> For the complete documentation index, see [llms.txt](https://boundaryai.gitbook.io/boundaryai-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://boundaryai.gitbook.io/boundaryai-docs/help/faq.md).

# FAQ

Answers to common questions about using BAI Analytics. For fixes to specific problems, see [Troubleshooting](/boundaryai-docs/help/troubleshooting.md); to reach us, see [Getting help](/boundaryai-docs/help/getting-help.md).

<details>

<summary>How do I export my data?</summary>

From any analysis, use **Download raw data** to export every response and its metadata to Excel. You can also turn a feedback group into a **report** and export it as PDF, HTML, or JSON. See [Reports](/boundaryai-docs/analysing-your-feedback/reports.md).

</details>

<details>

<summary>What languages does BAI Analytics support?</summary>

BAI Analytics can read and analyse feedback in **80+ languages**, so you can collect responses from a global audience. The analysis it produces (themes, summaries, and insights) and the interface are available in **English, French, and Spanish**. You can also publish a single survey in several languages with automatic translation, and each person can set their own interface language. See [Customizing a Survey](/boundaryai-docs/bringing-in-your-feedback/surveys/images-and-media.md).

</details>

<details>

<summary>What can my organisation use, and how do credits work?</summary>

There are no self-serve plans. The BAI Analytics team sets up each organisation with a set of **add-ons** (for example Connected platforms & API, Web & social monitoring, AI Visibility) and allowances; the **Organization access** panel in Settings lists what is enabled for yours.

Usage is metered in two separate ways: a **Usage Allowance** for AI analysis (themes, sentiment, Custom Monitoring, summaries, PII redaction) that resets monthly, and **Web & Social Media Credits** for scraping agents, which you can top up. To enable an add-on or change an allowance, contact your BAI Analytics representative. See [Settings](/boundaryai-docs/account-and-administration/updating-settings.md).

</details>

<details>

<summary>Can BAI Analytics remove personal data from feedback?</summary>

Yes. Organisation admins can switch on **PII redaction** in Organization details. Emails, phone numbers, card numbers, IBANs, IP addresses and custom patterns are removed by pattern matching, and an optional AI pass catches names, addresses, ID numbers, dates of birth, health information and anything you describe in custom instructions. Redaction happens as feedback arrives, before it is stored, and is permanent. Individual feedback groups can override the organisation setting. See [PII redaction](/boundaryai-docs/account-and-administration/updating-settings.md#pii-redaction).

</details>

<details>

<summary>Can I delete my data?</summary>

Yes. You can delete individual sources (a survey, upload, connection, or agent), which removes their responses and analyses, and you can delete whole feedback groups, all from within the app. To permanently remove all of your data, contact **<info@boundary-ai.com>** or your BAI Analytics representative.

</details>

<details>

<summary>How often does connector data update?</summary>

Support-platform connections with continuous import enabled sync automatically about every 30 minutes. Native connectors (CX platforms, data warehouses, file delivery) run on a schedule you choose per sync, hourly by default. To pull the latest data on demand, use **Sync now** on the connection or **Fetch new data** on the source's card. See [Connecting your tools](/boundaryai-docs/bringing-in-your-feedback/connect-to-your-existing-systems.md).

</details>

<details>

<summary>Can my team sign in with our company SSO, and can our IT team manage users?</summary>

Yes. Organisations can sign in through their own identity provider, and BAI Analytics supports **SCIM 2.0 provisioning** so your identity provider (Okta, Microsoft Entra ID, or any SCIM 2.0 compatible directory) creates, updates and deactivates users automatically. Setup is done with the BAI Analytics team; see [Single sign-on and automatic provisioning](/boundaryai-docs/account-and-administration/updating-settings.md#single-sign-on-and-automatic-provisioning).

</details>

<details>

<summary>Can I ask questions about my feedback from ChatGPT or Claude?</summary>

Yes. Open **Integrations** and choose **AI assistants**. Once an admin has enabled assistant access for your organisation, connect your own assistant, approve its read-only access, and start asking questions about your feedback groups. See [Connect AI assistants (MCP)](/boundaryai-docs/api-and-webhooks/mcp.md).

</details>

<details>

<summary>Is there a way to test my survey before sending it out?</summary>

Yes. Click **Preview Survey** at any point while building to see exactly how it will appear to respondents, and review the draft before you distribute it.

</details>

<details>

<summary>Where does the NPS benchmarking data come from?</summary>

Our **Net Promoter Score (NPS) benchmarks** are derived from a variety of industry reports, customer surveys, and workplace engagement studies. We collect data from:

1. **B2B Customer Feedback**: industries such as consulting, technology, and digital marketing, benchmarked using reports from sources like QuestionPro and other research platforms.
2. **B2C Customer Feedback**: sectors like retail, airlines, and car rentals, benchmarked using industry-wide consumer sentiment studies from organisations like Trustmary.
3. **Academic Feedback**: education and e-learning benchmarks, derived from academic research and higher-education feedback studies.
4. **HR Feedback (eNPS)**: employee engagement scores in industries like construction, hospitality, and technology, sourced from workplace satisfaction studies (e.g. HeartCount).

NPS benchmarks may vary based on survey methodology, sample size, and geographic region, so treat them as general reference points. For a walkthrough of the benchmarking screen, see [Benchmarking your NPS](/boundaryai-docs/analysing-your-feedback/analysing-results/benchmarking-your-nps-against-industry-standards.md).

</details>
