AI chat assistants are not getting used the way in which many marketers expect. Instead of acting as shopping tools or serps, they function primarily as support systems for cognitive tasks — writing, planning, analyzing and learning, according to a report by AI search engine optimisation agency Dejan. In 2026, optimizing for AI means understanding how users actually engage with assistants, not just assuming industrial behavior.
Dan Petrovic, director of Dejan, conducted an intensive evaluation of three.9 million conversational turns — covering 613 million words and 4.4 billion characters — to uncover patterns in real-world AI usage. The results challenge many assumptions in AI content strategy.
Usage patterns: most chats are short and task-oriented
- Median chat length: 2 turns (a single query and answer)
- Median session length: 430 words
- Over 80% of chats are under 1,000 words
- Only 4.2% of sessions exceed 2,500 words, typically involving editing, summarization, tutoring, coding, or data evaluation
- Mean session length: 732 words, skewed by long document submissions
- Assistant output volume: roughly 1.5 times that of user input
Petrovic noted that user contributions typically account for just 16–17% of the session, meaning the assistant generates the majority of the content.
Dig deeper: What happens when nobody clicks anymore
Intent breakdown: industrial use is proscribed
Petrovic analyzed 24,259 classified sessions across 42 intent categories. He found that 64.6% of sessions had no industrial intent. Users primarily engaged AI for tasks like:
- Brainstorming (7.7%)
- Planning (6.5%)
- Emotional support or conversation (6.2%)
- Analysis (5.7%)
- Learning (4.7%)
- Text transformation, including summaries and translations (4.6%)
- Content creation (3.9%)
Even among the many 35.4% of sessions showing industrial intent, the bulk were early within the funnel:
- Awareness stage: 10%
- Consideration: 8.5%
- Discovery and decision support: 6.9% combined
- Transactional support: 4.8%
- Post-purchase support: 5.1%, including troubleshooting and usage guidance
Key insight: AI conversations are cognitive workflows, not queries
Many marketers and SEOs are optimizing content with a search-first mindset — assuming AI chats mimic keyword-based queries. Petrovic’s findings suggest otherwise. AI assistants are more often used to support multi-step tasks, moderately than making immediate purchases.
“AI chat use is overwhelmingly non-commercial,” Petrovic stated. “Users treat assistants as co-pilots — not sales reps.”
Implications for AI content strategy
Marketers should adjust their approach to AI optimization by:
- Prioritizing content that supports awareness and early-funnel exploration
- Creating structured, high-context information that agents in longer workflows can reuse
- Avoiding overemphasis on transactional keywords in AI-facing content
Bottom line: The way forward for AI content visibility isn’t about gaming intent — it’s about meeting it. Assistants have gotten tools for cognition, not just conversion. If your content helps users learn, plan, write, or evaluate, it’s much more likely to show up in real-world AI usage.
The report. How do people use AI assistants?
The post Why AI content strategies need to focus on tasks not transactions appeared first on MarTech.
Read the total article here
