Your help system gives the same answer to everyone.
The new hire gets the same explanation as the 10-year veteran.
One leaves confused. The other leaves annoyed.
Same question. Different knowledge levels. Needs a different answer.
INTERMEDIATE - Builds on classification to personalize every interaction.
When a new hire asks 'how do I process a refund?' they need the full walkthrough. Step by step. Every detail. When a senior team member asks the same question, they just forgot which button to click. Same question, completely different needs.
Awareness level detection reads the signals: the words they use, the context they provide, what they've asked before. It categorizes the person as a beginner, intermediate, or expert on this particular topic. Not globally. Topic by topic.
This isn't about being condescending to beginners or cryptic to experts. It's about respecting everyone's time. The beginner gets the foundation they need. The expert gets straight to the answer.
Get it wrong and beginners feel lost while experts feel patronized. Get it right and everyone gets exactly what they need.
Awareness detection solves a universal problem: how do you give the right amount of detail when different people need different depths?
Assess the knowledge level first. Then calibrate the response. The same information gets packaged differently based on who is receiving it.
Click on any of these sample messages to see how the system detects knowledge level from the way someone asks.
Click a message to analyze
What words do they use?
Beginners say 'the thing that sends emails automatically.' Experts say 'the SMTP trigger.' The terminology someone uses reveals how much they already know. Detecting jargon vs. plain language is the first signal.
Works from the very first message
Can misjudge if someone uses borrowed terminology
How much do they already explain?
Beginners ask 'how do I do X?' Experts ask 'when I tried X with Y configuration, I got Z error.' The amount of context someone provides shows their familiarity. More specific context means more expertise.
Highly accurate for technical topics
Talkative beginners can look like experts
What have they asked before?
If they asked five beginner questions last week, they probably still need beginner-level answers this week. If they've never asked about a topic before, start at the foundation regardless of their general expertise.
Gets more accurate over time
Requires session or user tracking
Awareness detection sits between understanding what someone asked and crafting the right response. It takes the classified intent and adds the context of who is asking.
Hover over any component to see what it does and why it's neededTap any component to see what it does and why it's needed
Animated lines show direct connections · Hover for detailsTap for details · Click to learn more
Someone might be an expert at your core process but a complete beginner at the new tool you just added. Treating them as a 'general expert' means they'll get cryptic instructions for something they've never seen.
Instead: Assess awareness topic by topic, not person by person.
A new team member might have picked up jargon from their previous job without understanding what it means in your context. Matching terminology doesn't mean matching understanding.
Instead: Combine vocabulary with context depth and ask clarifying questions when signals conflict.
Someone starts as a beginner but learns fast. If your system keeps giving them beginner explanations after they've clearly progressed, they'll get frustrated.
Instead: Let users escalate or simplify explicitly, and watch for signals they need a different level.
You've learned how to assess what someone already knows before responding. The natural next step is using that assessment to actually personalize the response.