47 Visitors. 3 Ready to Buy. Do You Know Which?

Fred does.

Someone asked about your pricing, mentioned they needed it done before the end of the month, and wanted to know if you were licensed in their county. That conversation happened at 9 AM. By the time anyone looked at it, they had already called someone else.

Fred does not treat them the same.

Every conversation Fred has gets scored. Not by a human reviewing a chat log at the end of the day. By a system that read the conversation in real time, identified what signals were present, and produced a number that tells you exactly where that person was in their decision.

Zero to 100. And a breakdown of why.

What Fred Is Actually Listening For

The score is not a gut call. It is built from real signals in the conversation itself.

Buying language. Did the person use words that indicate purchase intent — pricing, quote, schedule, book, cost, how soon, what does it take to get started. Not did they visit your pricing page. Did they say those things to Fred.

Urgency. Did they mention a deadline. A problem that needs solving now. Language that suggests they are not researching for next quarter. They need someone this week.

Engagement depth. How much did they put into the conversation. A person who asks one question and reads the reply is different from a person who goes back and forth six times, follows up on your answer, asks a clarifying question. Depth is signal.

Whether Fred reached the right content. This one matters more than most people expect. If someone is asking about your process and Fred surfaces your process page, that is a match. If Fred cannot find relevant content and the conversation stalls, that suppresses the score. Not because the person was not interested, but because the conversation did not reach the information that would have moved them.

That last one is a content audit hiding inside a lead score. A pattern of high-intent visitors with low scores is not a traffic problem. It is a knowledge base problem. Fred is telling you something.

Two Passes on Every Conversation

The scoring runs in two layers.

The first layer is fast and deterministic. It runs the signals, applies weights, accounts for how many signals fired together because one strong signal in a thin conversation is less meaningful than four signals in a conversation with real depth. Math is honest about what it knows and what it doesn’t know.

The second layer reads the last stretch of the conversation and makes a call: was this person ready to act, still evaluating, or just browsing. That assessment adjusts the score. When the two layers agree, the score is telling you something real.

When the LLM call fails for any reason, the first layer result stands unchanged. The system does not guess. It reports what it can verify.

What Happens With the Score

The score and the breakdown go into your CRM. High-intent conversations route to high-priority queues. The rep who calls back the 87 is not working from the same list as the rep calling the 23.

You decide the thresholds. Fred decides who crossed them.

Industry matters here too. An insurance inquiry that includes coverage, policy, and renewal language scores differently than a general service inquiry. A home services conversation that includes urgency, licensing questions, and a timeline scores differently than someone asking about your hours. The signals are calibrated to what buying actually looks like in your business.

The Conversation You Were Not In

You were on a job. Or it was 9 PM. Or your team was slammed and the chat came in and someone marked it followed up when they sent a one-line reply from a text automation asking how you could help or to book a call. Friction.

Fred was in that conversation. Fred read every message, scored the intent, flagged the urgency, and sent the payload to your CRM before you knew the conversation happened.

You come in the next morning and the list is sorted. The people who are ready to act are at the top. The tire kickers are not gone. They are just not pretending to be something they are not.

That is what a pipeline scored in real-time looks like. Not more leads. Better information about the leads you already have.

Figure out how many conversations your business has in a month. How many contact forms get filled out. Figure out how many of those you actually follow up on with real speed and a strong idea of why they visited your site in the first place. The gap between those two numbers is what Fred’s lead scoring is closing.