Skip to main content

Frequently asked questions

Clear answers about TXNORA.

Here you can read what TXNORA does, where your data is processed, how integration works and how to get access to the API.

Product

What TXNORA does, which data it processes and who it is for.

What is TXNORA?

TXNORA is an API-based service for the automated processing of unstructured data. Its focus is transaction intelligence: through one API, TXNORA turns raw transaction data into normalized merchants, consistent categories, contextual labels and a confidence value.

It is built primarily for financial software providers: fintechs, accounting platforms, expense tools and analytics products. TXNORA is also built for companies, SaaS providers, platforms and data providers that process large volumes of unstructured data automatically. See Use cases for more.

What does TXNORA return for a transaction?

For each transaction, TXNORA returns four results: the merchant, a category, contextual labels and a confidence value.

A typical example: the raw description “PAYPAL *ADOBE” becomes the merchant Adobe, the category Software and the labels SOFTWARE and DIGITAL_SERVICE.

Can TXNORA process data other than transactions?

Yes. The same platform also processes text, documents and other records. Seven functions are available: Categorization, Labeling, Data Annotation, Data Classification, Information Extraction, Data Intelligence and Custom Endpoints.

For example, records can be assigned to categories automatically, large datasets can be annotated in a structured way, and relevant information can be extracted from unstructured text.

Are there custom endpoints for specific requirements?

Yes. With custom endpoints, TXNORA processes data according to your own prompts, rules and data structures. Custom models are possible as well.

This suits special data models, company-specific processes and large data volumes. We work out which adjustments your use case needs together after your request.

Is TXNORA a banking app, or does it give tax or accounting advice?

No. TXNORA is an API for teams building financial software and other data-driven products. It is not a banking app and not a tax or accounting service.

Categories and labels describe a transaction in a structured way. They are not a tax or accounting assessment, for example of whether an expense is deductible.

Data & privacy

Where TXNORA processes data and how this website handles your data.

Where is the data processed?

TXNORA processes data on LLM endpoints hosted locally or in the EU. Processing is GDPR-compliant.

Sensitive company data is not passed on to conventional public AI services.

What about data processing agreements, security attestations and availability?

We clarify these points with you individually, for example the data processing agreement, security attestations and availability commitments.

If they matter for your decision, feel free to mention them in your request. We will then address them in our reply.

Does this website use cookies or tracking?

No. This website sets no cookies and uses no analytics or tracking. Fonts are served by the website itself, without connecting to external font services.

For details, see the privacy policy.

Integration

How you connect TXNORA and use the results in your product.

Do I need my own AI infrastructure?

No. TXNORA provides the LLM endpoints, and your system calls them through a simple REST API. You do not need to build or run your own LLM infrastructure.

Processing is scalable, including for large data volumes.

How quickly can TXNORA be integrated?

Quickly: TXNORA works through simple REST API endpoints. Your system sends a request and receives structured JSON that you can use directly in your own logic.

How long the integration takes in your case depends on your system. With trial access and the developer documentation, you can try the integration right away.

What does a request to the API look like?

Your system sends a transaction with amount, currency, booking date, counterparty and payment reference. The response contains merchant, category, labels and a confidence value as JSON.

A simplified example for POST /v1/transactions/classify is on the Developers page.

How should the confidence value be understood?

The confidence value indicates how certain the model is about each assignment. Your product can then decide by its own rules which results to accept directly and which to send to review.

The values on this website, such as 98.4% for the Adobe example, are sample values, not evidence of measured accuracy.

Access & pricing

How to get access to the API and how pricing works.

How do I get access to the API?

TXNORA is a finished product in active use. Send us a short request through the contact form: within 24 hours, you receive free, time-limited trial access to the API, the developer documentation and our pricing. You don’t need an account.

The first step: Request trial access.

How much does TXNORA cost?

Pricing is usage-based, transparent and highly competitive: you pay for the processing you actually use.

Trial access is free. You receive our pricing after your request, together with your trial access.

How does a request work?

Use the contact form to send us your contact details, your use case and, optionally, a short message about what you are planning. Within 24 hours, we will get back to you by email with your free trial access, the developer documentation and our pricing.

How we handle your details is described in the privacy policy.

Next step

Didn’t find your question?

Tell us about your product and your use case. Within 24 hours, you receive free trial access, the developer documentation and our pricing, and we answer your questions along the way.