Search is changing. Fast. And not just because of AI Overviews or LLMS’ answers.

We’re witnessing a shift in how people make meaning online. Queries have stopped being simple keywords: they’re now liquid, personal, and shaped by context, history, and culture.

To understand this shift, we can borrow tools from a surprising place: semiotics, the study of signs and meaning.
Think Umberto Eco meets user intent.

In this post, I will explain how semiotics can help us rethink search behaviour, AI interfaces, and the way users navigate what Google calls the “Messy Middle.”
Whether we’re optimising content or steering growth strategy, these insights will help us move beyond rankings and toward relevance, conversions, and brand impact.

What semiotics brings to the table

Semiotics is the study of how meaning is created. Every word, image, or gesture is a sign, and a sign has two parts:

• The signifier (the form it takes: a word, image, symbol).
• The signified (the concept it refers to).

In Search, the user types (or says) a signifier. However, what they mean by it isn’t fixed. It’s fluid. Shaped by context, behaviour, device, time, and even mood.

Example: The “light company” query in Spain usually means “cheap electricity provider.” In the US, it could refer to an LED supplier or a minimal-brand design firm: same signifier, different signifieds.

SEO takeaway: Don’t optimise for the literal string. Map your content to the multiple meanings behind a query; use case studies, comparisons, and explainers to cover different user interpretations.

Umberto Eco’s “Open Work” and the fragmented web

 

Umberto Eco described art and literature as “open works”: things that don’t contain a single meaning but invite interpretation. This is how modern search also works.

Every query is a personal construction of meaning. Search results, especially those powered by AI, are assembled on the fly based on what the machine thinks the user means right now (and this is how Google wants AI Mode to work).

Example: A tourist searching “best breakfast in Valencia” gets filtered Google Maps results, review-driven listicles, and possibly TikTok embeds. A local with a fitness tracker might see healthy cafés that open early (note: did you ever read what other apps and data Google and Chatgpt have access to when installed on your phone? No? You should).

Brand takeaway: The more flexible your content is, the more often it will be reassembled into someone’s search experience.

Visibility doesn’t come from ranking #1; it comes from showing up in as many personalised SERPS as it is possible.

Actionable idea: Use modular content blocks that can stand alone: comparison tables, semantic snippets, definitions. Make your content fragmentable and remixable by AI systems.

The Messy Middle is a semiotic loop, but it starts and ends elsewhere

 

Google’s “Messy Middle” model describes the non-linear path users take when making decisions. Most interpretations focus on the zig-zag between exploration and evaluation, but two other moments are just as critical:

• The trigger, when the journey begins.
• The experience, where it ends.

If you’re ignoring either of those, you’re losing visibility at the top and conversions at the bottom.

Let’s see each phase semiotically and strategically.

Trigger: The moment meaning sparks

 

This is where everything starts. It’s not a query yet. It’s a need, a stimulus, an exposure to a concept or brand.

In semiotics, this is the pre-conscious phase: the emergence of meaning.

Examples:

• Seeing a TikTok about someone’s home energy savings.
• Hearing a podcast mention of a new electric car model.
• Glimpsing a brand in someone else’s buying journey.
• Reading an article on Discover.

This is where brand presence, associations, and signs matter most.

Strategic levers:

• Branded SEO (own your brand SERP, control the narrative).
• Digital PR (appear on high-authority, audience-relevant publications).
• Zero-click content (guides, definitions, stats that show up without requiring a click).
• Entity building (make your brand part of known topical clusters in Google’s graph).
• Collaborations with influencers, experts, or websites are already shaping user meaning.

If SEO starts here, then strategy must too. Your job is to place the right signs in the right minds, and before they type anything.

Exploration: meaning in motion

 

Once the need is triggered, users enter the exploration phase. They don’t know what they want yet, but they’re mapping the terrain.

Examples:

• “Is solar power worth it?”
• “Best electric vehicles for city driving”
• “How do energy ratings work?”

Here, users are decoding new terms, models, and categories. Semiotic friction is high, and every unfamiliar word is a potential dropout point.

Strategic levers:

• Glossary-style content, “X explained,” and guides.
• Structured formats (FAQS, PAA targeting).
• Internal linking that reflects the mental map users are building.
• Language mirroring: use the same vocabulary that your users already trust.

Evaluation: weighing the signs

 

Now, users are comparing options. Meaning stabilises. They’ve defined the question, and now they want the best answer.

Examples:

• “Tesla Model 3 vs BYD Dolphin”
• “Top 5 energy providers in Spain”
• “Best time of year to install solar panels”

This is where semiotics meets economics: symbols of trust, authority, and logic are all in play.

Strategic levers:

• Comparison content, reviews, and rankings.
• Brand signals (logos, awards, expert quotes).
• Schema for review snippets, product features, and pricing.
• CRO elements that reinforce credibility (UX writing, visuals, guarantees).

Experience: The final interpretant

 

The journey ends not with a click but with what happens after.

This is the landing page/PDP, the UX, the form field that won’t load.

It’s where meaning either converts or collapses.

Examples:

• Page loads in 6 seconds → bounce
• Page feels easy to scan and persuasive → conversion
• Content reflects the exact query they asked → trust is reinforced

This is the semiotic closure. If your site doesn’t confirm the meaning the user came in with, you break the journey.

Strategic levers:

• Page Experience (Core Web Vitals, mobile UX, visual stability).
• Message consistency (title tag → headline → CTA flow).
• Technical SEO hygiene and interaction design that reflects user intent (clear navigation, filters, CTAs).

Micro-Moments: intent archetypes revealed by the SERP

 

Micro-moments are compressed semiotic units, aka tiny expressions of intent, urgency, and context. And the most important clues lie not in the query, but in the SERP layout.

Let’s break them down properly:

• I want to go
  • To a place

SERP clues: Local Pack, Local Knowledge panel, Top Sights/Things to Do, Hotels…
Content match: Location pages, accurate Google Business page, reviews…

• • To a website

SERP clues: Homepage and PLPs ranking, Knowledge Graph, “Explore brands” Merchant features…
Content match: Optimised homepage and PLPs, branded search snippet, fast load speed…

• I want to know
  • Fresh knowledge

SERP clues: Top Stories, timestamps, Twitter/X carousels
Content match: Fast-published content, news angles, blog updates

• • Theoretical knowledge

SERP clues: AI Overview, Featured snippets, definitions, People Also Ask, Things to Know, Images Box, Knowledge Graph.
Content match: Glossaries, E-E-A-T-style explainers, longform articles.

• • Practical knowledge

SERP clues: AI Overview, How-to search results, image box, YouTube embeds and Videos, Discussion and Forums…
Content match: Visual guides, tutorials, structured walkthroughs

• I want to do

SERP clues: AI Overview, How-tos, tools, Shorts, Discussion and Forums…
Content match: Step-by-steps, calculators, checklists

• I want to buy
  • Need help deciding

SERP clues: Review packs, side-by-side comparison carousels, AI Overview, “Best X…” search results…
Content match: UGC, testimonials, “vs” pages, buying guides, landing pages

• Ready to buy

SERP clues: Shopping ads/ads, Popular product Merchant feature, Product Knowledge panel, product rich results with or without product carousel…
Content match: PDPs, “prices” pages, CTAs, delivery info…

Strategic insight: Think of each micro-moment as a distinct search stage with a visible footprint. The SERP itself reveals the meaning structure Google assumes. Follow that structure when building or refining your content.

AI Search is a dialogue, and the SERP is its interface

 

With AI-generated answers, search is no longer a static list of links (albeit if cluttered with Search features). It’s becoming a conversation, and the SERP is the interface where that dialogue begins. Queries are now seed signs and starting points from which meaning can branch in multiple directions.

Your strategic advantage lies in reading the SERP for what it is: a map of journeys yet to be taken. Google’s interface reveals more than results; it shows how meaning is constructed, extended, and reframed.

Here’s how to decode the clues.

Topic Filters (Tabs & Refiners)

Topic filters are Google’s explicit prompts for refining intent or, in other words, semantic pivots based on dominant user journeys.

Example: Search for “Star Wars: Legion” and you might see filters like:

• Fantasy Flight Games.
• Expansión.
• Asmodee.
• Español.
• Reviews.
• et al

These are curated subtopics, not based on keywords alone, but on real behavioural and contextual clustering.

Strategic move: Treat these as topical parent nodes. Build internal linking structures and content hubs that mirror these branches. Match both the terminology and content format.

People Also Ask (PAA)

People Also Ask feature surfaces anticipated follow-up questions and steps in a reasoning process. It simulates dialogue by modelling the clarifications or concerns a real person might express next.

Example: For “Como pintar Star Wars: Legion?” (“how to paint Star Wars: Legion?”) PAA might include:

• ¿Qué pintura se utiliza para pintar miniaturas? (“What kind of paint can be used for miniature painting?”)
• ¿Necesitas imprimar las miniaturas de Star Wars Legion? (“Do you need to prime Star Wars Legion miniatures?)
• ¿Qué escala es Star Wars Legion? (“What scale is Star Wars Legion?”)
• ¿Qué tipo de pintura se usa para pintar juguetes? (“What type of paint do you use to paint toys?”)

Each entry is a micro-dialogue fragment. Clicking one reveals more questions, like a dynamic branching tree of meaning.

Strategic move: Don’t just answer PAA: study their logic. How does the question reposition the topic? Use this to shape your content sections and UX flow.

People Also Search For (Related Searches)

People Also Search For or Related Searches appear at the bottom of the SERP and function as semantic drift indicators. They reflect how the user’s core topic could evolve into something else entirely.

This isn’t deepening; it’s broadening. These are alternative but adjacent interpretations of the original need.

Example: For “How to paint Star Wars: Legion”, related searches may include:

1. How to paint Star Wars Legion black
2. Star Wars: legion painting guide pdf
3. How to paint Star Wars Legion Clone Troopers
4. Star Wars Legion Core Paint Set
5. Star Wars Legion Yoda painting guide
6. Star Wars Legion Rebels painting guide
7. Star Wars Legion Clone paint schemes
8. Star Wars Legion miniatures

These often signal query reformulation paths, aka what users searched next after this didn’t satisfy them, or what people eventually typed after iterating.

Strategic move: Use related searches to structure supporting content and exit paths. They help predict where your user is headed next, before they bounce to someone else’s site.

Things to Know

Things to Know is a still-emerging search feature, but increasingly powerful, this feature distils a topic into structured sub-sections and learning paths.

Example: A “Things to Know” panel for “heat pumps” might include:

1. How they work
2. Types of systems
3. Cost range
4. Common misconceptions

Each heading is a topical contract, aka a promise about how information will be organised.

Strategic move: Match this structure 1:1 in your content. These clusters are machine-validated signals of how people cognitively break down the topic. They are ideal for pillar content architecture.

Image Tags (Google Images)

 

These tags, visible in Google Images, are often misunderstood as purely visual labels. But in reality, they offer structured insight into how people semantically expand a given entity.

If your seed query is a known entity (“heat pump,” “standing desk,” “electric car”), the image tags can reveal:

• Attributes most frequently associated with the entity (function, form, material, use case).
• Related entities that co-occur or are often compared or combined with the seed entity.

Example: For “heat pump,” image tags might include:

1. Air source
2. Mitsubishi
3. Thermostat
4. Installation
5. Outdoor unit

This tells us:

• “Air source” = an attribute (type).
• “Mitsubishi” = a related entity (brand).
• “Installation” = a topical context or adjacent query cluster.
• “Outdoor unit” = a partitive relationship (component of the entity).

In semiotic terms, these tags show how the entity is culturally constructed; what aspects people care about, how it’s broken down, and which other objects it’s linked to in user mental models.

Strategic move: Use image tags to enrich your entity maps and content modelling. They help you:

• Identify common modifiers for faceted navigation or product filters.
• Detect related entities to target in supporting content or comparison pages.
• Structure product or service pages to reflect real-world associations that people expect.
  

You’re not just looking at pictures. You’re analysing semantic proximity and attribute salience in the wild.

Persuasive Writing Is Meaning Alignment, Not Just Copy Tricks

Persuasion in search isn’t about clever headlines or emotional triggers alone. It’s about semantic alignment, which means matching your message to the user’s intent, cognitive frame, and bias structure. In other words, it’s semiotics at work.

Every user arrives with a meaning model already in place: what they think the problem is, how they’ve named it, what kinds of solutions they believe are valid. If your content speaks a different language – linguistically or conceptually – it won’t persuade, no matter how well-written it is.

Semiotics and Persuasion: Key Connections

• Sign systems: The words and visuals you choose are signs. Choose signs that are familiar, trusted, or desirable to your audience’s worldview.
• Codes and conventions: Different industries, cultures, and audiences share unwritten “codes” about how authority, trust, or value are expressed. Persuasive content works when it reinforces those codes, or gently subverts them in favour of something better.
• Mythologies (à la Barthes): People don’t just want products: they want meaning. Electric vehicles aren’t just transport; they’re environmental responsibility, status, and independence. Tap into that level.
  

Example: A landing page for solar panels might say:

“High-efficiency solar for any roof.” (ok, but not efficient).
“Take control of your energy costs: quietly, cleanly, and permanently.” (good!)

The first talks tech. The second reflects the user’s emotional logic.

Search Intent + Biases = Context for Persuasion

When we write for search, we often match the keyword but miss the biases underlying it. Semiotics helps us see them:

Intent Type	Likely Bias	Writing Strategy
“Best X for Y”	Fear of choosing wrong	Show comparisons, transparency, and reviews
“How to…”	Fear of looking incompetent	Use simple language, reassure about effort levels
“X vs Y”	Need to confirm existing belief	Start from both sides, guide toward the truth
“Is X worth it?”	Cost aversion, FOMO	Use framing, future gains, and social proof

 

 

 

 

Writing that responds to intent must also respond to psychological context. A good copy fills knowledge gaps. Great copy also fills emotional ones.

Practical Levers for Meaning-Aligned Writing

• Mirror the user’s vocabulary (use forums, Reddit, and review language as input).
• Use visual semiotics: icons, layout, and structure that reinforce message hierarchy.
• Layer your CTAs by decision stage (“Still deciding?”, “Need a quick quote?”, “Ready to compare options?”).
• Use analogy and metaphor to reframe complex ideas into culturally resonant terms.
• A/B test not just the CTA, but the semiotic framing of your value proposition.
  

What This Means for Teams

Your UX and technical writers and your SEO lead shouldn’t work in silos. Your content strategist should know your top 5 audience objections by heart. And your writers should study how your audience talks about your topic, not just what they search for.

Writing that persuades in the age of AI search isn’t louder. It’s clearer, closer, and more attuned to meaning.

E-E-A-T is a Semiotic Framework

E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) is often misunderstood as a checklist of ranking factors. It isn’t.

Google doesn’t score “E-E-A-T” directly through algorithmic signals. Instead, it uses E-E-A-T as a semiotic framework: a way to describe what quality content looks like in terms of how it communicates meaning, trust, and relevance.

From a semiotic perspective, E-E-A-T helps Google evaluate:

• How meaning is constructed (Is the information grounded in real experience?).
• How signs of expertise are encoded (Does the content use language, structure, and references that signal credibility?).
• How authority is inferred (Who is speaking? Are they cited by others?).
• How trust is visually and linguistically reinforced (Clear design, sourcing, intent).

Google’s systems don’t “measure” E-E-A-T directly, but they look for signs of it. These signs are cultural, contextual, and interpretive. That’s semiotics.

E-E-A-T is not a metric; it’s a model for how good content should mean what it says and promises.

AI and Semiotics: Symbolic AI

 

Symbolic AI (also called Good Old-Fashioned AI or GOFAI) is based on explicit rules, symbols, logic trees, ontologies, and if-then reasoning. It contrasts with statistical AI (like large language models), which is based on pattern recognition, probabilities, and training data.

Symbolic AI is about structured knowledge and reasoning, not just correlations.

How Symbolic AI connects to Search and SEO

Google’s Knowledge Graph is Symbolic AI
The Knowledge Graph is one of the most prominent real-world examples of Symbolic AI at scale. It maps entities and their relationships using:

• Ontologies
• Triples (subject–predicate–object)
• Structured rules and schema

This powers features like:

• Knowledge Panels
• Topic Filters
• “Things to Know”
• Entity disambiguation in AI Overviews
  

SEO that focuses on entities, structured data, and semantic relationships is essentially working within a symbolic system.

You’re optimizing not just for content, but for how your content fits into a machine-readable logic of meaning.

Symbolic AI bridges LLMs and factual grounding

LLMs hallucinate. Symbolic systems don’t, and they rely on defined knowledge structures.

The future of search (especially AI Overviews and retrieval-augmented generation) likely hinges on a hybrid model:

• LLMs for language, pattern recognition, and conversation.
• Symbolic systems for facts, logic, context, and disambiguation.

This means SEO strategies that treat content as discrete signs in a symbolic system – using schema.org, Wikidata, internal linking hierarchies, and concept-based content modeling – will perform better in these hybrid environments.

Semiotics and Symbolic AI work together

Your semiotic framework already interprets how meaning is made and transformed through signs, user behavior, and SERP design.

Symbolic AI is how machines try to encode and reason about those signs.
Google attempts to model understanding in a rule-based form.

So while LLMs offer statistical plausibility, Symbolic AI supports semantic precision. In the context of SEO:

• LLMs explain intent dynamically.
• Symbolic structures anchor it in a verified context.

Implications for SEO and LLM Search

• Entity-first SEO will only grow in importance. Treat content as nodes in a knowledge system.
• Structured data is not just technical SEO—it’s symbolic input for machine reasoning.
• Content that clearly expresses relationships, categories, comparisons, and definitions will perform better in hybrid (LLM + Symbolic) search interfaces.
• Future search optimization will require fluency in both language systems (for LLMs) and symbol systems (for knowledge graphs).

Meaning Is the New Metric

We’re not just optimising for clicks anymore; we’re shaping the way people build understanding.

Search is a meaning-making system.

If your content, structure, and strategy reflect that, you’ll earn more than visibility: you’ll earn trust, preference, and conversions.