Ops Usage

Operations (ops) are the unit of measurement for activity in Wexio. Each action that processes or sends messages consumes ops from your monthly allocation.

What Are Operations?

Your plan includes a monthly ops allocation that resets at the start of each billing cycle. Every message sent or received, every flow card executed, and every CRM sync consumes ops from this allocation.

Message Operations

Action	Ops Cost
Receive a message	1
Send a message (manual)	1
Send a message (bot/flow)	1
Send a message (AI)	1

Flow Card Operations

When a flow executes, each card that sends a message or performs an action consumes ops.

Basic Cards - 1 op each

Card	Description
Text	Send a text message
Button	Send a message with buttons
Media	Send a photo or video
Audio	Send an audio file
Document	Send a document
Question	Ask a question and wait for response
Flow	Trigger another flow
Update Contact	Update a contact profile field
WhatsApp Template	Send an approved WhatsApp template
WhatsApp Catalog	Send a catalog
WhatsApp Product	Send a single product
WhatsApp List Message	Send a list message
Assign Operator	Assign the conversation to an operator
Outbound Webhook Event	Emit a custom `flow.*` webhook event
Wait	Pause the flow
Reply to Comment	Reply to a social post comment

Advanced Cards - 2 ops each

Card	Description
Media Button	Message with media + buttons
Audio Button	Audio with buttons
Document Button	Document with buttons
AI Text	AI-generated dynamic content
AI Classification	Classify a message into one of your categories
AI Agent	Run an autonomous agent step
AI Text to Speech	Generate spoken audio
Conditional	Branching logic evaluation
Fetch	External API request
Wait Timeout	Wait for a reply with a timeout branch

CRM Card - 1 op per field

The CRM Card updates fields in your connected CRM during flow execution. Each field updated costs 1 op.

Fields Updated	Ops Cost
1 field	1
3 fields	3
5 fields	5

No-Cost Cards - 0 ops

Card	Description
Note	Internal note (not sent to contact)
Ghost	Placeholder for flow structure
Order	Order handling - no message is sent

Cards that send nothing cost nothing. That is the rule behind this table: an internal note, a structural placeholder and an order step consume no ops, while every card that puts a message on a channel costs at least one.

CRM Sync Operations

CRM import and export operations consume ops based on the number of contacts and field mappings.

Operation	Formula
Export to CRM	contacts × (0.2 × fields)
Import from CRM	contacts × (0.3 × fields)

Import costs more than export because it requires validation, deduplication, and field transformation. Only successfully synced contacts are charged - failed records don't consume ops.

Export Examples

Contacts	Field Mappings	Ops Cost
10	30	60
100	10	200
500	5	500

Import Examples

Contacts	Field Mappings	Ops Cost
100	10	300
500	20	3,000
1,000	30	9,000

AI Ops

All AI features - both AI Auto-Reply and AI Copilot - use a unified token-based billing model. AI usage is tracked as AI Ops and counted against your plan's AI limit.

How AI Ops Are Calculated

AI ops consumed = tokens used × token cost rate

Token cost rates vary by model tier:

Model Tier	Rate (per token)	1,000 tokens =	Example Models
Economy	0.0004	0.4 AI ops	GPT-4.1 Mini, GPT-4.1 Nano
Standard	0.0008	0.8 AI ops	GPT-4.1, Claude Sonnet
Premium	0.0015	1.5 AI ops	GPT-4.5, O1, Claude Opus

Example Costs Per Action

Action	Model Tier	~Tokens	AI Ops
Fix grammar / simplify text	Economy	~300	0.12
Suggest reply with context	Economy	~800	0.32
Suggest reply with context	Standard	~800	0.64
Summarize conversation	Economy	~1,500	0.6
AI auto-reply (with history)	Economy	~1,200	0.48
Custom prompt (long)	Premium	~2,000	3.0

AI Limits by Plan

AI limits scale with your plan's ops limit:

Plan	Ops	AI Limit
Free	100	0
Standard 10K	10,000	1,000
Standard 80K	80,000	8,000
Pro 10K	10,000	1,000
Pro 1M	1,000,000	100,000
Pro 2M	2,000,000	200,000
Enterprise	Custom	Custom

What Counts as AI

Both auto-reply and copilot share the same AI ops pool:

AI Auto-Reply - Bot sends AI-generated responses to contacts
AI Copilot - Operators use suggest reply, translate, summarize, tone change, etc.

Assistants using your own API key (BYOK) are not counted against your AI limit - you pay the provider directly.

Understanding Ops Consumption

Flow Execution Example

1. Text Card: "Welcome!"             → 1 op
2. Button Card: "Choose an option"   → 1 op
3. Conditional Card: Check selection → 2 ops
4. Text Card: "You selected X"      → 1 op
                              Total: 5 ops

AI Conversation Example

1. Contact sends message  → 1 op (receive)
2. AI generates response  → 1 op (send) + AI ops
                    Total: 2 ops + AI ops

Important Notes

Ops are charged when messages are sent - if a Question Card is waiting for input, ops are charged when the question is displayed, not when the response is processed.
Re-processing doesn't double-charge - if a contact provides invalid input and the same question is shown again, it's only charged once per actual message sent.
Failed operations - if a card fails to execute, ops are not charged for the failure.

Ops Limits by Plan

Plan	Monthly Ops
Free	100
Standard	10,000 – 80,000
Pro	10,000 – 3,000,000
Enterprise	Custom

What Happens When You Run Out

It depends on your plan.

Free plan - a hard stop.

Flows stop executing
AI responses are disabled
Manual inbox messages still work (with a warning)
Upgrade, or wait for the cycle to reset

Paid plans - usage credit, if you opt in.

Standard, Pro and Enterprise can switch on overage: instead of stopping when the included ops run out, work continues and the extra is billed as usage credit.


You set a budget	A spending cap for the period. Work continues until the budget is reached, then stops.
Rates follow your plan	The per-unit rate is derived from your plan's price and its included limits - a bigger plan has a cheaper marginal unit.
Ops and AI count separately	Each is measured against its own limit, so exceeding one does not spend the other's allowance.
Minimum budget	There is a minimum budget, and a small headroom allowance above it.

Overage is opt-in and only on paid tiers. If you have not enabled it, a paid plan behaves like the hard stop above. Set the budget deliberately - it is the only thing between a runaway flow and a bill.

Do Unused Ops Roll Over?

No - ops reset at the start of each billing cycle.

Monitoring Usage

You can track your ops and AI ops consumption in two places:

Click the Usage button in the bottom-right corner of the inbox to see a live summary of your current usage:

Operations - current ops used out of your plan limit (e.g. 39.50 / 10K)
AI Operations - current AI ops used out of your AI limit (e.g. 64.84 / 1K)

This updates in real time as messages are sent and received.

Settings - Billing Page

Go to Settings → Billing for a full overview of your subscription and usage:

Plan name and status (Active)
Renewal date and days remaining
Operations progress bar - used vs. allocation with percentage
AI Operations progress bar - used vs. allocation with percentage
Change Plan or Cancel Subscription actions

Optimizing Ops Usage

Combine Messages

Use one message with multiple lines instead of multiple Text Cards.

❌ Inefficient (4 ops):
  Text Card → "Hello"
  Text Card → "How can I help?"
  Text Card → "Please choose:"
  Button Card → [Options]

✅ Efficient (1 op):
  Button Card → "Hello! How can I help?\nPlease choose:" + [Options]

Other Tips

Use conditional logic wisely - place Conditional Cards strategically to avoid unnecessary message sends
Optimize AI usage - AI responses consume both ops and AI ops; use them where they add the most value
Choose the right AI model tier - Economy models cost significantly less AI ops than Premium models for routine tasks