> ## Documentation Index
> Fetch the complete documentation index at: https://docs.teamfollowup.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Use the public API base URL: https://api.teamfollowup.ai/api.
> Authenticate public API requests with Authorization: Bearer YOUR_API_KEY.
> Use product-owned terms: agents, calls, campaigns, contacts, phone numbers, billing, and configuration.

# What happens after the call

> Every call ends in an outcome. You decide what the campaign does next: send a text, add a label, book, or try again.

The call is done. The agent has decided **what happened** and written it down (see
[What your agent records](/agents/outcomes)). Now the campaign acts on it. This is
where a plain phone call turns into real follow-up.

## Outcomes are buckets

Think of every call dropping into a labelled bucket:

* **Booked** - they set a time.
* **Transferred** - they got handed to a person.
* **Callback** - they asked to be reached later.
* **Asked for info** - they wanted details, not a booking yet.
* **Needs a human** - a person should step in.
* **Opted out** - they asked you to stop.
* **No answer / voicemail** - nobody picked up.

Which buckets exist depends on the agent's [type](/agents/agent-types). A booking
agent has a "booked" bucket, a transfer agent has a "transfer" bucket, and so on.

## You decide what each bucket does

For each outcome, you set the follow-up. It's a set of "when this happens, do
that" rules:

* **When they ask for info** → text them your booking link.
* **When they book** → send a friendly confirmation.
* **When they need a human** → label them so your team picks it up.
* **When it's a voicemail** → leave a note on the contact.

You chain simple actions, send a text or an email, add or remove a label, add a
note, update a field, and move them in your pipeline. Small actions, wired to the
outcome that should trigger them.

## The system already handles the basics

You don't start from zero. On every call, the platform automatically does the
groundwork: it tags each person by what happened (booked, opted out, transferred),
keeps your dialer moving, and raises alerts when a transfer fails.

So your follow-up rules are the **extra** touches you add on top, not a list you
have to build from scratch. A brand-new campaign with no rules of your own still
tags people correctly after every call.

<Warning>
  Because the system tags people automatically, you don't need to re-create those
  tags yourself. Add your **own** labels and actions for the extra steps you want,
  and let the built-in tagging do its job.
</Warning>

## The automatic labels

These are the labels the system applies for you, so you can trigger your own
follow-up off them:

| What happened                     | The label it adds        |
| --------------------------------- | ------------------------ |
| Booked, confirmed, or rescheduled | `tfu_ai_booked`          |
| Cancelled                         | `tfu_ai_cancelled`       |
| Transferred to a person           | `tfu_ai_live_transfer`   |
| Callback or asked for info        | `tfu_ai_standby`         |
| Needs a human                     | `tfu_ai_human_needed`    |
| Opted out                         | `tfu_ai_kill_switch`     |
| Transfer failed to connect        | `tfu_ai_failed_transfer` |

Use these as triggers. For example, when someone gets `tfu_ai_standby`, that's
your cue to send information and check back in a few days.