Nowadays more and more attention is being drawn to the software business models, also known as SaaS (Software as a Service). Companies save time, effort and money by using out of the box SaaS solutions. For example, it is way more cost and time-efficient to outsource your ERP system than to develop one inhouse, given the fact that you are in a completely different business yourself. Klippa has been in the SaaS business for over 5 years now. All of the solutions Klippa provides have a shared core competency, the inhouse developed OCR and parsing API. We extract data from documents such as invoices, receipts, passports, etc., and deliver them in the desired format. This blog stretches the importance of an invoice parsing API that can convert any given picture or PDF of an invoice into structured data. The aforementioned API has record implementation time of a few hours and is suitable for data analysis, financial institutions, accounting software, and loyalty purposes.
How does the invoice parsing API work?
Parsing invoices consists of a few important steps. Given the speed and accuracy of our OCR engine, the process takes less than 3 seconds. First, a user takes a photo of an Invoice or submits a PDF file to our API. One of the initial checks that will be done is a document quality check. If that meets our criteria and the quality is good enough, the picture is being transformed into a raw text file. Now we are able to check for language and country of origin with algorithms based on years of machine learning data. Having reached that point we are now able to identify all the relevant data such as currency, date, VAT, amount, etc. When all the important information has been identified, a decision engine chooses the correct values, adds categories and converts the data into a JSON output. By default, this JSON output is what the client receives but customization is possible. All it takes is 3000 milliseconds, that’s 3 seconds! Fast right?
What does the process look like?
Following below, you can see a visual presentation of the aforementioned process with the input (Invoice), the transformation of it to raw text form and eventually the JSON output.




Whatever it is that you desire. The customization possibilities of our OCR are numerous. We have listed the out of the box fields below, but on request, we can add any additional fields for your unique case. Input can be JPG, PNG and PDF and the default output is JSON file.
Default Fields:
- Country
- Language
- Invoice number
- Date of purchase
- Merchant name
- Address
- ZIP
- Bank account number
- VAT number
- Phone number
- Website
- Tax Identification Number
- Chamber of Commerce number
- VAT amount per percentage
- VAT percentages
- Total VAT amount
- Total amount, VAT excluded
- Total amount, VAT included
- Currency
- Payment method
- Card number
- Line-items names (products bought)
- Line-items prices
- Line-items category
What languages and countries does Klippa support?
Any language, any country is possible. We are proud of the flexibility of our OCR parsing API. Out of the box, Klippa is optimized for languages like English, Dutch, French, Spanish, Italian, Portuguese, German and more. Do you have a case based on another country or language? Reach out to us. We always love a challenge.
Why parsing invoices is so important?
We are living in a very digital time. Nowadays, many complicated and repetitive tasks are being performed by software. Similar to how robots have been replacing certain human labor in factories, the software is replacing manual repetitive labor in administrative business processes. Every finance department processes invoices. In some finance departments, millions of invoices are processed each month. This means that properly educated people are spending their time on data entry instead of growing your business or reducing your costs. Using OCR to automate your invoice processing will enable your employees to work on what’s important while leaving the simple data entry tasks up to the software. Happy employees, faster processing and a huge cost reduction.
About Klippa
The team at Klippa has automated document processing in their souls. In 2014, when we started with a receipt scanning app, we could foresee ourselves doing the same for invoices. We quickly realized that only through flexibility we could fulfill all of our clients’ different needs with the same consistent accuracy. We are proud to state that our personal approach, speed, accuracy and on-demand fields recognition is our strength.
How to proceed
We love to see the reactions of our clients after we showcase our solutions. Through our website, you can find multiple sections where you can book yourself a free 30-minute demo. During the demo, we can guide you through the possibilities of our engine and you can ask whatever it is that you would like to know.
If all is clear and you still wish to continue you are granted an OCR Invoice parsing Key to test it for yourself! Ready when you are!