B2B customers increasingly receive their first explanation of a problem, market and suppliers directly from AI. They may therefore reach a company website only once they are already much better informed and are mainly looking for confirmation, evidence or the next step. Marketing thus gains two different audiences for its content: people and machines. The same page may not be optimal for both.
Traditional content marketing assumed that the website educated a visitor gradually. A person arrived from a search engine, read introductory content, then more detailed material and eventually reached a product or contact point. Generative AI splits this journey.
Roel Haanappel of SUSE describes an experience in which websites now serve two audiences with different needs. AI systems need enough structured, factual and interconnected content from which they can construct answers. A person who eventually arrives at the website after a conversation with AI, however, often already knows the basic information and mainly needs confirmation, evidence and a clear next step.
This changes traditional search-engine optimization. The author warns that optimization for generative systems cannot simply be understood as a new SEO in which keywords are replaced by static AI queries. A conversation with a model changes according to previous answers and the customer’s circumstances.
Topical authority is more relevant. A company needs to create an interconnected set of product pages, case studies, expert articles, research and external mentions that together give the system enough context about a particular subject.
At the same time, the significance of the website for human visitors changes. If a visitor arrives much later in the buying process, they do not need another general explanation. They may need a case study, a specific difference between alternatives, a reference, a price or a way to contact a salesperson.
Haanappel says that visitors arriving directly from large-language-model environments show substantially higher engagement at his company than ordinary organic traffic. He also points to a possible next stage: in the future, some content may be supplied directly to systems in structured form through interfaces, while the website remains optimized mainly for humans.
For marketing, this means changing content planning. It is not enough to create more pages for individual queries. More important is systematically building credible evidence around a small number of strategic topics.
Content that AI cannot easily replace by synthesizing other sources becomes particularly valuable: original research, specific numbers from case studies, expert experience and information based on direct access to customers.
The website therefore does not have to stop being important. Its role may, however, shift from the main place of education to the place where a well-informed customer verifies whether it makes sense to continue with the supplier.
KEY TERMS
- Topical authority: The strength and credibility of an interconnected body of content around a specific area of expertise.
- Optimization for generative systems: Adapting content so that AI can interpret it correctly and use it in answers.
- Bottom-of-funnel content: Material for people who already know the offer and need confirmation before making a decision.
- Structured content source: Central content that can be delivered through different interfaces, including websites and AI systems.