Monitoring & Credits
The Monitor section gives you visibility into everything running in your workspace: workflow executions, credit consumption, error rates, and system health.
Monitor dashboard
Access monitoring from the Monitor section in the left sidebar. The dashboard shows:
Execution overview
| Metric | Description |
|---|---|
| Total executions | Number of workflow runs in the selected time period |
| Success rate | Percentage of executions that completed without errors |
| Failed executions | Count of executions that errored out |
| Average duration | Mean execution time across all workflows |
Execution history
A chronological list of all workflow runs with:
| Column | Description |
|---|---|
| Workflow | Which workflow ran |
| Status | Success, Failed, or Running |
| Started at | When the execution began |
| Duration | How long it took |
| Trigger | What initiated the run (webhook, schedule, manual, etc.) |
| Credits | How many credits this execution consumed |
Click any execution to see the full step-by-step trace: which nodes ran, what data flowed between them, and where errors occurred (see Execution History).
Credit usage
Credits are the unit of consumption for premium features (AI, scraping, vision, enrichment). Monitor credit usage in Monitor → Credit Usage.
Credit costs by category
| Category | Node | Credits per execution |
|---|---|---|
| AI | TinyGPT | 10 |
| AI | GPT Writer | 15 |
| AI | GPT Researcher | 20 |
| AI | Tiny Search | 10 |
| Scraping | Scrape Page, Crawl Site, Extract Data, Map URLs, Web Search | 2 |
| Vision | OCR | 3 |
| Vision | Read PDF | 3 |
| Vision | Table Extract | 5 |
| Vision | Extract Fields | 5 |
| Vision | Ask Document | 10 |
| Enrichment | Person Enrichment | 10 |
| Enrichment | Company Enrichment | 10 |
| Enrichment | Email Enrichment | 20 |
| All others | Flow control, integrations, database | 0 |
Usage breakdown
The credit usage view shows:
- Daily usage chart: visualize credit consumption over time
- By workflow: which workflows consume the most credits
- By node type: which AI/scraping/vision nodes are most used
- Remaining balance: credits left in your current billing period
Managing credit costs
| Strategy | How |
|---|---|
| Add early filters | Put If-Else nodes before AI nodes to skip processing for irrelevant data |
| Cache results | Store AI results in a database. Check cache before calling AI again for the same input. |
| Batch wisely | A For Each loop with 100 items and a TinyGPT node costs 1,000 credits (10 x 100) |
| Use the right node | TinyGPT (10 credits) is cheaper than Researcher (20 credits). Only use Researcher when you need web sources. |
| Test with samples | During development, limit batch sizes to 5-10 items |
Error monitoring
Error types
| Error type | Cause | Resolution |
|---|---|---|
| Connection error | Integration auth expired or API is down | Re-authorize in Build → App Authorizations |
| Rate limit | Too many API calls to an external service | Add Delay nodes between requests |
| Timeout | A node took too long to respond | Check the external service; consider retries |
| Validation error | Input data doesn't match expected format | Add data validation before the failing node |
| Credit exhaustion | Out of credits | Upgrade plan or wait for next billing cycle |
Setting up error alerts
You can add error handling to any workflow to get notified when things go wrong:
Any node → Error branch → Send Email to Yourself (HIGH priority) + Slack (#alerts)
See Error Handling for details on configuring error branches and retry logic.
Workflow health indicators
On the main Workflows list, each workflow shows health indicators:
| Indicator | Meaning |
|---|---|
| Green dot | All recent executions succeeded |
| Yellow dot | Some recent executions failed (intermittent issues) |
| Red dot | Most recent executions are failing |
| Gray dot | Workflow is paused or hasn't run recently |
Set up a "meta-workflow" that runs daily, checks your credit balance via the Monitor, and sends you a Slack alert when credits drop below 20% of your monthly allowance.
Credit-consuming nodes still cost credits during test runs. Use the Test tab on individual nodes during development to avoid running the full workflow repeatedly.