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

# Configuring Budgets and Hours

> Control how much your agent spends and when it is allowed to run.

BDR agents run autonomously, so it's important to set clear boundaries on spending and scheduling. You configure these when creating the agent and can adjust them at any time from the agent's edit page.

## Monthly budget

Every agent has a **monthly budget** that caps how much it can spend. Once the budget is exhausted, the agent pauses automatically until the next billing cycle.

You set the monthly budget when creating the agent and can change it at any time. Choose an amount that suits your campaign — start conservatively and increase as you gain confidence in the agent's performance.

### What counts towards the budget

All agent actions — sending emails, creating contacts, updating statuses, escalating, and drafting replies — are **free**.

The agent's budget covers the AI processing that happens behind the scenes when the agent evaluates each company and decides what to do. Contact discovery and enrichment are billed separately as part of the [enrichment process](/discovery/enrichment).

<Note>
  **Personalised agents spend more per company than Template agents.** They do extra work — refreshing [prospect intel](/bdr-agents/personalised-outreach#prospect-intel-refresh), composing original copy, and running safety checks — so budget accordingly. Most of the heavier spend is during [setup and testing](/bdr-agents/personalised-outreach#a-note-on-cost); steady-state running is more modest.
</Note>

### Per-company budget tiers

The agent can also limit how much it spends on each individual company, based on how well the company matches your profile. Higher-confidence matches get a larger per-company budget, while borderline matches get less. You can customise these tiers on the agent edit page.

## Operating hours

You can restrict when the agent sends emails so outreach lands during business hours:

* **Enabled/Disabled** — toggle operating hours on or off. When disabled, the agent runs 24/7.
* **Timezone** — the timezone for the schedule (default: Europe/London)
* **Start and end time** — e.g. 09:00 to 17:00
* **Active days** — which days of the week the agent is allowed to run (default: Monday–Friday)

<Tip>
  Setting operating hours in the prospect's timezone ensures emails arrive during business hours, which improves open rates. Choose the timezone where most of your target companies are based.
</Tip>

When the agent hits a time outside its operating hours, it simply pauses until the next active window. No work is lost — it picks up where it left off.

## Run pace

The **run pace** controls how many companies the agent processes per cycle:

* **Default:** 1 company per run
* **Recommended for testing:** 1 company per run (lets you review each interaction)
* **After confidence builds:** Increase to 3–5 companies per run

A lower pace gives you more control and visibility. A higher pace means faster outreach but requires more trust in the agent's configuration.

<Tip>
  Pair a slow run pace with [simulation mode](/bdr-agents/how-the-agent-works#simulation-mode) (an agent with no connected email account) to review each decision before any real email is sent.
</Tip>

## Follow-up cadence

Configure how the agent follows up with prospects who haven't replied:

| Setting                         | Default | What it controls                                                       |
| ------------------------------- | ------- | ---------------------------------------------------------------------- |
| **Min days between follow-ups** | 3 days  | Minimum gap between follow-up emails to the same person                |
| **Max follow-ups per contact**  | 2       | How many follow-up emails before the agent stops                       |
| **Follow-up template groups**   | None    | Specific templates for follow-ups (if different from initial outreach) |
| **Dormant re-evaluation**       | 90 days | How long before the agent revisits a company that went quiet           |

<Note>
  The follow-up cadence applies per contact, not per company. If the agent finds multiple contacts at the same company, each gets their own follow-up sequence.
</Note>

<Note>
  In **Template** mode, follow-ups reuse your templates (or a constrained variation). In **Personalised** mode, follow-ups can introduce **new offers and angles** from your [playbook](/bdr-agents/personalised-outreach#multi-offer-follow-ups) so a sequence stays fresh rather than repeating the same ask.
</Note>

## Adjusting settings on a live agent

You can edit any of these settings while the agent is running — click the agent name in **Discovery → AI Agents**, then click **Edit**.

Changes take effect on the next processing cycle. The agent won't interrupt an action already in progress, but will use the new settings for everything going forward.

## When the budget runs out

If the agent hits its monthly budget limit:

1. The agent **pauses automatically**
2. You receive a notification (if escalation emails are configured)
3. The agent resumes at the start of the next month, or when you manually increase the budget

You can also manually pause and resume agents at any time from the agent detail page.