.
When the algorithm cools your ad, don’t panic.
Act quickly to identify whether the issue lies in quality, freshness, audience, or budget – and make deliberate adjustments, not hasty ones.
Turning creative fatigue into a performance signal
Creative fatigue, like death and taxes, is inevitable. Every ad has a beginning, middle, and end.
The key is recognizing those stages early through vigilant data monitoring, so you can extend performance instead of waiting for the crash.
While automation may be taking over much of marketing, ad creative, and copy remain one arena where humans still outperform machines.
Great marketers today don’t just make good ads. They know how to sustain them through smart refreshes, rotations, and timely retirements.
Because when you can see the whimper coming, you can make sure your next ad lands with a bang.
Q4 is here – and for ecommerce brands, that means the biggest sales opportunities of the year are just ahead.
Black Friday, Cyber Monday, Christmas – the biggest sales events are just around the corner. To hit your targets, preparation is key. It’s not too late to act, and the opportunities ahead are huge.
Use this checklist to get up to speed quickly and set your account up for success.
Website and UX
Review site speed
Start with a website audit to identify any red flags. Tools like PageSpeed Insights can help diagnose technical issues.
Encourage clients to review key pages and the checkout process on multiple devices to ensure there are no bottlenecks.
If resources allow, use heatmap or session analysis tools such as Microsoft Clarity or Hotjar to better understand user behavior and improve the on-site experience.
Confirm tracking setup
Double-check that all tracking is configured correctly across platforms.
Don’t just verify that tags are firing – make sure all events are set up to their fullest potential.
For example, confirm high match rates in Meta and ensure Enhanced Conversions is fully configured.
Add VIP sign-ups/pop-ups
Before the sales period begins, encourage users to join a VIP list for Black Friday or holiday promotions.
This can give them early access or exclusive deals. Set up a separate automated email flow to follow up with these subscribers.
Launch sale page early
Publish your sale page as soon as possible so Google can crawl and index it for SEO.
The page doesn’t need to be accessible from your site navigation or populated with products right away – the key is to get it live early.
If possible, reuse the same URL from previous years to build on existing SEO equity.
You can also add a data capture form to collect VIP sign-ups until the page goes live with products.
Display cutoffs clearly
If shipping cutoff dates aren’t clear, many users won’t risk placing an order close to the deadline.
Clearly display both standard and express delivery cutoff dates on your website.
Highlight sales sitewide with banners
Don’t rely solely on a homepage carousel to promote your sale.
Add a banner or header across all pages so users know a sale is happening, no matter where they land.
As mentioned with pop-ups, supplementing that strategy with lead generation ads can help grow your email list and build early buzz around your upcoming sale.
Launch simple, clear primary sale ads
These will be your Black Friday or holiday sale ads running for most of the campaign.
Keep the messaging and promotion straightforward. Any confusion in a crowded feed will make users scroll past.
Use strong branding, put the offer front and center, and include a clear CTA. On Meta, this often works best as a simple image ad.
Create Cyber Monday-specific ads
Many brands simply extend their Black Friday sale rather than creating Cyber Monday-specific ads and web banners.
Take advantage of the opportunity to give your campaign a fresh angle – both in messaging and offer.
Since it’s often the final day of your sale, you can go bigger on discounts for one day or add a free gift with purchases over a certain amount.
It’s also a great way to move slower-selling inventory left over from Black Friday.
Refresh primary ads with ‘last days’ urgency
Add urgency to your messaging as the sale nears its end by including countdowns or end dates.
This tactic works especially well for longer campaigns where ad fatigue can set in.
Finalize all creative assets early
November and December are busy months for ad builds and platform reviews.
Make sure all sale assets are ready several weeks before launch to avoid rushed builds and delays from longer approval times.
Advertising and data
Audit product feeds
Make sure item disapprovals and limited products are kept to a minimum. Double-check that your setup is current.
For example, if your return window has changed, update that information in Google Merchant Center.
Refresh first-party data and remarketing lists
Update any lists you plan to use this season.
If you don’t have direct integrations, upload new or revised lists manually.
Review your integrations and confirm that data is flowing correctly.
Build lookalike and custom audiences early
Start building audiences as soon as your first-party and remarketing lists are refreshed.
Create Meta Lookalike Audiences, Performance Max audience signals, and Custom Audiences.
If you run into volume issues, you’ll have time to adjust or explore alternatives.
Finalize budget by week, not just month
Agree on budgets early so you know your spending limits. Don’t plan just by month. Map out weekly spend, too.
You’ll likely want to invest more heavily in the final week of November than in the first.
Use title and description extensions or ad customizers
Updating search ad copy can be tedious and time-consuming.
These tools let you control and update copy dynamically without editing every RSA manually – saving hours in campaign builds.
Use ad assets, promo sitelinks, and GMC promotions
Enable sale-related sitelinks, callouts, and promotion extensions across search campaigns so your offers appear everywhere.
In Shopping, set up Google Merchant Center promotions to highlight deals and incentives in your Shopping ad annotations.
Apply countdown features
Add a dynamic countdown timer to search ads to show exactly when your sale ends.
This feature helps your ads stand out and adds urgency as the sale nears its close.
Launch search remarketing activity
Bid on generic keywords you wouldn’t normally target, but limit them to remarketing or first-party data audiences.
For example, people searching for “Black Friday deals” who have purchased from your site in the past 30 days already know your brand and are primed to buy again.
Apply seasonality adjustments
If you use Google Ads or Microsoft Ads with a target ROAS strategy, apply seasonality adjustments to prepare the algorithm for higher conversion rates during the sale period.
Remember to apply a negative adjustment once the sale ends to prevent unnecessary spend spikes.
Not every tactic will fit your business or resources – and that’s OK.
The key is to focus on what will have the biggest impact on your store.
By addressing most of the points in this checklist, you’ll build a solid foundation for a strong Q4 and set yourself up to capture more sales during the busiest shopping season of the year.
Preparation is everything. The earlier you audit, test, and launch, the smoother your campaigns will run when traffic – and competition – start to surge.
MiCA has given Europe a uniquely strong position to establish the regulatory gold standard for crypto, says Malta Financial Services Authority CEO Kenneth Farrugia, but regulators must work quickly and collaboratively to preserve the region’s advantage.
In the early days of SEO, ranking algorithms were easy to game with simple tactics that became known as “black hat” SEO – white text on a white background, hidden links, keyword stuffing, and paid link farms.
Early algorithms weren’t sophisticated enough to detect these schemes, and sites that used them often ranked higher.
Today, large language models power the next generation of search, and a new wave of black hat techniques are emerging to manipulate rankings and prompt results for advantage.
The AI content boom – and the temptation to cut corners
Up to 21% of U.S. users access AI tools like ChatGPT, Claude, Gemini, Copilot, Perplexity, and DeepSeek more than 10 times per month, according to SparkToro.
Overall adoption has jumped from 8% in 2023 to 38% in 2025.
It’s no surprise that brands are chasing visibility – especially while standards and best practices are still taking shape.
One clear sign of this shift is the surge in AI-generated content. Graphite.io and Axios report that the share of articles written by AI has now surpassed those created by humans.
Two years ago, Sports Illustrated was caught publishing AI-generated articles under fake writer profiles – a well-intentioned shortcut that backfired.
The move damaged the brand’s credibility without driving additional traffic.
Its authoritativeness, one of the pillars of Google’s E-E-A-T (experience, expertise, authoritativeness, and trustworthiness) framework, was compromised.
While Google continues to emphasize E-E-A-T as the North Star for quality, some brands are testing the limits.
With powerful AI tools now able to execute these tactics faster and at scale, a new wave of black hat practices is emerging.
As black hat GEO gains traction, several distinct tactics are emerging – each designed to exploit how AI models interpret and rank content.
Mass AI-generated spam
LLMs are being used to automatically produce thousands of low-quality, keyword-stuffed articles, blog posts, or entire websites – often to build private blog networks (PBNs).
The goal is sheer volume, which artificially boosts link authority and keyword rankings without human oversight or original insight.
Fake E-E-A-T signals
Search engines still prioritize experience, expertise, authoritativeness, and trustworthiness.
Black hat GEO now fabricates these signals using AI to:
Create synthetic author personas with generated headshots and fake credentials.
Mass-produce fake reviews and testimonials.
Generate content that appears comprehensive but lacks genuine, human-validated experience.
LLM cloaking and manipulation
A more advanced form of cloaking, this tactic serves one version of content to AI crawlers – packed with hidden prompts, keywords, or deceptive schema markup – and another to human users.
The goal is to trick the AI into citing or ranking the content more prominently.
Schema misuse for AI Overviews
Structured data helps AI understand context, but black hat users can inject misleading or irrelevant schema to misrepresent the page’s true purpose, forcing it into AI-generated answers or rich snippets for unrelated, high-value searches.
SERP poisoning with misinformation
AI can quickly generate high volumes of misleading or harmful content targeting competitor brands or industry terms.
The aim is to damage reputations, manipulate rankings, and push legitimate content down in search results.
Even Google surfaces YouTube videos that explain how these tactics work. But just because they’re easy to find doesn’t mean they’re worth trying.
The risks of engaging in – or being targeted by – black hat GEO are significant and far-reaching, threatening a brand’s visibility, revenue, and reputation.
Severe search engine penalties
Search engines like Google are deploying increasingly advanced AI-powered detection systems (such as SpamBrain) to identify and penalize these tactics.
De-indexing: The most severe penalty is the complete removal of your website from search results, making you invisible to organic traffic.
Manual actions: Human reviewers can issue manual penalties that lead to a sudden and drastic drop in rankings, requiring months of costly, intensive work to recover.
Algorithmic downgrading: The site’s ranking for targeted keywords can be significantly suppressed, leading to a massive loss of traffic and potential customers.
Reputation and trust damage
Black hat tactics inherently prioritize manipulation over user value, leading to poor user experience, spammy content, and deceptive practices.
Loss of credibility: When users encounter irrelevant, incoherent, or keyword-stuffed content – or find that an AI-cited answer is baseless – it damages the perception of the brand’s expertise and honesty.
Erosion of E-E-A-T: Since AI relies on E-E-A-T signals for authoritative responses, being caught fabricating these signals can permanently erode the brand’s trustworthiness in the eyes of the algorithm and the public.
Malware distribution: In some extreme cases, cybercriminals use black hat SEO to poison search results, redirecting users to sites that install malware or exploit user data. If a brand’s site is compromised and used for such purposes, the damage is catastrophic.
AI changes the game – not the rules
The growth of AI-driven platforms is remarkable – but history tends to repeat itself.
Black hat SEO in the age of LLMs is no different.
While the tools have evolved, the principle remains the same: best practices win.
Google has made that clear, and brands that stay focused on quality and authenticity will continue to rise above the noise.
Login
