Pricing
Simple, predictable
Built for SaaS and integrations—and the APIs your AI agents call. Free: 2 APIs, 7-day history. Pro: production guardrails, dependency map. Scale: teams, Slack, anomaly detection. No long-term contracts.
Upgrade or downgrade from Billing. No lock-in.
Estimate your API monitoring usage
See which plan fits your expected traffic. One event = one tracked API call.
Estimated API calls monitored per month: 40,000
Recommended plan: Pro
Example products
- Side project
~5k API calls/month → Free
- AI SaaS with OpenAI
~40k API calls/month → Pro
- Production SaaS with payments + AI
~300k API calls/month → Scale
Typical DependWatch usage
Realistic product scenarios and recommended plans.
Side project / indie SaaS
Typical usage: 5k–20k API calls/month
Recommended: Free
Hobby projects, prototypes, small side tools.
Early SaaS startup
Typical usage: 30k–120k API calls/month
Recommended: Pro ($29/mo)
Production SaaS: payments, AI inference, notifications.
Production SaaS platform
Typical usage: 300k–1M+ API calls/month
Recommended: Scale ($99/mo)
Teams running multiple services. Full observability over every API and tool their software depends on.
Free
Perfect for side projects and early experimentation.
Best for: Side projects, trials, 1–2 APIs
10,000 events included
Hard cap (no paid overage)
2 APIs · 7-day history
1 alert rule (no Slack)
Pro
For production SaaS monitoring critical APIs.
Best for: Production apps, 3–10 APIs, guardrails
100,000 events included
$5 per 100k events
10 APIs · 90-day history
Slack alerts (3 webhooks)
- 10 monitored APIs
- 90-day event history
- 10 alert rules, 3 Slack webhooks
- Operation-level analytics
- API Intelligence (cost drivers, reliability, slow endpoints)
- Guardrails: cost, error, latency spike
- Dependency map (reliability & cost per provider)
- Projected cost & cost-spike detection
- SDK + MCP / AI assistant integration
Upgrade when
You need team collaboration, unlimited Slack webhooks, 1-year history, or unlimited APIs.
Scale
For teams operating multiple services and needing full observability over every external API and tool their software depends on.
Best for:
1,000,000 events included
$3 per 100k events
Unlimited APIs · 365-day history
Slack alerts (unlimited)
DependWatch never stops monitoring your APIs. Pro and Scale: overage billed at period end at the rates above. Free: hard cap, no overage charge. Real API events only—test/demo excluded from billing.
What counts as an event?
One event = one tracked external API call. Most SaaS products generate between 10k and 200k API calls per month depending on traffic. Free (10k) fits side projects and low-volume apps; Pro (100k) fits typical production; Scale (1M) fits high-traffic or multi-service setups.
What counts as an API monitored?
An API monitored = a distinct external provider you send events for (e.g. OpenAI, Stripe, Twilio). We count unique providers per project. Operation-level tracking (e.g. openai.chat.completions, stripe.paymentIntents.create) happens inside each provider and does not increase the API count.
Feature comparison
Everything included in each plan.
| Feature | Free | Pro ($29/mo) | Scale ($99/mo) |
|---|---|---|---|
| Events included | 10,000 | 100,000 | 1,000,000 |
| Overage | Hard cap | $5/100k events | $3/100k events |
| APIs monitored | 2 | 10 | Unlimited |
| Event history | 7 days | 90 days | 365 days |
| Alert rules | 1 | 10 | Unlimited |
| Slack webhooks | 0 | 3 | Unlimited |
| Provider-level analytics | Yes | Yes | Yes |
| Operation-level analytics | — | Yes | Yes |
| API Intelligence (insights) | — | Yes | Yes |
| Guardrails (cost, error, latency spike) | — | Yes | Yes |
| Traffic anomaly detection | — | — | Yes |
| Dependency map | — | Yes | Yes |
| Projected cost & cost-spike detection | Dashboard only | Yes | Yes |
| Team workspace | — | — | Yes |
| Test event onboarding | Yes | Yes | Yes |
| SDK integration | Yes | Yes | Yes |
| MCP / AI assistant integration | Yes | Yes | Yes |
Why teams upgrade
- Free → Pro: More than 2 APIs, more than 7-day history, or need for cost-spike guardrails and operation-level analytics.
- Pro → Scale: Team collaboration, Slack alerts, 1-year history, or unlimited APIs. Scale adds anomaly detection.
Built for modern API-driven stacks. SaaS. Integrations. The APIs your AI agents call.
The observability layer for products that depend on these providers—and any other HTTP API.
Frequently asked questions
- What counts as an event?
- One event = one tracked external API call. Each time your app calls wrap() or track() and sends data to DependWatch, that’s one event. Batched requests count as multiple events (one per call in the batch).
- Can I upgrade or downgrade anytime?
- Yes. Upgrade or downgrade from your workspace Billing page. Changes take effect at the next billing cycle; we don’t prorate mid-cycle for simplicity.
- Do you charge overages?
- Pro and Scale plans: usage over your included events is billed at $5/100k (Pro) or $3/100k (Scale) and added to your next invoice. Free plan has a hard cap with no paid overage. Billable usage excludes test/demo events. See Billing for your current period usage.
- Can I monitor multiple services?
- Yes. Each project can send events from one or many services (same ingest key). APIs monitored = distinct providers (e.g. OpenAI, Stripe) you send events for. Scale has unlimited APIs; Free caps at 2, Pro at 10.
- How do alerts work?
- Free: 1 alert rule, no Slack. Pro: 10 rules, 3 Slack webhooks. Scale: unlimited rules and webhooks. Configure thresholds (latency, error rate, budget) in Project → Settings → Alerts. Alerts fire when a threshold is exceeded and are sent to your Slack webhooks; cooldown reduces noise.
- How accurate is cost forecasting?
- Projected monthly spend is extrapolated from your events: (total cost in period / days in period) × 30. Accuracy depends on you sending estimated_cost_usd per call where applicable. We use it for trends and guardrails, not invoicing third-party APIs.
All plans include latency and error metrics. No lock-in. Keys hashed; we never log request bodies.