Entities Talked About in 2010 and What Still Applies Today (A Hamsterdam History Lesson)

Last updated:

November 5, 2026

Welcome to another edition of Hamsterdam History!

This is a series of posts where we look at “vintage” SEO articles to celebrate their contributors, gain historical knowledge, and discuss how things have changed (or haven’t) since.

In this edition, we’ll be looking at “Search Engines and Brand Entities – Whiteboard Friday” posted by Aaron Wheeler and featuring Rand Fishkin for SEOMoz on November 11, 2010.

Entity optimization is a primary toolkit within modern-day SEO/GEO strategies. It ties back to semantic search and knowledge graphs.

What’s interesting is that, while Google’s Knowledge Graph came out in 2012 and the Hummingbird Update in 2013, the origins of entities really kind of start back in July 2010 when Google acquired Metaweb, including its Freebase database and entity-relationship technology.

Rand references this development in his Whiteboard Friday:

“So, let’s try and tackle the question, what makes a brand? What is it that separates a brand from a non-brand in the minds of the search engines when it comes to domains, when it comes to websites and pages? You can think of a lot of different things. Certainly Google has put out some patent applications that suggest some of the things they might look at. They made an acquisition of a company called Metaweb that does a lot of these things, including a service called Freebase that kind of makes entity associations from context and text and word usage.”

In fact, Rand’s video includes quite a few prescient ideas that we can relate to modern-day AI search optimization tactics:

“These things can include stuff like appearance and repetition of text content. You can imagine that Adidas, Puma, and Vibram, these show up on the Web a lot more than tennis-shoe- store.info or whatever it is. There is kind of this idea, “Huh, maybe that’s a brand, maybe that’s not.” And then there is context of use and positioning of that text and content.”

That last part about context of use and positioning of the text and content is one of the most forward-looking parts of the transcript.

We know today that it’s not enough for an AI system to just encounter a company’s name. That system also needs to understand what that company does, which categories it belongs to, and how it relates to other entities.

Take a Manhattan architecture firm as an example. Mentions that consistently connect the firm with residential renovations, Upper East Side apartments, and interior architecture establish a clearer topical association than generic references to it as a “design company.”

The practical takeaway is that we should prioritize relevant, contextual third-party coverage over accumulating generic brand mentions as an entity-building strategy.

Rand actually continues on that point:

“You can see that those brands are all mentioned in news and they’re mentioned in blogs. They’re in stores. They’re in different stores both on and off the Web. They’re in eCommerce shops. They’re featured in traditional media outlets, online and offline. You see them in offline media as well. They show up in links. They show up in advertising.”

This is essentially a description of a brand’s digital footprint.

The principle remains useful, which is that independent sources can help establish a company’s identity, activities, reputation, and expertise.

What’s changed is our understanding of quality, meaning that a mention in a relevant industry publication might provide better independent corroboration than dozens of syndicated press releases or low-quality directory listings.

Relatedly, Google’s newer AI search guidance discusses information about products and services appearing in blogs, videos, and forums, while cautioning against inauthentic mentions.

As a practical takeaway, we should try to earn credible coverage that actually explains a brand’s expertise, products, services, or distinguishing characteristics.

Rand also raises a fair point with regard to actionability in the transcript:

“A lot of this is sort of interesting theoretical stuff, but I know that many SEOs are going to be asking the question, “Well, what do I actually do with this data?” So, some good things to keep in mind is that we as SEOs sometimes ignore branding.”

Ignoring branding is perhaps one of the most disastrous moves an SEO/GEO can make today, if their strategy involves sustainable visibility over short-term wins.

The fact is that SEO/GEO, PR, content marketing, social media, and other marketing disciplines can collectively establish the associations people make with a brand. For AI search, those associations also create opportunities for a company to be considered when people ask for recommendations or comparisons.

So, as a practical takeaway, we shouldn’t treat entity optimization as a standalone technical SEO/GEO project, but rather we should connect it with broader positioning, distribution, and reputation-building efforts.

Lastly, as a practical tip, Rand suggests what we’d now call today “entity clarity“:

“You know, as part of that, finally, I would say, try and work on making your site and your product and the naming conventions that you use as brand friendly, as branding friendly, as possible. All of those things are going to potentially impact the way your brand is perceived.”

When it comes to entity optimization, we’re looking for semantic consensus, not necessarily verbatim repetition. In short, you want your brand entity to be associate with related entities that are relevant to your business, both in your owned media as well as throughout earned media.

This consistency can extend to structured data, as well. Google recommends Organization schema to help disambiguate organizations and understand their information, while supported properties include name, alternateName, url, logo, and sameAs.

As a practical takeaway, we should maintain a consistent, identifiable presence across our website, organizational profiles, and relevant independent sources, supported by structured data.

That explains the consistencies between 2010 and today when it comes to entity optimization. Some practical takeaways are strewn above, but let’s consolidate the most actionable advice below:

Build a clear entity identity: Make your organization’s official name, website, description, services, locations, and relevant attributes consistent and easy to verify, and use appropriate structured data where helpful.
Strengthen meaningful entity associations: Develop content and earn relevant coverage that connects the brand with its actual areas of expertise while focusing on accurate relationships, not repetition or mention volume.
Build credibility outside your owned website: Seek independent reviews, industry coverage, expert commentary, and other authentic references that substantiate what your brand claims about itself.
Optimize for retrieval as well as recognition: Publish useful, original, accessible content that directly addresses relevant questions and comparisons.
Measure brand visibility and source visibility separately: Track branded searches, the queries where you appear, AI-generated mentions, recommendations, and citations; available tools include Google’s branded-query filter in GSC.

Takeaways

In the 2010 SEOMoz Whiteboard Friday transcript, Rand’s ideas around entities were largely prescient. In short, he was describing a transition from optimizing individual pages to establishing a recognizable brand within the broader information environment.

What’s changed since 2010 is the sophistication of the AI systems interpreting that environment, along with the different ways those systems can surface information.

Back in 2010 we might ask, “How does Google recognize that a website represents a real, established brand?”

Today, we’d ask additional questions, such as:

“What does an AI system understand about our brand?”
“How does the AI system verify or retrieve information about our brand?”
“Under what circumstances does the AI system mention, recommend, or cite our brand?”

Perhaps the primary takeaway from today’s Hamsterdam History lesson is that entity optimization has evolved from just establishing brand recognition to also establishing a clear, credible, retrievable, and contextually relevant identity across the whole information ecosystem.

Outro

ChatGPT got overly ambitious with generating a featured image for this post and gave me this very detailed infographic. I thought it was useful so I decided to include it: