Frequently asked questions
Your questions about Zebra AI, answered
Get answers to common questions about AI data analysis with Zebra AI – from Power BI, Excel and database connections to automated reporting, calculation accuracy, recurring reports, pricing and security.
On this page
- 01Is Zebra AI right for me?
- 02How Zebra AI works with your data
- 03Trust and analysis
- 04Plans and pricing
- 05Sharing your analysis
- 06Security and data
Can’t find your question?
Get in touch
01 — What Zebra AI does
Is Zebra AI right for me?
What Zebra AI does, what you can use it for, and how it differs from general-purpose AI tools.
What is Zebra AI?#
Zebra AI is an AI analytics platform that turns business data into decision-ready analysis.
Connect your data and Zebra AI automatically builds a structured data story showing what changed and what is driving it. Then ask follow-up questions in plain language, explore the numbers further, and share the result as a live story or presentation.
The output uses Zebra BI’s IBCS-based business visuals, aligned with the international reporting standard ISO 24896:2026.
What can I use Zebra AI for?#
Zebra AI is designed for three common jobs: recurring reporting, deeper analysis and planning, and answering business questions without waiting on a BI team.
Refresh recurring management reports, investigate variances and root causes, drill into costs or performance, model scenarios, or answer questions such as “Why are we below plan?”, “Which regions drove the decline?” or “What would it cost to hire five more people?”
The analysis does not end with the first dashboard. Keep asking questions, drill further into the data, and refresh the same story when the next reporting cycle arrives.
Who is Zebra AI for?#
Zebra AI is for people who regularly work with business data but do not want every question to become a BI project.
For finance, that includes management reporting, P&L and variance analysis, planning, costs and headcount. Revenue and sales teams investigate pipeline and performance; operations teams explore cost, delivery and operational drivers.
You don’t need to belong to a particular function. If your job involves understanding business numbers, finding out what is driving them, and communicating the answer, Zebra AI is built for that workflow.
Do I need to be a data analyst?#
No. Connect your data and Zebra AI builds the first analysis for you. From there, ask questions in normal language instead of writing SQL or DAX or manually configuring every chart.
For analysts, Zebra AI also removes repetitive work from report building and recurring analysis.
Do I need Power BI?#
No. Zebra AI works directly with Excel and CSV files, SQL databases, Snowflake and Databricks, as well as Power BI semantic models.
If you already use Power BI, Zebra AI works with the governed model your BI team has built, including its measures and row-level security. If you don’t have a BI stack, start directly from a file or database connection.
How is Zebra AI different from ChatGPT or other general-purpose AI tools?#
General-purpose AI is designed to answer almost anything. Zebra AI is purpose-built to analyse business data.
Zebra AI computes calculations with deterministic tools against the data you connect rather than asking a language model to generate the numbers. Every figure is traceable to the underlying data, and the analysis is structured around the drivers, variances and comparisons that matter in business reporting.
It also produces structured, IBCS-based business visuals and keeps the analysis reusable. Instead of starting again with a new prompt next month, refresh the same story for the next close, QBR or business review.
02 — How Zebra AI works with your data
Will Zebra AI work with my data?
What Zebra AI connects to, what your data needs to look like, and how it handles your terminology and your updates.
What data sources can I connect?#
Excel and CSV files, Power BI semantic models, Microsoft Fabric, SQL databases including Microsoft SQL Server, MySQL and PostgreSQL, Snowflake and Databricks.
Whether your starting point is a governed Power BI model your team maintains or a spreadsheet you exported five minutes ago, Zebra AI works with both.
How should my data be structured?#
Your data doesn’t need to be analysis-ready, but good structure matters. Zebra AI works best with flat datasets with clearly labelled columns, consistent values within each column, and data arranged as a table rather than a formatted report.
In practice, a tabular export from your ERP, CRM or warehouse is a good starting point. A report designed for people to read — with merged cells, subtotal rows or several layers of headers — usually needs some restructuring first.
Does Zebra AI use our fiscal year?#
Yes. In Data Preview, set the fiscal year used by your organization so period comparisons and drill-downs follow your reporting calendar rather than the calendar year.
How does Zebra AI work with Power BI?#
Zebra AI connects to the Power BI semantic model your BI team already built and maintains. It works with the governed model rather than asking you to recreate it elsewhere, and respects existing row-level security.
You don’t need to write DAX to start analysing it.
Can I connect Snowflake or Databricks?#
Yes. Both connect directly, so Zebra AI queries your warehouse or lakehouse at the source rather than working from an extract. No copied dataset to maintain, and no BI layer required in between.
Can I teach Zebra AI our terminology, KPIs and business logic?#
Yes. Add context about what your columns, terms and abbreviations mean, how they relate, and any known trends or seasonality Zebra AI should understand. Row-level comments can also be read alongside the numbers.
For Power BI and database connections, this context lives with the dataset connection, so every story built from it starts with the same agreed definitions rather than rebuilding that understanding each time.
Can I add my own calculations and KPIs?#
Yes — and Zebra AI can suggest useful measures for you too.
Zebra AI analyzes the columns in your data and identifies KPIs that can be calculated from them, such as margins, rates, ratios or other common business measures. You can review those suggestions and decide which ones should become part of the analysis.
You can also create your own calculated measures for company-specific KPIs that Zebra AI could not infer on its own — for example an internal efficiency metric, custom margin definition or business-specific performance indicator.
Once added, calculated measures become part of the data model and are available across charts, calculations and chat, so Zebra AI understands and uses the same KPIs throughout the analysis.
What happens when my data updates?#
You refresh, and the same analysis rebuilds on current numbers — same structure, same definitions, new data. Build once, refresh every cycle.
If you’ve made manual changes to the story, refresh protection catches them and asks whether to keep or overwrite. A refresh never silently discards your work.
Do I need to reconnect my data each session?#
Your story — structure, charts, chat history and folders — is retained. The underlying dataset isn’t: it’s deleted on logout or after six hours of inactivity, so you reconnect to resume. Established dataset connections to, e.g., Power BI are available in the sidebar and you can create a new story out of them without having to configure them from scratch.
That deletion is deliberate and is covered in the security section below.
03 — Trust and analysis
Can I trust the analysis?
How the numbers are produced, how to check them, and how much of the output you control.
How does Zebra AI calculate my numbers?#
Two separate systems do two separate jobs. The language model interprets your question and writes the commentary. Dedicated calculation tools compute the aggregations, variances and comparisons against the data you connected.
The language model does not perform the calculations. It interprets the question and explains the result; the underlying figures are computed separately against your data. That separation is fundamental to how Zebra AI works — and one of the key differences from using a general-purpose AI tool to analyse a spreadsheet.
Can I verify the numbers Zebra AI gives me?#
Yes. Every figure traces back to the data behind it, and you can open the thought process to see how your question was interpreted and which calculations ran as well as the settings on each visual to see raw numbers and key statistics. That means your team can review the full picture before it reaches leadership.
The numbers being checkable is what makes the analysis reviewable rather than something you have to accept on trust.
How does Zebra AI visualize my data?#
Zebra AI uses Zebra BI Charts and Tables to turn each insight into a business-ready visualization. It recommends a visualization based on the data and the type of analysis, while also giving you alternative ways to present the same insight.
You can follow the recommendation or choose a different option — for example, switch from a chart to a table or select another chart type that better fits the message you want to communicate.
The visuals are built on IBCS business-reporting conventions, which form the basis of ISO 24896, the international standard for business reporting notation. This gives the analysis a consistent visual language for comparisons, variances and other common business-reporting patterns.
Can I customize the charts and dashboard?#
Yes. Zebra AI gives you a recommended first version, not a fixed dashboard.
For each insight, you can choose from alternative visualizations, switch to another chart type or table, and adjust the individual visual settings. You can also add, remove or rearrange visuals as you shape the story.
Story Settings let you customize the dashboard at a broader level, including changing its colors to better match your company branding or presentation style.
The goal is to give you a strong starting point while keeping you in control of how the final analysis looks and how the message is communicated.
Does Zebra AI understand hierarchies in my data?#
Yes. Zebra AI detects hierarchies in your data and builds them into the analysis, two levels deep, with expand and collapse. Region to country, channel to campaign, category to product — and for P&L data, a full waterfall from revenue down to EBITDA.
You can also adjust the visualized hierarchies manually.
04 — Plans and pricing
Plans and pricing
How the trial works, what happens afterwards, and what counts toward your AI usage.
Can I try Zebra AI for free?#
Yes. Zebra AI runs a 7-day reverse trial: you start with access to the premium features, and when the trial ends you continue on the free plan rather than losing your work. No credit card is required to start.
What happens after my 7-day trial?#
Your account moves to the free plan automatically. The stories you created during the trial remain in your workspace; your available features and usage then follow the limits of the free plan.
What counts as an AI interaction?#
An AI interaction is an action that triggers new AI analysis or generation — for example generating a dashboard, asking a follow-up question, or adding a chart through AI.
What’s the difference between the free and paid plans?#
The free plan is designed for lighter individual use. Paid plans increase usage and unlock additional capabilities for teams and larger deployments, while Enterprise adds deployment and security options for organizations with more complex requirements.
05 — Sharing your analysis
Can I share and reuse what I build?
Getting the analysis in front of the people who need it — this cycle and every cycle after.
Can I export my analysis to PowerPoint or Excel?#
Yes, in one click. PowerPoint export includes the charts, titles, summaries and speaker notes, already formatted. Excel export includes the questions and answers, the charts, the data slices behind the analysis, and a story overview.
Exports use Zebra BI for Office, so the charts remain editable and dynamic rather than becoming flat screenshots. Keep working with them after export — no rebuilding the deck at 11pm, no copy-pasting screenshots.
Can I use Zebra AI for recurring reports?#
Yes. Recurring reporting is one of the core workflows Zebra AI is built for. Set the analysis up once — structure, definitions, business context — and every cycle after is a refresh rather than a rebuild.
Month one you build the story. Month two onward you review and send. Monthly close, board pack, QBR, forecast review: the work compounds instead of resetting each period.
Can Zebra AI generate analysis in other languages?#
Yes. Titles, summaries, commentary and chat responses are generated in your team’s chosen language, switchable in the preview. Your data and column names stay as they are.
06 — Security and data
How does Zebra AI handle security and data?
What the AI sees, where your data is processed, and the deployment options available for enterprise teams.
Is my data used to train AI models?#
No. Zebra AI does not train models on customer data, and the AI-providers we use guarantee the same for their APIs.
Where is my data hosted, and how long is it retained?#
Zebra AI runs on Microsoft Azure in Western Europe. For uploaded data, the underlying dataset is retained only for the active session and deleted on logout or after six hours of inactivity.
How is my data encrypted?#
Data in transit is encrypted using TLS 1.2 or higher, and data at rest is encrypted with AES-256.
What deployment options are available?#
Zebra AI is available in the browser, as a workload inside Microsoft Fabric, and as an Azure Managed Application for enterprises that need deployment inside their own Azure environment.
The Azure Managed Application lets your organization retain control over identity, networking, governance, monitoring and compliance controls while Zebra AI remains managed and supported.
Connect your data and see it for yourself
Upload a file or connect your existing model, and get a complete dashboard with written analysis in minutes. Free to start, no credit card.
Still have a question? Get in touch or browse the knowledge base.