AI
OpenAI
AI API provider for GPT-4, o1, embeddings, and other foundation models used in production AI applications worldwide.
What gets synced
Roiva writes these metric observations on each sync. Reference the key in a value formula to use this data in your ROI calculations.
API Usage
openai.usage.input_tokens
Input Tokens
count
openai.usage.output_tokens
Output Tokens
count
openai.usage.total_tokens
Total Tokens
count
openai.usage.request_count
Request Count
count
openai.usage.<model_slug>.input_tokens
Per-model input tokens (dynamic, e.g. openai.usage.gpt_4o.input_tokens)
count
openai.usage.<model_slug>.output_tokens
Per-model output tokens (dynamic)
count
openai.usage.<model_slug>.total_tokens
Per-model total tokens (dynamic)
count
openai.usage.<model_slug>.request_count
Per-model request count (dynamic)
count
Common use cases
- Track API spend per initiative to measure the cost side of AI ROI
- Monitor token consumption across models to identify efficiency opportunities
- Measure request volume growth as a proxy for AI feature adoption
- Compare cost-per-task before and after model or prompt optimizations
How to connect
- In OpenAI: Go to platform.openai.com/api-keys and generate an API key
- In OpenAI: Ensure the key belongs to an owner or admin in your organization to access usage data
- Back in Roiva: Paste the key here and save the connection
- Back in Roiva: After connecting, link this connection to an initiative on the Initiatives tab
Tips for capturing value
- Usage data is available for the past 90 days via the API — older data requires exporting from the OpenAI dashboard
- Cost-per-unit (per document, per call, per user) is more meaningful than total cost for ROI measurement
- Pairing OpenAI cost data with outcome data from HubSpot, Zendesk, or Stripe gives you a complete ROI picture
OpenAI
Sign up to connect OpenAI and start tracking AI initiative outcomes.
Start free trial → Sign inRelated integrations
