Your AI. Your data. Our sales operation.
For large operations with multiple teams and IT requirements. The models, keys, and data are yours. The orchestration, service, and qualification are ours, from the first hello to a lead that's ready for your team.
No charge per lead or handoff Usage cap set by you 7 days to withdraw, with a full refund
From your company
- The AI models
- The access keys
- The contract with the provider
- Where the AI processes the data
From AttendTo
- 24/7 service across channels
- BANT qualification and ideal profile
- Every reply checked
- Pipeline, CRM, scheduling, and handoff
Compatible with
the models your company already uses.
- OpenAI
- Anthropic
- Google Gemini
- Microsoft Azure
- Amazon Bedrock
- Meta Llama
- Your own server
Brands belong to their respective owners. Technical compatibility only, with no commercial affiliation.
INSIDE
How your AI
works in here.
Every lead message goes through the orchestration. Your models do the language work; AttendTo provides the context, checks the reply, and guides the lead all the way to the sale.
Your models
Company context, knowledge base, lead memory, reply checking, and rules.
- Intent
- BANT qualification
- Ideal profile
- Reply check
- Next best action
- Handoff
Where the sale happens
Personal data masked. Email, phone, CPF, and CNPJ are masked before every call.
Backup for every task. If one model fails, another takes over and the conversation continues.
Everything in the audit trail. Who changed what, when, with the before and after.
WHAT YOUR COMPANY GETS
Control for IT.
Predictability for finance.
Everything AttendTo does on a standard account, running on the models your company chose.
The models and keys are yours
Connect the models your company already subscribes to, or one that runs on your own server. The key is stored encrypted, and the screen shows only its first and last characters.
Data stays with your provider
The AI processes conversations with your company's own provider, with email, phone number, CPF, and CNPJ masked. You choose whether we store the text of the calls or only the numbers.
Costs without surprises
A fixed monthly fee with messages included, and no charge per lead, qualification, or handoff. Alerts at 80% and an overage cap, if you want one.
No lead left unanswered
If your models go down, the AttendTo AI replies in their place or the conversation goes to your team. You decide.
Each task on the right model
One model to chat, another to double-check, another to qualify. Plus rules by lead type: a lead that matches your ideal customer profile can go to the strongest model.
Audit trail and LGPD
Every change to a model, route, or rule is logged and can be exported to a spreadsheet. Consent, opt-out, and data deletion work the same as on any account.
Two-step verification
By default, only the account owners can change your own AI, and only with two-step verification turned on. The rest of the team can view it.
Model health at a glance
Response time, success rate, and calls for each model. An hourly check alerts you when one goes offline.
WHO IT'S FOR
The same conversation as Custom plans.
At a larger scale.
The specialist reviews your operation and tells you whether it calls for Enterprise. The proposal covers what your company needs, with no fixed package.
High volume
Tens of thousands of conversations per month, with alerts at 80% and an overage cap you define.
Multiple teams
Different teams, queues, and numbers in the same operation, with rules by lead type.
IT requirements
A model on your own server, two-step verification, and a full audit trail.
CAN BE INCLUDED IN THE PROPOSAL
What Enterprise adds.
- Dedicated support from our team
- 99.9% SLA in the contract
- Extra WhatsApp numbers
- Contract with your company's clauses
How does onboarding work?
Our team supports you from the start: we connect your models with you, run the connection test, define who handles each task, and take the operation live. The timeline depends on how many models and integrations your company brings, and it goes into the proposal along with the onboarding fee.
How is this different from Custom plans?
It's the same platform, and in both the AI runs on your company's models. Enterprise is for larger operations: it adds dedicated support, a contractual SLA, and custom clauses. Your specialist will recommend what makes sense for you.
Is there a minimum contract term?
Yes, 12 months. After the 7-day withdrawal period, ending the contract before the minimum term is arranged with our team.
What if I change my mind?
You have 7 days from the first payment to withdraw. During that period we refund everything you paid: onboarding fee, monthly fee, and overage (Brazilian Consumer Code, art. 49).
What happens if my models go down?
The task's backup model takes over instantly. If the backup also fails, the AttendTo AI replies in its place, as long as you keep the fallback turned on; those messages appear separately in Usage, at the rate stated in your proposal. With the fallback turned off, the conversation goes to your team. An hourly check alerts you when a model goes offline.
Who pays for AI usage?
Your company, directly with your provider, under the contract you already have. AttendTo does not resell AI usage: the monthly fee covers the orchestration and the entire sales operation. SMS, email, and WhatsApp message templates are still billed separately because they carry third-party costs.
What counts as a message?
Every reply the AI sends to a lead on WhatsApp, Instagram, or web chat. If you go over the messages included in your proposal, the overage follows the agreed rate, with alerts at 80% and when you pass the limit. If you set an overage cap, once it is reached the conversations go to your team.
How do security and my data work?
Keys are stored encrypted and never appear in full on screen. Before every text call to your models, email addresses, phone numbers, CPF, and CNPJ are masked. You choose whether we store the text of the calls or only the usage numbers. Each company sees only its own data, and changing your own AI requires two-step verification.
Can I use a model on my own server?
Yes. Besides the market providers, you can connect a model that runs on your own infrastructure, as long as it offers a standard chat API. Our team checks compatibility during onboarding.
Do voice and knowledge base search also use my models?
They do. Transcription of lead voice messages, the AI's voice, and knowledge base search all run on the models your company connects. If your provider doesn't offer one of these features, our team helps you connect one that does during onboarding. Without the fallback turned on, nothing runs on the AttendTo AI.
ENTERPRISE WITH YOUR OWN AI
Your models.
Our sales operation.
Tell us how your company uses AI today. A specialist shows how the operation would run on your models.
So is the lead.