Name It to Claim It

If you open ChatGPT, Claude, or Perplexity and type: "Explain Jobs to Be Done theory," you'll get Harvard professor Clayton Christensen's name mentioned in the first few sentences.

Every time. Every model.

Ask the same tools to explain the core idea without naming it: "Why do people buy products based on the progress they're trying to make rather than demographics?"

The answer gets muddier, less attributable, and anchored to no one. The AI tool will construct something useful from general knowledge…but it won't tie the idea to a particular person when it does.

That gap is what this whole edition is about.

Christensen didn't invent the observation that people hire products to solve problems; researchers and consultants had been circling that insight for decades. But he named it, published about it under that name repeatedly, and, over time, the label stuck.

"Jobs to Be Done" became a discrete entity in the training data, a handle that the model can grab onto, and a thing LLMs (and people!) can point to and say: this idea belongs to this person.

Your unnamed expertise doesn't get that treatment, no matter how original it is. This fact is incredibly important to your AEO strategy.

Today's newsletter is brought to you by Ahrefs and Agent A, their new AI marketing agent powered by Ahrefs data. You can ask Agent A to research competitors, find content opportunities, build reports, analyze performance, and turn repeatable marketing tasks into tools and automations. It’s part chatbot, part analyst, part app builder, with full access to everything inside Ahrefs.

What LLMs need to cite you

Over the past two months, I've been telling you about the Source Signal Stack: a framework for understanding which content signals LLMs trust when they decide who to cite.

The four layers (Brand, Executive, SME, Community) map the infrastructure side: Where to publish, which voices to leverage, and how far from brand control a signal needs to originate to carry weight within the AI citation equation.

Today, I want to dig into the prerequisite beneath all four layers that determines whether the content you produce through any of them actually becomes citable, or just gets absorbed into the model's general knowledge without attribution.

That prerequisite is entity resolution.

For an LLM to cite a person, a company, or a framework, it needs three things:

A discrete label. A name for the thing (idea, framework, POV) that’s specific, searchable, and serves as a two-to-five-word “sticky” descriptor of your concept.
Repeated co-occurrence. That label needs to show up alongside a specific person or brand, consistently, across enough documents that the model builds a statistical association between the name and the idea.
Cross-platform verification. LLMs need to see this same label appearing on LinkedIn, in a podcast transcript, in a newsletter, in a trade publication, in a conference talk, as this is a cross-verification that helps prove that multiple independent sources reinforce the same association of person + idea.

Without a name, there's nothing to resolve. Your brilliant insight about B2B buyer psychology, developer tooling, or sales enablement gets absorbed into the model's general understanding of the topic.

It becomes background. The LLM might use your thinking to construct an answer, but it won't point at you when it does. And that’s a problem.

With a named framework, however, the math changes.

The label becomes a node in the model's knowledge graph. Every new document using that label alongside your name strengthens the association. Every new platform adds a verification layer.

Named vs. unnamed frameworks: The receipts

This pattern plays out across B2B with striking consistency: The practitioners who own citations in AI search are the ones who’ve packaged their expertise into a named, repeatable concept and published it across enough surfaces for LLMs to pick it up.

A few worth examining:

"Zero-Click Content" (Amanda Natividad). Every content marketer was already feeling the reality that social platforms suppress outbound links, but Amanda named it. The handle now gets attributed to her across ChatGPT, Perplexity, and Google AI Mode. She didn't discover the phenomenon, but she named it, owned the association, wrote about it under that labeled concept repeatedly, and it stuck.
"They Ask, You Answer" (Marcus Sheridan). Thousands of content agencies were already practicing the approach of building content around customer questions, but Marcus gave it a name, wrote a book about it, and built a business around the label. Ask any LLM about content strategy for small businesses, and his framework surfaces.
"Dark Social" (Alexis C. Madrigal). Coined in a 2012 Atlantic article to describe traffic shared through private channels analytics tools couldn't track. The concept existed long before the article, and marketers knew about it, but Madrigal named it, and 14 years later, LLMs still attribute the term to him.
"10x Content" (Rand Fishkin). Simple concept: to rank, your content needs to be ten times better than what's currently on page one. Fishkin put a name on it during a Whiteboard Friday. Two words turned a subjective quality standard into a retrievable, citable entity.

Now think about the consultants, VPs, and experienced practitioners who have ideas from their hands-on work that are just as sharp as any of these.

Think about your own professional experience. What are the uncovered concepts, truths you’ve found via experimenting and testing your ideas?

The reality is: You probably just haven’t uncovered it yet.

You (or the subject matter experts you work with) have these valuable insights, but haven’t had the brainspace to sort through them yet to pin them down. It’s all still floating around as a gray, fuzzy, ambiguous idea (just waiting to be turned into a concrete, named framework.)

Why named frameworks a part of a smart AEO strategy

In April, I wrote about the content monoculture problem: AI-generated content is flooding every platform with unoriginal takes and zero human perspective or experience-informed insights.

When every B2B SaaS blog publishes the same "7 Best Practices for X," it’s just adding more noise to the pile of mediocre, bland content that already exists.

There’s no reason for LLMs to cite that content because it doesn't add anything new to the conversation; it's just regurgitating what's already been said in slightly different words.

This is a great way to spend a lot of time and money on an expensive tool or poorly planned “AEO initiative” that doesn’t actually move the needle or result in more sales, and keeps you running in place for the next 12 months.

Named frameworks can cut through the monoculture precisely because they are, by definition, distinct. A named framework signals something to the model (and to the human reader): this person thought about the problem differently enough to give it a label.

That's a differentiation signal. It tells the LLM this isn't generic category content. It's a specific, attributable point of view.

There's also a compounding dynamic worth flagging.

As more companies invest in AEO and build out their Source Signal Stacks (and they will, because the data is impossible to ignore), the advantage of simply having source signal infrastructure narrows.

Many companies will eventually have a program to get employee SMEs posting on LinkedIn, and B2B creators will have established long-term partnerships with the brands that were smart enough to see the future and lock them into a deal to create content around their products and services on an ongoing basis.

The bottom line here is finding (and naming!!) the IP inside the stack. The named, ownable frameworks no one else can claim because you coined them.

Source signals get you into the conversation, and from there, named frameworks help make you the one the AI points to as you work to own your topic.

The anatomy of a citation-worthy named framework

If naming is the leverage point, what actually makes a framework citable? Not every label sticks. Plenty of people have tried to coin terms that went nowhere.

The frameworks I’ve seen LLMs consistently cite tend to share three key traits:

1. A clear, specific name (2-5 words, memorable, searchable).

This sounds obvious, but the number of experts who describe their methodology as "our approach" or "what we do" is staggering. "Our approach" is not an entity, nor is it searchable. An LLM cannot build a knowledge node around it (and to be frank, no human is going to find that very sticky, either.)

The name needs to be specific enough to be distinct and short enough to function as a label.

2. A specific claim that takes a position.

The framework can't just describe what exists; it has to assert something.

"Jobs to Be Done" isn't just a label for customer research. It's a claim that the fundamental unit of analysis in product development should be the job the customer is trying to accomplish, not the customer's demographics. That's a position; you can agree or disagree with it.

"Zero-Click Content" isn't just a label for posts without links. It's a claim that the dominant content distribution strategy (drive clicks to your site) is broken on social platforms, and that the winning move is to deliver value entirely within the feed. That's a position.

Frameworks without a claim are just categories.

3. Published evidence from your actual work.

This is the big one. Frameworks need to be rooted in real, tested, hard-won experience. Not theoretical or aggregated from other people's research, but rooted in direct experience and data to back up the claims you’re making.

Sheridan built "They Ask, You Answer" out of the content strategy that actually saved his pool company during the 2008 recession.

The evidence doesn't need to be a peer-reviewed study, but it does need to be first-party. "I observed this pattern while doing this specific work" carries more weight with LLMs and human readers than "research suggests."

The Source Signal Stack itself checks all three boxes: a clear name, a specific claim (that the further a signal originates from brand control, the more weight LLMs give it), and evidence from my real client work diagnosing why content programs produce zero AI citations.

Digging up the idea

Here are three questions I want you to sit with this week, either for yourself or for your marketing department:

Is there a concept you explain to clients or colleagues repeatedly that you've never seen anyone else articulate the same way? Not a common best practice, but something you developed through your own work.
Do people come to you for a specific type of advice where you've developed a consistent, repeatable approach? Key word: consistent. If you give different advice every time, you don't have a framework yet. If you find yourself walking people through the same logic, the same steps, the same diagnostic questions over and over, that's the signal.
Have you ever drawn something on a whiteboard (or in a doc, or on a napkin) to explain your thinking that made a room go "Oh, I get it now"? If you've externalized your thinking into a visual model that creates clarity for other people, you're closer to a named framework than you think.

If you said yes to any of these, you may have a possible framework…you just haven't named it.

And until you do, it's invisible to the systems that increasingly decide who gets cited, who gets found, and whose expertise shows up when buyers go looking for answers.

Most experts I work with are sitting on something nameable. The hard part isn't having the expertise, it's knowing which piece of it to extract, how to package it, and what to call it.

Coming soon

I'm building something for practitioners, consultants, and SMEs who know they have expertise worth citing but haven't figured out how to extract and name it.

More details on that soon, but if you want to be first to hear about it, reply to this email with "interested" and I'll add you to the early access list.

'Til next time,

Kaleigh Moore