Blog: See Exactly What Your AI Agents Cost — Introducing the Cost Intelligence Dashboard

See Exactly What Your AI Agents Cost — Introducing the Cost Intelligence Dashboard

Release v1.0.152

Running a company powered by 21+ AI agents is powerful. But until now, it came with a blind spot: you couldn't see what those agents were actually costing you.

Today that changes. With v1.0.152, SaaS Factory ships the AI Credit & Cost Intelligence Dashboard — a full observability layer for your agent spend, complete with burn rate analytics, budget alerts, monthly forecasting, and actionable recommendations to cut waste.

The Visibility Gap

Here's what was happening behind the scenes: every time an agent ran, the observability router calculated a cost estimate — $3 per million input tokens, $15 per million output tokens — and stored token usage against each job in the database. The math was there. The data was there.

But none of it was visible to users.

No dashboard. No alerts. No way to know whether your marketing researcher was running three times a day burning through your Pro plan credits, or which of your products was the most token-hungry. The pricing page prominently advertises credit limits (50/month on Free, 2,000 on Pro), but users had no way to track burn in real time.

That's the gap this release closes.

What You Can See Now

Burn rate by agent type. The Cost Intelligence Dashboard ranks every agent role by credit consumption. Architect running hot? Research agent unusually active? You'll know, and you can act on it.

Burn rate by product. If you're operating multiple products on SaaS Factory, spend is broken out per product. No more mystery about which workload is driving your monthly bill.

Projected monthly spend. Based on your recent usage trajectory, the dashboard forecasts where you'll land by end of month. If you're trending 40% over your plan limit by day 12, you'll see that — not on day 30.

Budget threshold alerts. Set the number and let the platform come to you. Configure alerts at any percentage of your credit limit and get notified before you hit the wall.

Agent cost recommendations. This is the part that goes beyond reporting. The platform analyses cost-to-value ratios and surfaces specific agents it recommends pausing or throttling — with the reasoning to back it up. An expensive agent doing repetitive low-impact work in a pre-launch product is a waste. Now you can see it and fix it.

Why It Matters for Enterprise

Anyone who's been through an enterprise procurement process knows the question that always comes up: "How do we control and forecast AI spend?"

Competitors like Devin answer this with per-session billing and session-level cost breakdowns. That's a start. SaaS Factory now goes further with agent-level granularity, cross-product attribution, forward-looking forecasts, and automated recommendations — all in one place.

For finance teams that need to sign off on AI tooling, this dashboard is the answer to their standard checklist:

✅ Can we see what we're spending? Yes.
✅ Can we set budget limits? Yes.
✅ Will we get warned before we go over? Yes.
✅ Can we identify and cut wasteful spend? Yes.

Credits and Tokens, Unified

One subtle but important improvement in this release: the users.credits field — the quota that determines your plan tier — is now connected to the token-level cost data tracked by the observability router. For the first time, you can see both your plan credit balance and the underlying token consumption that drives it, side by side.

This matters because credits are an abstraction. Token usage is the reality. Understanding how they relate helps you make smarter decisions about plan sizing, agent configuration, and workload scheduling.

Get Started

Head to Observability → Cost Intelligence in your project dashboard to explore your spend data. If you haven't already, set a budget threshold alert under Settings → Budget Controls — it takes thirty seconds and could save you a surprise overage.

As always, the agents keep running. Now you can see exactly what that's worth.