Last October, the Ahrefs Evolve conference. It was a gathering of SEO experts from around the world to discuss "how to survive in the era of AI search." A wide range of perspectives were exchanged, from the extreme view that "SEO is dead" to the argument that "GEO (Generative Engine Optimization) is the new standard."
I kept thinking on the plane returning from my business trip. How should marketing change in the age of AI? The conclusion was nothing new.
In the age of AI, we must return to the most fundamental questions of marketing
In what situations do customers think of our category?
At that moment, does our brand come to mind?
This is the essence of Category Entry Points (CEP). This concept, introduced by Byron Sharp 10 years ago, is emerging as the center of marketing strategies in the AI era.
So today is a bit long. However, if you are a B2B marketer preparing for 2026, this is a must-read.
In 2025, the landscape of search changed
Everyone is experiencing this in their daily lives, aren't they?.
According to the results of a McKinsey consumer survey last October
- of consumers intentionally use AI-based search.
- It is projected that $750 billion in revenue will be generated through AI search by 2028.
- AI summaries are already displayed in 50% of Google searches, and this will expand to over 75% by 2028
Asking ChatGPT, researching with Perplexity, and having Google AI Overview summarize the answers has now become a daily routine. (With the launch of the shopping feature on November 27, even bigger changes are expected in the future.)
As a marketer, it is only natural to feel confused. What will become of the SEO strategies we have worked so hard to build over the past decade? Will all the effort we put into getting our blog posts to rank high become meaningless?
The experts I met at Ahrefs Evolve all said the same thing.
Return to the essence
Category Entry Point (CEP) and positioning—concepts from 10 years ago—are becoming important again.
What is Category Entry Point, or CEP?
Category Entry Point (CEP) refers to the 'needs,' 'situations,' and 'motivations' that make consumers think of a specific product category. It is a concept that refers to the internal and external "triggers" that trigger purchasing behavior. Category Entry Point is a framework first introduced in 2015 by Australian author Byron Sharp in his book *How Brands Grow*.
B2C Example:
- “When you feel thirsty” → Coca-Cola
- “When you need energy after a workout” → Gatorade
- “When you need caffeine on your way to work” → Starbucks
B2B Example:
- “When team collaboration is difficult” → Slack, Google Docs
- “When customer management becomes complex” → Salesforce
- “When the project goes off the rails” → Notion, Jira
4 Reasons Why CEP Is Important Again in the AI Era
With the recent rise of search powered by generative AI , the landscape of search is rapidly shifting from a "keyword-centric" approach to one centered on "natural language questions and context ." Consumers no longer simply enter a few keywords into the search bar; instead, they describe their problems or needs in detail using everyday language .
For example, while in the past people searched for things like “recommendations for damaged hair shampoo,” now it is common to make queries in the form of sentences longer than 23 words , such as “what shampoo is good to repair hair damaged after dyeing?”
In other words, the search paradigm has shifted to a structure where AI finds a suitable solution (product/brand) when you describe the “Current Situation (CEP).”
Ultimately, the traditional brand strategy question of “which CEP should we connect our brand to?” valid in the AI era. What has changed is that we must now implant that “link” not only in the minds of human consumers but also in the knowledge base of AI .
1. Integration of the Search Journey: Context Has Become More Important Than Channels
Traditionally, we have established a channel strategy based on the stages of 'Awareness → Consideration → Purchase.' Customers become aware through Google ads, understand the utility of the service via blogs, and explore features in detail on various sub-pages of the website before signing up or leaving a purchase inquiry.
However, AI compresses this journey into a single conversation.
User: “Recommend a CRM suitable for our company (a 30-person startup). Tell me the price range and show me implementation examples.”
AI: “If you are a startup with 30 employees, a CRM that is easy to use, scalable, and low-cost,. Here are 3 key recommended options…”
Now, ' in what context we are mentioned' has become more important than channel strategy
2. Hyper-personalized context: Must occupy more specific situations
AI search reflects the user's very specific situation.
General Search vs. AI Search:
AI Search: “Good collaboration tool” vs. AI Search: “Time difference-aware collaboration tool for global teams with frequent asynchronous collaboration”
General Search: “CRM Program” vs. AI Search: “CRM suitable for a manufacturing company of 100 employees, compatible with existing ERP”
Potential customers now search like this: 'A collaboration tool usable by a marketing team of 30, integrated with Notion, supporting Korean, and with a monthly budget of under 1 million won'
If your brand isn't mentioned in this specific context, you've already fallen behind in the competition.
3. Criteria for Defining a Customer: No Customer Without CEP
Failing to define CEP is the same as failing to define the customer.
Ambiguous Target : Small and Medium-sized Business Marketer
An in-house marketer working for
a CEP-based team of 3–10 people who needs to achieve SEO results despite limited time for content creation. Someone working at a company with growth potential but a limited outsourcing budget
The latter is precisely one of the CEPs defined by Elifunt. Defining it specifically like this makes the content, message, and channel selection clear.
4. LLM is a 'context matching' engine
a recent AirOps study (2025), 85% of brand exposure in AI search comes from third-party content.
In the North American market, how our brand is mentioned on external channels such as Reddit, Quora, review sites, news articles, blogs, YouTube, and industry forums is important, and LLM learns signals from across the web indicating, “This brand is suitable for this situation.”
Ahrefs Evolve speakers add the following regarding this phenomenon.
“LLM prefers popular brands. This is because LLM cites trusted brands, and that trust is built in the same way humans build trust—through reputation.” Kevin Indig
“Brand is the new SEO. The best marketers are training people and the internet to associate their brands with category terms.” – Carrie Rose
How to Specifically Define CEP: The 7W Framework
Then, how can CEP be defined? The person who emphasizes the importance of CEP the most in Korea is CEO Park Se-yong of Ascent Korea . CEO Park when defining CEP the 7W frameworksuggests using
1. WHY - Why are people looking for our category?
2. WHEN - When do they think of our category?
3. WHERE - Where does this need arise?
4. WHILE - What are they doing when they feel the need?
5. WITH/FOR WHOM - With whom? For whom?
6. WITH/FOR WHAT - With what product/service?
7. HOW (feeling) - In what emotional state?
By elaborating on the seven customer situations, we can define our brand's positioning strategy more precisely.
I will introduce an example of the CEP definition case from Elifunt's client , Roobook
Lubook is a hotel sales solution. The potential customers are general managers of 5-star hotels and sales representatives for rooms and banquet halls. By applying the 7Ws, this can be concretized as follows.
- WHY: When you want to increase hotel room sales efficiency
- WHEN: Before peak season, when there are many vacant rooms
- WHERE: At the hotel sales team office
- WHILE: While analyzing sales data
- WITH WHOM: Sales Team Manager, Hotel Management
- WITH WHAT: Integration with existing PMS
- HOW: Sales pressure, manual labor fatigue
By making it concrete in this way, Lubook was able to clearly position itself as a “partner that structures sales pipelines.”.
→ See more of Lubook's content growth marketing success stories
An era where marketing performance improves the more clearly you target
When I ask people to precisely define a customer's problem situation, many ask this question.
If you define CEP so specifically, wouldn't the target actually become smaller and thus reduce performance?
No! On the contrary.
Specificity is scalability.
The moment you say “a solution for everyone,” AI will not recommend you in any situation.
If you cannot identify customers on the web, for what purpose are you providing information? In the age of AI, occupying a specific category and establishing expertise within itis the way to build a brand's minimal worldview.
Expanding Category Entry Point (CEP)
Strategy:
- Clearly occupy 1-2 key CEPs
- Expand to 3-5 adjacent CEPs
- Build a network with 5-10 expandable CEPs
Applying this to the case of Class101 B2B, another client of Elifunt, it is as follows
- Key CEP: “When You Need an Employee Welfare Program”
- Adjacent CEP: “When looking for team-building activities”
- Extended CEP: “When You Want to Increase Satisfaction Among MZ Generation Employees”
By building a multi-layered CEP in this way, you can create various entry points while keeping each message specific and sharp
Redefining Channel Strategy in the GEO Era
What is GEO (Generative Engine Optimization)?
As such, GEO is a strategy that optimizes brands to be exposed and accurately represented in generative AI engines (ChatGPT, Claude, Gemini, Perplexity, Google AI Overview, etc.).
The Key Difference Between SEO and GEO
| division | SEO | GEO |
|---|---|---|
| target | Search result ranking | Include in AI answer |
| measurement | Clicks, Rankings, Impressions | Mention, contextual accuracy |
| Content Location | Mainly our own site | The entire web |
| Optimization target | Keywords, Backlinks | Brand Entity, Context Signal |
| User behavior | Visiting multiple sites | Get integrated answers in one place |
Google's Danny Sullivan says this.
AI Overview still relies on traditional SEO signals. The fundamentals of creating trustworthy, fresh, and structured content have not changed
GEO is an extension of SEO, not a replacement, Search Everywhere Optimization
There was also discussion regarding how to define the AI era. The Ahrefs team redefined SEO Search Everywhere Optimization. This is because, in reality, consumers no longer search only on Google. They tend to search on channels that match their User Event Error (CEP).
- Search for product reviews on TikTok (40% of Gen Z)
- Check out real user opinions on Reddit
- Watch tutorial on YouTube
- Explore B2B solutions on LinkedIn
- Integrated questions for AI chatbots
Channel-specific CEP Mapping Strategy
If specific channels emerge within each customer's different CEP, the content strategy should also be tailored to those channels and messages. For instance, a company developing collaboration tools can cite these channel and content strategies as examples.
| ChannelNal | Google/AI Overview | online community | YouTube | Review site | |
|---|---|---|---|---|---|
| CEP | Collaboration Tool Comparison, Ways to Improve Team Productivity | Real user reviews, what are the downsides? | Actual usage, setup method | Tools tailored to our company, ROI verification | Verified solution |
| Content | Comparison Guide, Best Practices | Sincere participation, sharing customer success stories | Tutorials, customer interviews | Thought Leadership, Data-Driven Insight Content | Inducing customer success stories |
| reference | Reddit in North America, and Cliang, DC Inside, etc. in Korea | In North America, platforms like G2 and Capterra are used; in Korea, in-platform reviews are used, etc. |
GEO cannot do it alone.
PR, marketing, tech, community managers, etc., must become one team.
SEO and GEO cannot be done alone. Roles scattered across different channels must come together to organize channels and plan content toward a common goal. The traditional role of "marketer" may even be renamed. ( Bryan Casey, a speaker at Ahrefs Evolve, even proposed the role of "General Manager of Inbound.")
- PR: Media coverage, news coverage
- Content Marketing: Blogs, Guides, Resources
- Social Media: Community engagement, leading conversations
- Technical/Engineer: Technical optimization, structured data
- Brand Marketing: Building Brand Awareness and Trust
- Community Management: Management of external communities, such as online communities and open chat rooms
AI is re-examining the basics of marketing
If you have identified what is important in the GEO era in Part 1, it is now time to put it into action.
Part 2: 5 Practical Strategies to Make AI Recognize Your Brandintroduces specific implementation methods.
✅ How to Make AI Accurately Understand Your Brand
✅ How to be discovered first in the context the customer wants
✅ How to maintain a consistent brand message across the web
Elifunt is turning this change into an opportunity together with its clients. Is your brand also ready to prepare for 2026?
When do customers think of our category?
At that moment, does our brand come to mind?
Do we truly understand the customer's problems?
If you can answer these questions, you have already started GEO. The Elifunt Team will be with you to develop strategies for acquiring customers even in the AI era!
