Understanding Lenses
Understanding "Lenses" for Machine Learning Categorization
🧠 What Are Lenses?
Lenses are structured semantic perspectives that help categorize a topic based on how the user or customer sees it when they land on a webpage, search for a solution, or consume content.
LLMs aren’t just decoding topics — they’re interpreting user intent. So we need to match descriptions to the perspective a user has when they arrive at a certain page.
For example:
- A user searching for "Apollo.io" is likely trying to learn what the company does (Brand lens)
- A user searching for "cold email deliverability" is trying to solve a problem (Solution lens)
- A user searching for "InboxRamp" is researching a product (Product lens)
By writing with the correct lens, we help the model:
- Match the intent behind the topic
- Find high-signal, relevant URLs
- Avoid noise from comparison pages, forums, or irrelevant media mentions
🔹 The 6 AudienceLab Lenses
Lens | Focus | Cold Email Software | Sales & Marketing CRM | Conversational AI Tools |
Brand | Company or named entity | Instantly.ai | Salesforce | Drift |
Product | Specific product or feature | Smart Inbox | HubSpot Sales Starter | Intercom Inbox |
Solution | The outcome the user wants | Cold email deliverability | Revenue pipeline visibility | 24/7 automated support |
Function | What the tool actually does | Domain warm-up rotation | Lead scoring & CAC payback tracking | Natural language intent detection |
Service | How the solution is delivered | Email outreach as-a-service | CRM onboarding and consulting | Bot-building services |
Event | Where users learn or connect | Cold Email Bootcamp | Dreamforce | Conversational AI Summit |
✅ Choose ONE lens per topic to minimize ambiguity and maximize precision in model behavior.
Did this answer your question?
😞
😐
🤩
Last updated on August 6, 2021