🦄 “Unicorn” Clicks & the Magic of Etymology for SEO Research, a Hamsterdam Marketing Lesson
By Ethan Lazuk
Last updated:

Welcome to a new week of Hamsterdam Marketing! 🐹
If you’re new here, this is where we look at topics related to marketing and the social sciences and apply them to SEO strategies to grow as holistic marketers.
Last time, we examined how cultural anthropology could inform SEO tactics, specifically for keywords, content, images, and page experience.
Part of that post referenced syntax and semantics, two of the four subsystems of every language (along with phonology and morphology). 🗣️
Linguistics is a branch of anthropology, like cultural anthropology (my college major). 📚
As the study of language and its structure, linguistics is especially relevant to natural language processing (NLP) and today’s search engine systems and generative AI models. 🤖
One area of linguistics that’s always fascinated me is etymology, the study of word origins.
I went on a bit of an etymology tangent in a recent post about epistemic and aleatoric uncertainty in LLMs.
For example, I explained how my surname, Lazuk, could break down as “Laz” (an ethnic group, partly in Turkey) + “-uk” (“son of” in Ukrainian) to describe a diaspora community, or “little Laz.”
More generally (and practically), studying word origins can inform aspects of our SEO work.
It can have practical benefits for keyword research, audience personas, and content strategy, but we can also use etymology for research within our field.
In this article, we’ll apply etymology to the example of “unicornClick,” an attribute in the Google API Content Warehouse. 🦄
We’re not going to explain what a unicornClick does, exactly.
Instead, we’ll show one way to help figure out its meaning using etymology. 🧐
But first, here’s what got me thinking about “unicornClicks” and etymology.
Marie Haynes recently published a nice blog post about Navboost, in which she mentions “unicornClicks.”
Specifically, it’s in the context of “Navboost Craps.”
Ok, ok.

You’re right, Dale’s dad.
Professionals never do childish things on purpose. 🦁

Touche. 🤺
In a past Hamsterdam recap, I included a tweet from someone who said they knew what “craps” stood for.
Their post was deleted, so I won’t blow up their spot 🕵️♀️ … but to paraphrase:
“Craps doesn’t refer to gambling (or other mental images …). It’s an acronym for ‘Click and Results Prediction System’ (CRaPS) and is related to research into CTR prediction derived from …”
– A long-gone tweet.
The rest got cut off, but I believe it also referenced Google Ads. (But don’t quote me.) 🤷
The larger point is that we associate words with meanings based on our cultural upbringing, surroundings, or other factors.
Words can have both denotative (literal 🌹) and connotative (figurative ❤️) meanings. (Writers are eternally grateful for that.) ✍️
BTW, did you know Shakespeare’s sonnets and writing used to rhyme better, but languages evolve, so now they don’t as much …
TIL Shakespeare’s works have lost many of their rhymes and puns as a result of the evolution of the English language and it’s pronunciation
byu/seanleephoto intodayilearned
When I heard “craps,” for instance, I thought of a play on the word “carp.” Being Montana born, I remember people colloquially calling it “crap” fish:

It’s a play on the name and a reference to the resulting cuisine.
It’s good to be a vegan. 🌱
Ok, bring on the jokes …

😭
Moving on!
So, we have the possibility of “craps” being just an acronym.
Then again, it could refer to a game of chance, secondarily 🎲 …
Tech people use analogies, historical references, and even double entendres all the time.
I recently wrote about the origins of the Apple Safari browser, which was rumored to have had a codename of Alexander, short for Alexander the Great, an homage to the “Konqueror” open-source browser. ⚔️
And recall the origins of Inktomi, the first web crawler used at MSN. It was a Lakota legend about a trickster spider-like man. 🕷️ 🦸♀️
I also wrote about the etymology of “aleatoric” uncertainty (randomness). It comes from the Latin word “aleator,” meaning a dice player. 7️⃣ (RIP, Snot Boogie.)
But on that point …
Prompt: “Could clicks in a search engine be described as aleatoric uncertainty?”

Games of chance 🎰 …
Aleatoric uncertainty 🤨 …
An ads connection (maybe) 💰 …
I could see some clever engineers having some fun with that one.
More broadly, words have cultural context.
In the language of a Google engineer, “craps” has a specific meaning.
Much like how we hear “snippet” and think “chunk of text,” rather than “haircut” ✂️ or …

As part of that cultural context, word origin stories (etymology) can give us further insight into their meanings.
If we considered the etymology of “unicorn” literally, it would look like this:

We see the origins are Latin (“uni-” for single and “cornu” for horn, or “unicornis”), with Old French and then Middle English adoptions.
But honestly, who these days studies Latin? Other than …

👏
So let’s apply some more practical etymology to figure out (or get closer to) the meaning of “unicornClick.” 🦄
First things first …
What does “unicorn” mean?
This is probably the first concept that came to mind:

An association with fantasy, horses, and rainbows. 🧙 🐴 🌈
After all, it is Pride Month currently. 🏳️🌈
But also note the PAA questions in that screenshot.
“Unicorn” has polysemous (and context-dependent) meanings that may apply here. 🙌
Here’s a figure from a recent TDS article on geometric deep learning (my “sneaky pick” in the last Hamsterdam recap), which includes contributions from some prestigious minds in the ML space (both private sector and academia):

And what’s that in the lower corner?

Let’s now try something interesting with Gemini …
Prompt: “Hey Gemini, why might the authors of this paper (link) have a unicorn emoji on this image (attached)? Is there any significance to the concept of a unicorn in the space of tech companies and academia?”

In case you’re on mobile and can’t see that, it says unicorn in this context can “symbolize the elusive nature of an ‘ideal’ graph foundation model.”
More broadly in the tech space, “a ‘unicorn’ is a privately held startup company valued at over $1 billion.” 🧑💻
While in academia, “the term ‘unicorn’ might also allude to a theoretical concept or a solution that is highly desirable but not yet realized.” 🧠
“Additionally,” Gemini said (or predictively replied), “unicorns often represent innovation and disruption.” 🤯
So let’s now ask about “unicornClicks” directly, sort of …
Prompt: “Ah, interesting! So let’s say we had a search engine named Fanny that runs on today’s technology and has one system called BuddhaBowl. This system stores queries and user interaction data like clicks to help adjust rankings and SERP features. The clicks get categorized as goodClicks, badClicks, lastLongestClicks, and other data, but it also has a click designation for ‘unicornClicks.’ What do you think ‘unicorn’ would refer to in this context of BuddhaBowl?”

Unlike our first response from Gemini, this one doesn’t have verifiable sources (red highlights) because it’s a term we don’t know about (publicly). 🚦
Side note nerdery: here’s more about that (post-hoc) citation process, if you’re interested:
But more to the point, here’s what Gemini’s response said a “unicornClick” could mean:
In the context of a search engine’s click data, ‘unicornClicks’ likely refer to:
- Extremely rare clicks: Unicorn clicks could represent interactions that happen very infrequently, perhaps on specific types of queries or for unique user journeys.
- High-value clicks: These clicks might signify interactions with exceptionally relevant results, indicating a perfect match between the user’s intent and the content provided.
- Clicks leading to conversions or desired actions: Unicorn clicks could represent those that lead to a user taking a valuable action, such as making a purchase, signing up for a newsletter, or downloading a resource.
In essence, ‘unicornClicks’ would be a way for the search engine to identify and track the most impactful and valuable user interactions, which could be leveraged to improve search algorithms, ranking models, and SERP features.”
– Gemini response to above prompt (6/25). (My highlights, using colors from a how-to-draw-a-unicorn YouTube video.) 🎨
That kind of makes sense!
We didn’t get an exact answer.
However, we got several logical possibilities given the etymology of “unicorn” in the cultural context of tech companies and academia.
A “unicornClick” could be 1) a perfect click (which Jason Barnard recently elaborated on in SEL), 2) a valuable click for conversions (corroborated in some way, I’d imagine), or 3) just some uncommon behavior (or specific query or user journey types).
Hard to say without the full context. 🙃
🚨 We have an update from a week later. (New Older insights!)
When I searched for [what is a unicorn click in SEO] using the Google app last night (7/3), I found an article by Larry Kim for SEL in 2017 that gives us additional definitions. 🕺 🪩
Wait. Why didn’t I do that in the first place?
Well, it’s a long story.
I guess it all started when I was a child who wanted to be different. Those habits stuck throughout adolescence.
Now, when I write about SEO, I do research, sure, but I also try not to get influenced by …

Oh, snap!
Back to it:
So, the AI Overview kind of mixes up the nuance of what Larry wrote:

Remember how earlier we speculated that CRaPS might be related to Google Ads?
Here’s Larry’s actual second paragraph:
“After all, branded keywords tend to be ‘unicorns’ — in PPC, these are keywords which generate ridiculously high CTRs (click-through rates) (40, 50 or 60 percent) and Quality Scores (9 or 10); in SEO, these are keywords that rank at the top of the SERP. Meanwhile, non-branded keywords perform like donkeys — in other words, average at best.”
– Larry Kim, SEL. (My highlights.)
Click those AIO sources, people. 😇
But if a unicorn click is, in fact, a paid click with a high CTR, that would align with number 2 from earlier, or a valuable click for conversions.
On the SEO side, if a unicorn click goes to a top-ranking result, that would be more similar to number 1, or a perfect click (assuming it fully satisfies the user’s search intent).
Either way, this is a good reminder, as part of etymological SEO research, to not be afraid to check those SERPs. 🫣
Summing up where we’re at, so far — we have several possible meanings for unicorn click. 🦄
That’s epistemic uncertainty, which stems from a lack of knowledge that could be resolved with more context or clarification.
It’s not just us, though.
Google’s ranking systems have it, too. That’s why we also get a triple featured snippet, which can be shown if there’s no single answer or multiple perspectives:

Something we can feel certain about, though, is that knowing the origins of terminology in relevant cultural tech circles could help inform us about attribute names from the API docs.
And here’s something else to consider, a little out of the box …
That mention of “unicornClicks” was still in reference to Navboost “craps,” right?
There’s another place in the commercial universe where those overlap:

Where are the Ghost Busters when you need ’em?! 👻 🚨 🚓

But there’s a serious angle here, kind of …
What’s the origin of those products? 🍭
Prompt: “Hey Gemini! How might this strange collection of products related to unicorn [redacted] have gotten started? (Attachment.) Where did the idea come from?”

There are a few buzzwords in that response (which I won’t be transcribing, as I can already see the resulting GSC query data now …): “magic,” 🪄 “fantasy,” 🧟 and “social media.” 🤳
Let’s try one last follow-up …
Prompt: “I noticed you mentioned magic, fantasy, and social media. Do you see any connection between those products and the tech community’s associations with the word ‘unicorn’?”

Here’s part of Gemini’s answer (with a choice redaction):
“Yes, there’s a definite connection between the whimsical unicorn [redacted] products and the tech industry’s use of the term ‘unicorn.’ …
In both cases, the unicorn serves as a powerful symbol that resonates with people’s imagination and aspirations. It represents the pursuit of something extraordinary, whether it’s a groundbreaking technology or a playful and whimsical product.“
That connection is a little tenuous 🧶, and likely a bit of a stretch to make …
However, we can see how “unicorn” 🦄 can have multiple meanings that apply in different cultural domains, which people might co-inhabit 🏘️, like tech and fantasy fandom.
This etymological approach applies to other areas of SEO research, too, such as research papers and patents.
We’ve been talking about “clicks” because that’s our SEO lexicon. It’s what get reported in Google Search Console, for example. ✅
But word and concept meanings can vary based on their cultural context, like internally at a company.
Have you ever worked for a brand with a mascot or its own lingo (and emoji meanings 🐿️)?
Or it could be broader than that.
Like the difference between colloquial speech and corporate jargon ☕️:
@morningbrew Corporate talk is a whole different language #corporatelife #corporate #corporatetok #officelife #business ♬ original sound – Morning Brew
Or legalese 🧑⚖️:
Or academic writing vs. whatever is going on here 🐢:
As AJ Kohn (the wordsmith behind “It’s Goog Enough!”) recently explained in a webinar:
“If you look back at the foundational patents, you can’t look for ‘clicks.’ The magic words are ‘implicit user feedback.’“
– DesignRush roundtable (21 min mark). (My formatting of the YouTube transcript.)
🪄

And looking forward, how much will clicks matter for the future of search?
You’re asking me?! Well, to circle back with you (☕️), Marie Haynes linked to some great sources in her post.
Also, recall this line from Google’s antitrust post-trial debrief:
“the overwhelming majority of the quality gap between Google and Bing cannot be explained by the different volumes of user interaction data available. And the testimony of Google engineers confirmed that Google’s use of user interaction data has always been but one of many inputs into Google’s systems, has decreased over time, and that the reliance on that data is in the midst of a dramatic change given the rise of large language models that employ fundamentally different techniques to solve the same task, i.e., predicting the usefulness of documents given the query. For all of these reasons, there is no evidence that any share shift would generate any meaningful quality improvement.”
– Google Post-Trail Debrief, Page 76 of 123
And this other line from several pages earlier:
“new general search engines can avoid many of the costs historically borne by an incumbent like Google. As Neeva recognized, advances in the ability of computers to understand language ‘could be used as a short circuit to make ranking better,‘ and ‘when it comes to figuring out concepts for a query, related queries, or correcting misspellings that people often have when they type in or speak queries, [Neeva was] able to very, very successfully do that.’”
– Page 42 of 123
While we don’t have Neeva anymore, we do have Perplexity.
Here’s an excerpt from an interview Perplexity’s CTO and co-founder, Denis Yarats, recently gave on a machine learning podcast (which I’ve formatted/rephrased a tad for clarity):
“Because our index is not as large yet, we kind of spend a lot of time to build this trust score for a domain or a web page, and so we only sort of prioritize in ranking stuff that is high quality, and then the LLM plays a big part of it. …
The way people play is they try to figure out the ranking signals … and actually there was like a couple of weeks ago there’s a leak right, and then so they you know build things to sort of like trick the ranking system, so maybe it’s going to be like more clicks or like whatever other stuff …
Because those things are not part of our ranking system, it’s kind of like that content doesn’t go up as high as otherwise.”
– Denis Yarats (Gradient Dissent podcast). (My transcription.)
I put a fuller transcript in the last Hamsterdam recap. However, it’s a natural convo and thus hard to transcribe, so I’d suggest checking out the video:
So, what did we learn here? 🙋
Well, words can have multiple meanings. 🙌

And research of word origins (etymology) can help us situate those meanings in their cultural context. 💅
The next time you hear “unicorn in the tech space,” for example, what image will your mind conjure?

Close enough.
We also know tech folks like to play with semantics, literally and figuratively. 🤓
Still, Kendrick’s got us all beat as far as entendres go …
Outro
Thanks for reading this week’s Hamsterdam Marketing lesson about etymology and SEO research, specifically looking at unicornClicks! 🦄
We didn’t figure out the meaning of “unicornClick,” exactly, but at least we got a few possibilities to ponder: 1) perfect click, 2) valuable click for conversions, 3) uncommon behavior (or specific query type), 4) high CTR ads click, or 5) top-ranking organic result click. 🔥
Feel free to “click” the comment button to share your thoughts:

Well, that took a turn … 😆
Or contact me with your explicit user feedback, and check out related posts below.
Until next time, enjoy the vibes:
Thanks for reading. Happy optimizing! 🙂
Related posts
Epistemic vs. Aleatoric Uncertainty in LLMs, via a Google DeepMind Paper, “To Believe or Not to Believe Your LLM,” & Why SEOs Should Care (Likely)
In this Hamsterdam Research, we’ll explain Google DeepMind’s paper “To Believe or Not to Believe Your LLM” and why it’s relevant to SEOs (likely).
Language, Art & Culture: SEO Lessons from a Cultural Anthropology Textbook (for Keywords, Content, Images & Page Experience)
This Hamsterdam Marketing lesson examines keyword, content, image, and page experience lessons for SEO gleaned from a cultural anthropology textbook.
USER-LLM: Contextualizing LLMs with User Embeddings for Enhanced Personalization (via Google Research), & Why SEOs Should Care (Maybe)
This rendition of Hamsterdam Research explores USER-LLM, a novel framework from Google Research for contextualizing LLMs with user embeddings of user interaction data.
Leave a Reply