# FAQ ## General **What is Finora?** A FinOps dashboard for AI spend. Connect OpenAI, Anthropic, AWS Bedrock, Azure AI Foundry, Google Cloud, and Cursor — Finora pulls your costs together, normalizes them, and shows you what's happening with your money in clear charts and reports. **How is Finora different from cloud cost tools like Cloudability or CloudZero?** Finora is purpose-built for AI workloads. Cloud cost tools cover everything (compute, storage, networking, etc.) and treat AI as just another line. Finora has model-aware metrics like cost per 1k tokens, AI-tuned anomaly detection, and direct integrations with foundation-model providers (OpenAI, Anthropic, Cursor) that general cloud tools don't natively support. **Does Finora replace my provider's own billing dashboard?** No — it sits on top of them. Provider dashboards show their own usage. Finora consolidates all of them in one place, adds tags, alerts, forecasting, and FOCUS-compatible exports. ## Plans & pricing **Is there a free tier?** Yes — every account starts on **Trial** with full UI access and capped capacity (2 providers, 7-day history, 2 budget rules). See [Plans & Billing](/account/plans-and-billing.md) for the full comparison. **Can I try Scale or Enterprise before committing?** There's no automatic trial of paid tiers, but Enterprise customers typically run a paid pilot — contact . **How do I get an invoice with my company's PO number?** Email with the PO and we'll re-issue the next invoice with it included. ## Connecting providers **How long until I see data after connecting a provider?** Within one full refresh cycle: 24 hours on Trial and Growth, 1 hour on Scale, 15 minutes on Enterprise. **Can I track Claude.ai or ChatGPT consumer plans?** Not as billing integrations — those only cover the API platform. For flat-rate consumer or team seats, use the [AI Subscriptions tracker](/core-features/ai-subscriptions.md). **What about self-hosted models (Ollama, vLLM)?** Self-hosted inference doesn't have a provider bill, so there's nothing to read. If you run on a cloud (e.g. GPU instances on AWS), the cloud provider's bill captures the spend — but Finora's current AWS integration is scoped to Bedrock specifically. General-purpose AWS GPU cost tracking is on the roadmap. **Does Finora support OpenRouter, xAI, or GitHub Copilot?** Not yet. They're on the public roadmap. You can track flat-rate Copilot Business via the AI Subscriptions tracker today. ## Data & retention **How long is my cost data retained?** Trial 7 days, Growth 90 days, Scale 365 days, Enterprise unlimited. Older data is hidden from charts but not deleted — upgrading restores access. **Where is my data stored geographically?** Your account's primary region. Enterprise customers can request region pinning. **Can I export my data?** Yes — Standard CSV and FOCUS 1.1 CSV are available on Growth and above. See [exports](/exports/standard-csv.md). **How do I delete all my data?** Email . We'll confirm and then remove everything: your costs, tags, dashboards, alerts, and stored credentials. Stripe billing records are retained for legal and tax reasons. ## Alerts & notifications **When does an alert fire?** Budget alerts fire at 80% and 100% of threshold. Anomaly alerts fire only for **high severity** anomalies on the current day. **Where do alerts go?** Email (always) and Slack (Scale and above). Configure both in **Settings → Alert Notifications**. **Can I get SMS alerts?** Not in v1.0. SMS is on the roadmap. **Why didn't I get an alert for a spike I noticed?** Anomaly alerts only fire for **high severity** events. Lower-severity anomalies show up in the Anomalies widget but don't notify. The high threshold requires both more than 3× the 7-day average **and** an absolute spike of more than $500. See [Alerts & anomalies](/core-features/alerts-and-anomalies.md). ## Tagging **Are tags shared across users in my workspace?** Yes — tags are global per workspace. The intent is shared vocabulary; user-scoped tags aren't supported. **How do I bulk-tag historical events?** From the Overview page: filter to the events you want, select all, click **Bulk tag**. On Enterprise, you can also create automatic tag rules that apply on save and backfill history once. ## Reports **Can I schedule a recurring report email?** Yes — on Scale and above. Set the **Schedule** field on a report to Daily / Weekly / Monthly and add email recipients. The report is regenerated and emailed automatically. **Are reports printable?** Yes — choose the **Print-friendly** report style in **Settings → Profile & Account** for black-and-white output. ## Security **How are my provider API keys stored?** Encrypted in a secure vault, scoped to your account. Never shown back in your browser after first save. Never logged. **Does Finora have SOC 2 / ISO 27001 / HIPAA?** SOC 2 Type II is in progress (target: H2 2026). Enterprise customers can request the bridge letter. ISO 27001 is on the roadmap. HIPAA isn't supported today. **How do I report a security issue?** Email with subject `Security report`. We acknowledge within 24 hours. ## Troubleshooting **The Overview is empty even though I connected a provider.** Wait one full refresh cycle. If still empty, check **Settings → API Keys** — the provider may show "needs attention" if the credential failed. **My forecast says "Low confidence."** You have less than 30 days of history or recent volatility is high. The forecast is still useful as direction but shouldn't drive a tight finance plan. **A KPI shows `NaN` or `Infinity`.** Most likely a divide-by-zero in a custom formula. Wrap with a guard: `total_inferences > 0 ? total_cost / total_inferences : 0`. For more, see [Troubleshooting](/reference/troubleshooting.md). --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.finora.services/reference/faq.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.