One agent, honest pricing
Every plan is the full product: same agent, same security model, same approvals. Plans differ only in included AI budget and connection count.
Starter
One business, one growth loop, fully run.
- $15/mo included AI model budget
- Up to 3 connected accounts
- 2 daily digests (morning + evening)
- Approvals via Slack, Telegram, email
- Full audit log + one-click export
- Kill switch
Growth
For businesses where marketing is the engine.
- $50/mo included AI model budget
- Unlimited connected accounts
- 3 daily digests
- Opus escalation for hard problems
- Everything in Starter
Scale
Agencies and multi-channel operators.
- $200/mo included AI model budget
- Unlimited connected accounts
- 3 daily digests
- Opus escalation for hard problems
- Priority support
- Everything in Growth
How overage works
Model usage beyond your included budget bills at our cost × 1.3, metered daily to Stripe. Guardrails are automatic: a heads-up email at 80%, the agent economizes (switches to a lighter model for non-urgent work) at 100%, and it pauses entirely at 120% until you explicitly resume it. The worst case is bounded and you chose it.
Why it can be priced like this
Your agent routes work to the right model for the job: a fast, inexpensive model triages every signal and reads every digest, the flagship model does the real thinking, and the biggest model is reserved for genuinely hard problems (Growth and Scale). Between wakes the agent's machine suspends entirely — you're not paying for an idle VM to sit there. That's what makes an always-on agent affordable at a flat monthly price.
Compare plans
| Starter $199/mo | Growth $499/mo | Scale $1,499/mo | |
|---|---|---|---|
| Included AI model budget | $15/mo | $50/mo | $200/mo |
| Connected accounts | Up to 3 | Unlimited | Unlimited |
| Daily digests | 2/day | 3/day | 3/day |
| Missions (goal-driven outreach) | ✓ | ✓ | ✓ |
| Tools via MCP (Apollo, Hunter, custom servers) | ✓ | ✓ | ✓ |
| Approvals via Slack, Telegram, email | ✓ | ✓ | ✓ |
| Opus escalation for hard problems | — | ✓ | ✓ |
| Full audit log + one-click export | ✓ | ✓ | ✓ |
| Kill switch | ✓ | ✓ | ✓ |
| Priority support | — | — | ✓ |
Questions people actually ask
What counts against the AI model budget?
Only the AI model tokens your agent consumes — reading results, writing replies, generating creative. Your ad spend is your own and never touches our billing. Every model call is metered to the cent and visible live on your Billing page.
What happens when I hit 100% of the included budget?
Nothing dramatic. At 80% you get a heads-up email. At 100% the agent switches to a lighter, cheaper model for non-urgent work and further usage bills as overage at cost × 1.3. At 120% the agent pauses itself entirely and waits for you to explicitly resume it. You can never wake up to a runaway bill.
How does the free trial work?
14 days, full product, card required at signup but nothing is charged until day 15. We email you three days before the trial ends. Cancel from the Billing page in one click — no emails, no calls.
Can I change plans later?
Yes, from the Billing page via the customer portal. Upgrades take effect immediately (your included budget updates with them); downgrades apply at the next cycle.
What about my data if I leave?
Export everything — workspace, structured data, full logs — as a bundle any time from Settings. Deleting your tenant permanently purges the workspace, credential vault, history, and destroys the sandbox machine. Both are one click, self-serve, and documented in our Terms.
Do you spend my money without asking?
No. Financial and irreversible actions are approval-gated by construction — the agent physically cannot execute them without a signed, single-use approval token that only you can create. You also set a hard daily ad-spend cap; by default it's zero.
Something else? See the full FAQ, read the billing docs, or email hello@waveguide.adwave.com.