What the lead intake bot costs

Key takeaways · verified May 2026

Fixed cost is essentially zero. Quiet weeks bill nothing.
Variable cost is pennies per lead, dominated by Bedrock tokens for the extractors and the composer.
~100 leads/month → under three dollars total.
~1000 leads/month → under ten dollars total.
A $10 AWS Budgets alarm catches anything strange before it surprises you.

Fig 6. Three tiers of cost. The bill scales with how many leads land; the floor is nearly zero.

The fixed cost is essentially zero

There is no per-lead licence, no minimum monthly fee, no “starter plan.” If your campaigns are quiet for a week, the AWS bill matches. Lambda, DynamoDB pay-per-request, S3, CloudWatch, EventBridge, and SNS all sit in or near the always-free tier at the volumes a small business sees. Even running the Drive-sync schedule every five minutes and the Google Ads poll every hour, all month long, costs a vanishingly small amount. EventBridge scheduled invocations are essentially free at this volume, and each fire is a single Lambda call that does a small amount of work.

The biggest single line item in tier one is usually CloudWatch log storage. You control that by retaining seven days instead of forever. After day eight the logs are gone and the bill stops compounding.

The variable cost is per-lead pennies

Three things scale with lead volume:

Bedrock Haiku tokens for the extractors. Each lead pays for three small model calls (intent, urgency, fit) running against the message text. At small-model rates and the short input lengths most lead messages have, the all-in cost per lead for extraction lands at fractions of a cent.
Bedrock Haiku tokens for the composer. Hot-route drafts and warm follow-up drafts each pay for one composer call. The input is the structured lead plus the relevant pricing, voice, and ICP excerpts retrieved from the Knowledge Base. The output is a short reply. Even on the most expensive moves this is well under a cent per lead. The composer doesn’t fire on nurture or reject, so cold leads cost only the extractor tier.
Titan Embeddings for the policy docs. One-off per file change. If you edit your pricing doc twice a month and your voice doc once, that’s three embeddings runs totalling tiny fractions of a cent. The embeddings power the citation gate — finding the right policy passage to ground a reply.

Add it up at typical small-business lead volumes (fifty to a few hundred leads a month across all sources). Tier two lands in the under-three-dollars range. Tier one floors are under fifty cents. The all-in monthly bill is a coffee a month. At higher volumes — a busy B2B campaign at a thousand leads a month — tier two becomes the headline number. Even there it stays in the “phone bill, not Netflix subscription” range.

Three traps you’re avoiding

Per-seat sales-engagement tools. Many off-the-shelf lead-routing tools charge $50–$300 per seat per month, regardless of lead volume. They assume your SDR team is permanent and full-time. You’re trading a flat per-seat bill for pay-per-use that mostly comes in pennies. The bot runs on the same shape regardless of how many reps are on call.
Always-on infrastructure for polling. A naive setup might run a small server that polls Google Ads or Meta every minute. That’s a $30+/month idle bill before it processes a single lead. EventBridge schedules trigger Lambda — you pay only when the schedule fires, and Lambda scales to zero between fires.
One large model on every read. A frontier-class model would spend ten to twenty times more per lead for what is, at this scale, a short-input short-output extraction task. Haiku-class models are correctly sized. Reaching for a bigger one because it sounds smarter is the most common way to triple the bill on a system where the per-call output is a small structured object.

When this stops being cheap

The math changes if the campaigns scale up dramatically (a paid acquisition push, a product launch that surfaces in many directories at once) and lead traffic goes into the tens of thousands per month. At that point the tier-two number becomes the headline, and there are levers: cache policy embeddings (already cached after the first lookup), batch the extractors into a single multi-output call, or move the urgency extractor to a smaller dedicated model. Most small businesses, including small teams running paid ads, never need any of those levers.

For everyone below that — and that’s most small businesses — a $10 monthly AWS Budgets alarm catches anything strange before it becomes a surprise.

In plain words

The fixed bill is nearly zero. The variable bill is pennies per lead, dominated by tokens. A typical setup runs at coffee-money for the whole month. Set a budget alarm that fits your expected volume, and the bill can’t surprise you.

All posts