Hardcoded Message Workflows

Hardcoded Message Workflows allow you to define fully deterministic conversation paths inside a chatbot, without relying on AI-generated responses for every step. Instead of asking the model to “figure out” what to say next, you explicitly define what message is sent, when, and why.

These workflows are rule-driven, predictable, and intentional.

---

What a Hardcoded Message Workflow Is #

A hardcoded workflow is a predefined conversation flow made up of:

• triggers (when the workflow starts)
• fixed messages (what is sent)
• actions (what happens next)
• optional AI involvement (if and when allowed)

The key idea:
👉 The chatbot follows your script, not the model’s creativity.

---

How Hardcoded Workflows Differ from AI Chat #

AI chat:

• generates responses dynamically
• can vary phrasing
• may hallucinate
• depends on model behavior

Hardcoded workflows:

• send exact messages you define
• behave the same every time
• do not hallucinate
• are fully predictable

They can coexist in the same chatbot.

---

Where Hardcoded Workflows Are Used #

Hardcoded message workflows are configured inside the chatbot configuration, usually alongside triggers and actions.

They are evaluated:

• during conversations
• after trigger conditions are met
• before or instead of AI responses

They do not affect frontend placement or UI.

---

Typical Workflow Structure #

A workflow usually consists of:

1. Trigger
   Example:
   • conversation start
   • user sends a keyword
   • user is logged in
   • user is on a specific page
2. Hardcoded Message
   Example:
   • “Hi! How can I help you today?”
   • “Please provide your email to continue.”
3. Action (optional)
   Example:
   • wait for user input
   • collect lead data
   • append system instructions
   • end conversation
   • hand off to AI
4. Next Step
   • another hardcoded message
   • switch back to AI chat
   • trigger another workflow

---

Common Use Cases #

Hardcoded workflows are ideal for:

• welcome flows
• onboarding sequences
• GDPR or consent notices
• lead qualification
• data collection (email, name, phone)
• FAQ-style fixed answers
• routing users to the right bot
• escalation to human support
• enforcing rules the AI must not break

Any situation where consistency matters more than creativity.

---

Combining Hardcoded Messages with AI #

Workflows can:

• completely replace AI responses for a period
• temporarily override AI behavior
• inject instructions into the AI prompt
• hand control back to the AI after completion

Example pattern:

• workflow asks qualifying questions
• once complete, AI resumes with enriched context

This allows structured flows without losing flexibility.

---

Deterministic by Design #

Hardcoded messages:

• are sent exactly as written
• are not rewritten by the AI
• are not “improved” by the model
• are not influenced by temperature or creativity

If the message changes, it’s because you edited it, not because the AI decided to.

---

Scope and Isolation #

Hardcoded workflows:

• belong to a single chatbot
• do not affect other chatbots
• apply only when their triggers fire
• do not persist beyond their defined scope

They do not globally change chatbot behavior unless designed to.

---

Priority and Ordering #

If multiple workflows could match:

• evaluation order matters
• earlier workflows may block later ones
• misconfigured workflows can conflict

Design workflows carefully and keep them simple.

---

What Hardcoded Workflows Do Not Do #

They do not:

• train the AI
• change the model
• modify personas permanently
• override placement rules
• bypass consent or privacy settings
• automatically understand user intent

They execute logic, not reasoning.

---

Performance and Cost #

Hardcoded messages:

• do not consume AI tokens
• execute instantly
• reduce API usage
• improve response reliability

They are ideal for high-traffic or cost-sensitive flows.

---

Common Mistakes #

• overusing workflows for everything
• building long, rigid flows that frustrate users
• forgetting to hand control back to AI
• overlapping triggers across workflows
• mixing business logic and conversational logic excessively

Use workflows where structure is needed, not everywhere.

---

Best Practices #

• keep workflows short and focused
• clearly separate AI chat and scripted flows
• document workflow intent
• test with real users
• use workflows for enforcement, AI for conversation
• prefer multiple small workflows over one large one

---

Summary #

Hardcoded Message Workflows let you define exact chatbot behavior at specific moments, using fixed messages and deterministic logic instead of AI-generated replies. They are predictable, fast, cost-efficient, and ideal for onboarding, lead collection, compliance, and routing. Used alongside AI chat, they give you precise control without sacrificing flexibility.