How we identify red flags
Outdoo AI uses a proprietary machine learning model to detect red flags in prospect emails. Here is how it works.
Contextual understanding of language
The model is built on a context-aware understanding of human language. It was pre-trained using millions of text samples from diverse online sources.
Specific red flag training
The model was trained with hundreds of real-world examples from sales emails across multiple industries. Examples of red flag phrases include:
- "Although the team feels strongly about your products, they are unsure about the budget aspect."
- "I have talked to the decision-makers, and they haven't acted interested at this time."
- "We are cutting our budgets across the business and can't commit to additional costs."
- "This project has dropped down the priority list."
Comprehensive email scanning
The model evaluates the full context of prospect emails to identify red flags and surface relevant insights.
Does Outdoo AI rely on keywords?
No. Outdoo AI does not depend on keywords alone. It analyzes the entire context of an email to assess whether it presents a red flag. For instance:
- Words like "unsure" or "priority" on their own do not necessarily indicate risk.
- The system interprets the intent behind the full sentence, so it can recognize risk even when phrased in unfamiliar ways.
How accurate is red flag detection?
Outdoo AI's red flag detector achieves 90% precision. That means 9 out of 10 flagged risks are accurately identified based on test data. Occasionally, up to 10% of flagged risks may not align with expert judgments, but the model is continuously updated to improve accuracy by incorporating new examples.