Decision Graph

Quick Definition

The Decision Graph is a network data structure that maps all possible paths a customer can take during their journey. Unlike a static funnel, it uses nodes to represent decision states and edges to represent the probability of transition between these states, feeding the Conversational Forecast.

In Simple Language

Imagine that your customer's journey is not a straight-line slide, but a railway network with various detours, returns, and stops. The Decision Graph is the complete map of this network. It understands that a customer can skip steps or go back, and it calculates the chance of them closing a deal based on each interaction ("turn") that happens in the conversation.

Why This Concept Exists

The concept was born to solve the failure of the Linear Sales Funnel. In the real world, customers do not progress sequentially (Discovery -> Consideration -> Decision). They hesitate, compare, have doubts again after receiving the price, and seek social proof at unexpected moments. The Decision Graph names the actual complexity of the human journey, allowing the company to manage statistical uncertainty through Probabilidade Térmica (pT).

Educational Metaphor

Think of a Chess Grandmaster observing the board. He doesn't just see the piece that moved; he visualizes a "tree of possibilities." Each move opens new paths and closes others. The Decision Graph is the "living board" of MCI, where each customer message is a move that alters the value of all future victory (conversion) probabilities.

Practical Example

A customer contacts via WhatsApp interested in management software.

1. Initial Node (Trigger): They ask for the price.
2. Transition Edge: The agent (AI or human) provides the price and asks about the team size (Context).
3. Movement in the Graph: If the customer answers "We are 50 people," they move to the Exploration/Scholar node with a high pT (Probabilidade Térmica). If they respond "I'm just looking," the graph repositions them in the No Need/Tourist node, reducing the probability of an immediate sale and triggering a specific nurturing sequence for that state.

Anti-example

The Decision Graph is not a Service Flowchart (IVR). A flowchart is rigid: "If X, then Y." The Decision Graph is probabilistic and dynamic: it accepts that the customer can transition between any decision state at any time without the system "breaking" or forcing the user back to the start of a menu.

How it Appears in Operation

• Opportunity Stock Visibility: You know exactly how many customers are in the "Comparison State" and the real chance of them moving to "Purchase."
• AI Contextualization: Generative AI consults the graph to know which node the customer is in before choosing the tone of voice and content for the next message.
• Friction Reduction: The system does not ask something the customer has already answered in a previous node (combating Operational Amnesia).

How to Apply in MCI

In MCI, the Decision Graph is the brain that orchestrates the Dynamic Journey.

• Convenience (8Cs): Allows the customer to dictate the pace of the conversation.
• Content (8Cs): Delivers the exact information for the current Decision State (e.g., do not send a contract to someone still in the Exploration state).
• Guardião do Ciclo: The Graph provides the data so the Guardião knows when to intervene to prevent a customer from getting "stuck" in an indecision node.

Related Metrics

• pT (Probabilidade Térmica): The percentage chance of conversion at each node.
• Node Dwell Time: How long the customer takes to transition between states.
• Conversation Score: The quality of the interaction that drives state changes in the graph.
• Conversational Forecast: The revenue forecast based on the sum of pTs of all active customers in the graph.

Diagnostic Questions

• Do we know how many customers gave up on the purchase at the moment we asked for their ID/Tax info (Closing Node)?
• Does our sales force treat all leads with the same priority or focus on those with higher Probabilidade Térmica in the graph?
• Does the service AI know how to differentiate a customer in the "Exploration" state from one in "Comparison"?
• Where are the "black holes" in our graph, where customers enter and never leave?

Related Terms

• Dynamic Journey: The actual path executed over the graph.
• Probabilidade Térmica (pT): The numerical weight of the graph's edges.
• Decision States: The points (nodes) of the graph.
• Operational Amnesia: What happens when the graph has no memory of previous turns.

Executive Mode

For the C-Level, the Decision Graph represents the transition from "intuition" to "conversational data science." It transforms subjective conversations into a predictable financial asset. By visualizing the graph, the CRO (Chief Revenue Officer) can see the pipeline not as a raw volume of leads, but as a network of probabilities that allows for much more accurate forecasting and optimized resource allocation (CAC).

Operational Mode

For managers, the Decision Graph is the definitive diagnostic tool. If conversion rates drop, the manager doesn't just look at the "end of the funnel"; they identify at which node transition the bottleneck is occurring. This allows for team training or adjusting the AI prompt specifically for the "pain" of that decision state.

Technical Mode

Architecturally, the Decision Graph is a Directed Graph where each node maintains the context state and edges are probability functions based on intent vectors (embeddings). With each new user input, the inference engine reclassifies the current state and updates the global pT, triggering webhooks or automation actions depending on the user's new position in the graph topology.

Playful Mode

Imagine a smart GPS. In the traditional funnel, if you miss a turn, the GPS says "Return to start." In the Decision Graph (the MCI GPS), if you miss a turn, it recalculates the route instantly. It knows where you want to go (the Purchase) and knows all the shortcuts and alternative paths, no matter how many turns you take.

Executive Summary

The Decision Graph is the evolution beyond the traditional marketing funnel. It maps the non-linear customer journey through a network of states and probabilities, transforming the flow of conversations into a predictable, auditable, and highly efficient data structure for closing deals at scale.