Let’s Talk About “Brand Positioning” in the Context of SEO/GEO: A Hamsterdam Marketing Lesson

By Ethan Lazuk

Welcome to another edition of Hamsterdam Marketing! 🐹

This is a project where we examine marketing topics to learn about their fundamental concepts and apply them to the context of SEO/GEO.

This week, we’ll be looking at brand positioning.

In short, brand positioning is about defining an organization’s position in a market relative to its competitors, with the goal of communicating the brand’s value and differentiation to a target audience.

Why I think brand positioning matters to SEO/GEO has to do with web consensus. In essence, web consensus is the process of making a brand’s positioning well-supported across the information ecosystem so that search engines and AI assistants can confidently understand and reproduce it — it’s a degree of corroboration.

When people ask me, “What’s the shortest way to achieve AI visibility?” my answer is generally to get consensus about your brand across the areas of the web where your target audience is present.

That’s the value of brand positioning.

Let’s define brand positioning a bit more, with an illustration.

For starters, we can use this template:

For [target audience], Brand X is the [category/frame of reference] that provides [important differentiated benefit] because [reason to believe].

For example, let’s imagine a fictitious software company:

Northstar is the payroll platform for small nonprofit organizations that makes nonprofit payroll and compliance easier because it is specifically designed around nonprofit employment requirements.

There are several separate pieces in that statement:

• Target: small nonprofits
• Category: payroll software
• Problem: payroll and compliance complexity
• Differentiation: nonprofit specialization
• Benefit: easier payroll administration
• Evidence: nonprofit-specific features/expertise

Incorporating the above elements matters because positioning isn’t merely a slogan.

If we were to say, “Payroll without the headaches,” that’d be messaging.

But if we said, “Payroll software for small nonprofits,” that’s much closer to positioning.

Positioning statements also align with the concept of extractability, or making content easier for AI assistants to parse and understand, such as by using semantic triples (subject > predicate > object) and entity-rich information.

Positioning requires both similarity and differentiation.

A core tenet of positioning is that a brand must belong to a recognizable category before it can successfully differentiate itself within that category.

We can describe this categorization using points of parity and points of difference. Successful positioning requires a brand to have a competitive frame of reference to establish that it possesses the necessary attributes to be considered a legitimate member of its category, and only then can the brand communicate why it’s different.

A company like Volvo can’t merely say, “We are safe.”

The brand first needs to establish its category to consumers as an automobile manufacturer. Only then can it say, “We are Volvo, an automobile manufacturer associated with safety.”

Those points of parity and difference matter for SEO/GEO, especially when we’re talking about entity clarity.

An AI assistant must determine:

• What entity is this?
• What category does it belong to?
• What attributes distinguish it from other entities in that category?

What makes matters tricky is that positioning isn’t determined by what’s on your website alone (your owned media).

Rather, positioning succeeds or fails outside the company.

A brand can declare its ideal positioning, but the market determines whether that positioning resonates successfully with the brand’s target audience.

In other words, brand positioning exists in the consumer mindset.

There’s a whole range of vocabulary around brand knowledge, of which positioning is just one part:

• Brand identity: What a brand says it is.
• Brand positioning: What a brand intends to stand for relative to alternatives.
• Brand messaging: How a brand communicates its positioning.
• Brand image: What people actually believe about a brand (it’s reputation).
• Brand equity: The value created because those associations exist.

This creates a basic marketing feedback loop:

Company → communicates position → audience experiences brand → audience forms associations → market repeats/reinforces associations.

The strongest brands reach a point where the market effectively does some of the positioning work for them.

You’ll see this with brands that are known to be “helpful for beginners,” “great for enterprise companies,” or “premium options.”

When a brand gets to that stage, its positioning has migrated from a brand claim to market perception.

And that is where the connection to SEO/GEO gets really interesting.

The web has become a machine-readable proxy for brand perception.

Historically, marketers would study brand positioning through methods like surveys, focus groups, customer interviews, sales data, or brand studies.

They still should.

But nowadays, search engines (and the AI assistants that use them through RAG) have another enormous dataset: the web itself.

Consider everything that may describe a brand on the web:

• Its website
• Its About page
• Product pages
• LinkedIn
• YouTube
• Wikipedia or Wikidata (when appropriate)
• News coverage
• Industry publications
• Reviews
• Reddit discussions
• Directories
• Marketplace listings
• Analyst reports
• Partner websites
• Association memberships
• Conference speaker bios
• Podcasts
• Research citations
• Customer case studies
• Job listings
• Google Business Profiles
• Social content

Those documents collectively contribute to entity clarity, describing what the entity is, what it does, who it serves, what products it sells, how people perceive it, what topics it is connected with, how trustworthy it appears, and what differentiates it.

In short, your brand positioning gets translated through web consensus, if you’re lucky.

I’ve seen cases where negative sentiment reported in an authoritative source influenced the positioning of a brand in AI-generated answers years afterward.

Google’s Knowledge Vault research illustrates this underlying concept well. It investigated constructing structured knowledge by combining information extracted from text, tables, page structure, annotations, and existing knowledge repositories, then calculating probabilities around factual correctness.

That doesn’t mean Google’s current ranking systems use the Knowledge Vault, necessarily.

But it does illustrate how machines can collect assertions about entities from multiple sources and reconcile them into representations of what appears to be true.

A survey of machine knowledge published on ArXiv similarly describes how knowledge bases and graphs can be constructed from web content and text sources, including entity identification, entity canonicalization, classification and the extraction of properties associated with entities.

That research is conceptually close to what marketers are trying to accomplish with brand positioning.

So what exactly is “web consensus”?

Web consensus is the degree to which credible, independent (and machine-accessible) sources consistently associate an entity with the same facts, categories, attributes, expertise, reputation, and use cases.

The two words that stand out from that definition are “independent” and “consistently.”

Hypothetically, a brand could create 50 press releases stating its brand positioning.

That’s repetition, but it’s not consensus.

Now imagine if an industry association mentions the brand’s positioning, G2 categorizes the brand, another industry publication reviews the brand, several customers independently review the brand, an industry conference describes the brand, and a third-party comparison article does as well, all aligning with what’s mentioned on the brand’s website content and structured data.

That’s much closer to true consensus — different sources, created for different purposes, repeatedly arriving at roughly the same association.

We could even call that semantic consensus, where the language may differ, but the underlying entity relationships remain consistent.

That gives AI assistants a much richer set of independent evidence from which to understand the brand.

Think of positioning as an entity graph.

We can visualize brand positioning as a set of relationships:

• Brand → is a → category
• Brand → serves → audience
• Brand → solves → problem
• Brand → specializes in → topic
• Brand → offers → product/service
• Brand → possesses → differentiator
• Brand → associated with → attribute
• Brand → trusted by → customer/category
• Brand → recommended for → use case

Now ask yourself, how many credible places around the web are reinforcing those edges for your brand today?

That’s a far more useful question for your SEO/GEO strategy than simply asking, “How often are we mentioned?”

SEO has always contained elements of brand positioning and web consensus.

Just because we’re in the GEO era doesn’t mean every tactic is brand new.

SEO has long involved external signals that extend beyond the website itself.

Consider links; the web’s link graph is fundamentally an external endorsement structure.

Local SEO provides another example: Google says local results largely depend on relevance, distance, and prominence or popularity. Google’s local SEO guidance has historically explained that prominence can incorporate information Google has about businesses from elsewhere on the web, including links, articles, directories, and reviews.

That’s essentially the consensus framework.

For example, a restaurant can’t simply put on its homepage:

“We are a famous Italian restaurant in Brooklyn.”

Google has to observe the wider ecosystem to know if it’s true:

• Restaurant directories
• Reviews
• News articles
• Links
• Google Business Profile information
• Other references to the restaurant

The brand’s self-description is only one component.

Before attempting to build consensus around positioning, a brand should first create consensus around identity.

Google’s Organization structured-data documentation recommends giving the search engine information such as an organization’s name, alternate name, URL, logo, address, and telephone, along with other relevant attributes.

But that schema type also supports sameAs, which can point toward profiles on third-party websites containing information about the same organization.

That’s effectively entity disambiguation.

A machine can’t confidently understand that:

Apple is an excellent product for students.

It first needs to understand which Apple?

This need produces a logical hierarchy:

Identity consensus → category consensus → attribute consensus → reputation consensus.

If a brand’s identity is confused, everything layered on top becomes more opaque.

So, how does this influence how we think about on-page SEO?

Traditional SEO might start with the question, “What keyword should this page target?”

A brand-positioning approach starts one level higher: “What should this organization become associated with?“

Keywords (and prompts) are still important because they represent how users express their needs, but the strategic objective becomes association building.

Suppose our brand positioning centered around “technical SEO for enterprise ecommerce brands.”

We wouldn’t merely create one page on our website targeting that phrase.

Rather, we’d want the entire content ecosystem to reinforce connections between our brand and related attributes like:

• Technical SEO
• Enterprise websites
• Ecommerce
• JavaScript
• Crawl management
• Faceted navigation
• Large-scale migrations
• International ecommerce
• Product structured data
• Log-file analysis

This way, the website establishes a topical neighborhood around the brand.

That is both a search strategy and a positioning strategy.

So, if you’re stuck trying to come up with topics for your content strategy, don’t merely look at keywords; think about your brand positioning and the attributes related to it. Those are your playground.

But again, it’s not only the content on your website that counts. It’s web consensus, especially for GEO strategies.

GEO expands SEO into PR, brand, and reputation.

While SEO historically operated primarily through website architecture, content, and links, GEO increasingly intersects with SEO, content, digital PR, communications, social, reputation management, product marketing, customer experience, and branding.

This is because each discipline creates different pieces of the information environment an AI assistant might encounter.

While conversations over ownership are interesting, the truth is no one department should own GEO. It’s a collective effort, and brand positioning is at the center of it.

In GEO, brand positioning should be evidence-producing.

To create brand positioning, you wouldn’t merely distribute your positioning statement and call it a day. Instead, you should distribute evidence that allows external parties (your target audience) to arrive at that positioning independently.

This isn’t about having consistent messaging across all channels. In fact, you shouldn’t.

Rather, you should adapt your messaging to each channel’s culture and audience, but ensure semantic consistency.

It’s not about 50 websites reciting your press release verbatim. It’s about independent sources arriving at related conclusions based on the evidence you provide.

That’s brand positioning in the GEO era.

And original research can be particularly powerful here.

While you could publish 10 pages proclaiming yourself a “GEO professional,” that’d lack evidence and wouldn’t contribute to web consensus.

Alternatively, you could conduct an independent study on 10,000 prompts across ChatGPT, Gemini, and Perplexity, then publish your research, methodology, dataset, etc. and let your target audience come to their own conclusions.

Do enough of that independent research over time, and you’ll naturally demonstrate your GEO professionalism with evidence that gets reinforced throughout the web as semantic consensus, i.e., brand positioning.

A quick note: consensus doesn’t mean ubiquity.

One mistake I’ve seen SEO/GEO practitioners make is manufacturing inauthentic mentions across the web.

The fact is that 10 relevant sources can be more meaningful than 10,000 irrelevant ones for achieving web consensus.

The web isn’t a democracy where every URL receives one vote. Relevance counts.

This line of thinking is consistent with decades of information retrieval and knowledge-fusion research, where sources are known to differ in reliability, so multiple observations do not necessarily mean independent observations. For instance, Google’s Knowledge Vault work deals with reconciling web extractions of varying reliability rather than simply counting duplicated claims.

We’ve been talking a lot about the connections between brand positioning and web consensus in this article. They are related.

That said …

Brand positioning and web consensus ultimately solve different problems.

Positioning answers: What does a brand want to be known for?

SEO answers: Can the relevant information be discovered, understood and ranked?

Entity optimization answers: Can the system correctly understand who or what a brand is?

Digital PR answers: Can credible third parties validate a brand’s expertise and relevance?

Reputation answers: What does the market actually say about a brand?

While GEO answers: When an AI constructs an answer to the questions a brand’s audience asks, is that brand sufficiently retrievable, relevant and supported to become part of the answer?

They’re all addressing separate questions. Web consensus unifies them.

The final question is: how to measure brand positioning and web consensus?

I’ve written extensively about prompt-tracking for AI visibility measurement on my blog.

Prompt-tracking can answer the question, “Were we mentioned or cited as a brand?”

That’s a good starting place, but to go deeper on brand positioning and web consensus measurement, there’s a host of other questions that need to be addressed:

• Entity accuracy: Did the AI assistant correctly identify us?
• Category association: What category did it put us in?
• Audience association: Who does it think we’re for?
• Attribute association: What are we known for?
• Competitive association: Which companies appear beside us?
• Recommendation context: For which use cases are we recommended?
• Sentiment: How does it characterize us? (Not just the score out of 100.)
• Evidence: What claims does it make?
• Citation provenance: Which sources appear to influence those claims?
• Source diversity: Are multiple domains independently reinforcing the claims?
• Prompt stability: Do those associations persist across different prompts and repeated runs?

This also changes how we do competitive analysis.

Traditional SEO might ask, “Why does Competitor A rank #1?”

Whereas a GEO who cares about positioning might ask, “Why does the internet seem to believe Competitor A is one of the best companies for this problem?”

That question invites a slew of investigative questions:

• What categories does the competitor appear in?
• What reviewers discuss them?
• Which publications mention them?
• Which attributes repeatedly occur near their name?
• Who links to them?
• What experts talk about them?
• What do Reddit users say?
• What do YouTube reviewers say?
• Which comparison pages include them?
• What original research did they publish?
• What entities are they connected with?
• Which prompts trigger them?

But the biggest question is around category ownership. Who has it, and why?

Are we entering an era of brand-positioning GEO?

Let’s look at the evolution of tactics and principles within SEO/GEO:

First, old-school SEO was about optimizing a document so it ranks for a keyword.

Then modern SEO was about building a useful, technically accessible information ecosystem around audience needs.

Later came entity SEO to help machines understand the people, organizations, products, topics and relationships involved.

GEO then emerged to help generative systems retrieve enough high-quality information to accurately incorporate a brand and its content into synthesized answers.

Now we may be entering an era of brand-positioning GEO, where the goal is ensuring that when machines synthesize information about a market, the truthful associations a brand wants to own are repeatedly supported by the broader information ecosystem.

Takeaways

Brand positioning is what you want your brand to be known for relative to your competitors. That includes information about your category, audience, expertise, and points of differentiation.

For SEO/GEO, your positioning should be reinforced across the web, not just on your own site. This is the idea behind web consensus.

Web consensus is built from consistent, credible third-party corroboration across sources like reviews, forums, directories, and industry sites. The goal isn’t more mentions, but stronger associations.

Search engines and AI assistants should repeatedly encounter the same core story about who you are and what you are known for. That’s semantic consensus.

See you next time!

I hope you’ve found this discussion around brand positioning helpful.

This is only one of a few Hamsterdam Marketing lessons so far, with a new one to come next week hopefully.

If you want a hand creating web consensus around your brand positioning, get in touch with me. I’m an independent SEO/GEO consultant based in New York City.

Until next time, enjoy the vibes:

Thanks for reading. Happy optimizing!