With automated invoice processing powered by advanced machine learning, all your invoices can be automatically coded, assigned to the appropriate workflow, and routed electronically for approval – no matter how many you have. Once you approve, invoices flow directly into your accounting system for payment processing. When an imported invoice can't be submitted to workflow successfully, the system will remove it from further automated processing. An accounts payable clerk can review and edit the invoice before the automated process attempts to rerun the process for that invoice. Currently, unsuccessful invoices must be resubmitted for processing one at a time. This feature lets you restart processing for multiple invoices where the reason for the failure can be resolved by a single fix for those invoices. Invoice processing automation involves transforming paper invoices into electronic formats by using scanners that digitize the paper documents and apply optical character recognition (OCR) to enable the digital documents to be read, routed, manipulated, approved, and stored electronically. AI Builder Invoice Processing Automation Starter Kit is a reference implementation of invoice processing automation provided by Microsoft to bootstrap your implementation. This is an installation package including two Power Automate flows orchestrating the processing of invoices with AI Builder, several Common Data Service entities to store the information, a role and a Power App to customize the starter kit and monitor the processing and edit missing or inaccurate extracted data. Nanonets Invoice Automation allows AP teams and businesses to process invoices quickly and more efficiently. Supercharge your procurement process and strengthen relationships with suppliers. Here are some more reasons for going with Nanonets invoice automation: Smart 3-Way Matching.

Automated invoice processing is the process of seamlessly extracting data from invoices entering your system and pushing it into your ERP so that processing a payment is a process of a few clicks. This blog is a comprehensive overview of the latest technologies that can enable you to do that.

Processing payments for finance departments is a chaotic process with invoices coming from in every format possible: paper invoices, PDFs, scanned PDFs, emails, etc. The volumes also fluctuate with seasons where peaks become extremely hectic for finance personnel. Invoice capture using manual data entry into systems during these times lead to common issues like lost documents and tally mismatches.

In this blog we talk about what an ideal automated invoice processing system looks like and what are the pros and cons of such a system.

Want to build your custom Automated invoice processing workflow? Head over to Nanonets and get started for free!

Invoice Automation Software

The Procure to Pay cycle typically involves the following steps:

There are 3 steps in any invoice processing:
- Scanning the physical invoice
- Interpreting the invoice data according to business rules
- Filing of data into any ERP
Now there are two kinds of invoice processing softwares in the market :

  • Rule-based OCR softwares like Kofax, Tipalti, Abby
  • Intelligent OCR softwares like Nanonets, Rossum

The software should intelligently extract the required information and validate it according to business rules like mapping to a Purchase order number before pushing the final details into the ERP for payment processing. The rule-based systems don't scale well as businesses keep adding/churning vendors and writing new rules everytime is an inefficient process.

These softwares also offer integration with most ERP softwares like SAP, Oracle, Microsoft Dynamics 365.

Automated Invoice Processing Workflow

Any automated invoice processing software needs to run a 3-way match purchase order generated, invoices received against purchase order and receipts generated against invoice. Each of these documents has a unique number which needs to mapped against each other.

Here is an example of an automated invoice processing flow:

  1. Invoice arrives in specific email folder as an attachment (JPEG, PDF, EDI, PNG etc)
  2. The document enters under 'Review' tab
  3. Invoice automation software extracts data from attachment
  4. Validate extracted data of vendor invoices with business rules from Master Vendor file
  5. If extracted values pass all validation rules, move to 'Verified'
  6. If extracted values don't pass validation rules, documents remain in 'Review' section unless they are manually reviewed and moved to 'Verified'
  7. Final report is generated containing all invoice data
  8. Export as .csv or integrate API with ERP system

Nanonets OCR API has many interesting use cases. Talk to a Nanonets AI expert to learn more.

Invoice Cognitive Capture

Using the latest advances in artificial intelligence coupled with the reduction in compute costs, Nanonets invoice OCR solution is able to capture invoice data regardless of the format. You no longer need to setup rules

Table Extraction

This is the phase where the information is extracted after the tables are identified. There are a lot of factors regarding how the content is structured and what content is present in the table. Hence it’s important to understand all the challenges before one builds an algorithm.

  • Dense Content: The content of the cells can either be numeric or textual. However, the textual content is usually dense, containing ambiguous short chunks of text with the use of acronyms and abbreviations. In order to understand tables, the text needs to be disambiguated, and abbreviations and acronyms need to be expanded.
  • Different Fonts and Formats: Fonts are usually of different styles, colors, and heights. We need to make sure that these are generic and easy to identify. Few font families especially the ones that fall under cursive or handwritten, are a bit hard to extract. Hence using good font and proper formatting helps the algorithm to identify the information more accurately.
  • Multiple Page PDFs and Page Breaks: The text line in tables is sensitive to a predefined threshold. Also with spanning cells across multiple pages, it becomes difficult to identify the tables. On a multi-table page, it is difficult to distinguish different tables from each other. Sparse and irregular tables are hard to work with. Therefore, graphic ruling lines and content layout should be used together as important sources for spotting table regions.

Benefits

  • Costs: Reduce actual keystrokes typed by almost 90% and take advantage of discounts on early payments
  • Productivity: By removing the task manual tiresome data entry, your finance team does the more important tasks 75% faster
  • Analytics: Using machines you can now extract a lot more data a lot faster leading to deep analytics into business processes inefficiencies and bottlenecks
  • On-premises: We understand this is sensitive data to your business. Nanonets AP Automation software can run on-prem

Evolution of the invoicing process

The process of reviewing invoices has evolved a lot over time. The growth in technology has seen the process of invoice processing move through three major phases.

Phase 1: Manual Reviewing

Consider a use case where an organisation is going through it's process of reimbursing its regular vendors for the expenses of the month.

Automated Invoice Processing Software

The following steps are followed to process invoices -

  1. People are expected to submit several invoices in person to the concerned organisation's point of contact.
  2. This person would in turn forward all the invoices to a reviewer who will review every document entirely. This includes writing down or entering each detail into a software like name of the person making the purchase, name of the store purchased from, date and time of purchase, items purchased, their costs, discounts and taxes.
  3. The sum total of each invoice calculated, again manually or if the data entry software is specifically designed for accounting purposes, using said software.
  4. A final bill/receipt is made with the final figures and the payments are processed.

Phase 2: Invoice Scanning and Manual Reviewing

With the advent of OCR techniques, much time was saved by automatically extracting the text out of a digital image of any invoice or a document. This is where most organisations that use OCR for any form of automation are currently.

  1. Digital copies of invoices are obtained by scanning invoices or taking pictures using a camera.
  2. The text is extracted from these invoices using OCR. This is able to provide digital text that makes data entry a little easier. But a lot of work still needs to be done manually.
  3. The OCR results of each invoice have to be parsed appropriately to find the relevant data and discard the irrelevant data.
  4. Once this is done, the data has to be entered into a software which provides the reviewer with a template to make his task easier. This template is unique to each use case, organisation and mostly for each different kind of invoice. While the OCR process helps the invoice processing, it doesn't solve many tedious parts due to the unstructured results of OCR.
  5. The data entered is put through manual review to correct errors. This process takes some time since it goes through multiple reviewers due to poor performance of currently available OCR tools.
  6. Finally, the calculations are done and the payment details are forwarded to the finance division.

How to digitize invoices better?

By using OCR and deep learning, we have enabled machines to perform as well and in some cases even better than humans.

Digitizing invoices involves several human moderated steps :

  1. Digital images of invoices taken and uploaded by the user.
  2. Image verified to be fit for further processing - good resolution, all data visible in the image, dates verified, etc.
  3. Images checked for fraud.
  4. Text in these images extracted and put in the right format.
  5. Text data entered into tables, spreadsheets, databases, balance sheets, etc.

Phase 3: Deep Learning and OCR

Deep learning approaches have seen advancement in the particular problem of reading the text and extracting structured and unstructured information from images. By merging existing deep learning methods with optical character recognition technology, companies and individuals have been able to automate the process of digitizing documents and enabled easier manual data entry procedures, better logging and storage, lower errors and better response times.

Several tools are available in the market and the open-source community for such tasks, all with their pros and cons. Some of them are Google Vision API, Amazon Rekognition and Microsoft Cognitive Services. The most commonly used open-source tools are Attention-OCR and Tesseract.

All these tools fall short in the same manner - bad accuracy which requires manual error correction and the need for rule-based engines following the text extraction to actually be able to use the data in any meaningful manner. We will talk more about these problems and more in the coming sections.


Validation

The data extracted from any invoice needs to be verified to ensure proper sums, VAT rates, VAT IDs, currencies, seller details and more. Most companies have workflows of multiple levels of approvals to ensure that an error doesn't slip through it's cracks where reconciliation later is a very painful process. A good invoice processing automation software allows you to setup some of these basic rules and in case any extracted value violates that, it raises a flag for a manual review. So now you're adding a layer of Artificial intelligence led checks. In these redundant repetitive processes, machines have been said to been humans hands down.

Handling mismatches

In case there is a discrepancy in the information extracted against said business rules, one can create a resolution handling workflow where there is manual verification of this data. You can set confidence score thresholds to handle such automation - if a value is below a pre-decided confidence score number you route to a manual workflow, else push it to the automated payment workflow.

How Nanonets can help you

Nanonets is an artificial intelligence based OCR engine that helps business across the world automate manual extraction. Our AP Automation software makes monthly invoice processing a hassle-free process. Using invoice OCR API you can cut down upto 90% of manual tasks of data entry into your accounting softwares. We are a valued partner of over 20,000 businesses in 4 continents.

The Nanonets OCR API allows you to build OCR models with ease. You do not have to worry about pre-processing your images or worry about matching templates or build rule based engines to increase the accuracy of your OCR model.

You can upload your data, annotate it, set the model to train and wait for getting predictions through a browser based UI without writing a single line of code, worrying about GPUs or finding the right architectures for your deep learning models. You can also acquire the JSON responses of each prediction to integrate it with your own systems and build machine learning powered apps built on state of the art algorithms and a strong infrastructure.

Using the GUI: https://app.nanonets.com/

You can also use the Nanonets-OCR API by following the steps below:

Step 1: Clone the Repo, Install dependencies

Step 2: Get your free API Key
Get your free API Key from http://app.nanonets.com/#/keys

Step 3: Set the API key as an Environment Variable

Invoice Automation Software

Step 4: Create a New Model

Note: This generates a MODEL_ID that you need for the next step

Step 5: Add Model Id as Environment Variable

Note: you will get YOUR_MODEL_ID from the previous step

Step 6: Upload the Training Data
The training data is found in images (image files) and annotations (annotations for the image files)

Step 7: Train Model
Once the Images have been uploaded, begin training the Model

Step 8: Get Model State
The model takes ~2 hours to train. You will get an email once the model is trained. In the meanwhile you check the state of the model

Step 9: Make Prediction
Once the model is trained. You can make predictions using the model

Further Reading

  • Overcoming the Top Pitfalls of Manual Invoice Processing
  • In-depth Guide to Automating Invoice Processing in 2021

You might be interested in our latest posts on:

Update:
‌ Added more reading material about different approaches using OCR in automating invoice processing workflows

Start using Nanonets for Automation

Try out the model or request a demo today!

TRY NOW-->

This topic describes the capability for automating your vendor invoice processing and the benefits of using an automated process. This capability consists of features that are turned on in Feature management. These features apply only to vendor invoices, not to invoices that are processed using the Invoice journal or Invoice register journal page.

Organizations often work with third parties to process paper invoices by using an optical character recognition (OCR) service provider. The service provider returns machine-readable invoice metadata. To help with automation, the Accounts payable automation features let you consume these artifacts from Accounts payable.

You can automate some Accounts payable vendor invoicing processes. These processes include submitting imported vendor invoices to the workflow system and matching posted product receipt lines to pending vendor invoice lines. The automated process shows information about the progress of a vendor invoice as it moves through each of the processes. This capability can help Accounts payable clerks and managers process vendor invoices more efficiently. It also helps reduce the errors and inefficiencies that can occur when information is manually entered and processed.

Invoice Automation Tool

The automation processes can be used to perform these tasks:

  • Automatically submit imported invoices to the workflow system.
  • Match product receipts to pending vendor invoice lines.
  • Simulate posting before a vendor invoice is posted.
  • Quickly and efficiently view workflow and automation history.
  • View and analyze the results of automating vendor invoice processing.
  • Resume automated processing for multiple invoices.

Submit imported vendor invoices to the workflow system

As part of a touchless Accounts payable invoicing process, you can have the system automatically submit an imported invoice to the workflow system. The process will run in the background, at a frequency that you specify (either hourly or daily). The capability to automatically submit imported invoices to the workflow system requires that your process begin with an imported invoice. To ensure that the invoice can be processed from start to finish without manual intervention, an automated posting task must be included in the workflow configuration.

Invoice Automation

Invoices that are related to purchase orders (POs), and invoices that contain a non-PO procurement category and non-stocked lines, can automatically be submitted to the workflow system. Invoices that are manually entered and invoices that are created using the Vendor collaboration invoicing workspace must be manually submitted to the workflow system. Prepayment application processing must be performed manually for imported invoices. You can manually apply prepayments before or after posting the imported invoice. You can manually apply prepayments to unposted standard invoices using the Vendor invoices page. After posting, the settled prepayment will be available to manually apply to other invoices from this vendor on the Vendors page (Accounts payable > Common > Vendors > All vendors > Invoice tab > Apply).

The automation feature provides a flexible framework that lets you define company-specific rules for submitting imported vendor invoices to the workflow system and matching posted product receipt lines to pending vendor invoice lines.

Match product receipts to invoice lines that have a three-way matching policy

The system can automatically match posted product receipts to invoice lines that a three-way matching policy is defined for. The process will run until the matched product receipt quantity equals the invoice quantity. As part of this process, you can specify the maximum number of times that the system should try to match product receipts to an invoice line before it concludes that the process failed. The process will run in the background, either hourly or daily. You can run the automated matching process as part of the process for submitting invoices to the workflow system. Alternatively, you can run it as a standalone process.

Pre-validate vendor invoice posting

Posting simulation completes the validation steps that are done during the posting process for vendor invoices, but no accounts are updated. To run the process, you can select either a single invoice or multiple invoices on the Pending vendor invoices page.

Enhanced experience for viewing workflow and automation historical information for vendor invoices

An easy-to-read view of vendor invoice workflow history is provided. Vendor invoice workflow history can be accessed directly from the vendor invoice. Therefore, fewer clicks are required to find that information. If your organization has enabled the ability to automatically submit imported vendor invoices to workflow, the automation history is provided for the imported invoices. The automation history helps you identify the current process step, as well as the steps that have already been completed. When a step is unsuccessful, the system provides detailed information to help you understand the reason for the failure.

Analytics and metrics

The Vendor invoice entry workspace lets you focus on vendor invoices that didn't make it through the automated process. Tiles on the workspace list information about vendor invoices that weren't successfully submitted to the workflow system, imported, or matched to product receipts. Microsoft Power BI metrics are also provided to give Accounts payable managers insight into the efficiencies of vendor invoice automation.

Resume automation processing for multiple invoices

When an imported invoice isn't successfully submitted to workflow using the automated process, the system will remove it from further automated processing. An accounts payable clerk can review and edit the invoice before the automated process resubmits it to workflow. When a failure reason can be resolved by the same fix for multiple invoices, you can restart the automated process on the Resume automated invoice processing page.

Tracking the Invoice received date value

The Invoice received date value indicates the date when the company received the invoice from the vendor. It provides a starting point for tracking the invoice's progress through the automation processes. This value can be included in the imported data for a vendor invoice. For invoices that were manually created, you can specify the date. If no value is entered, the current date is used by default.

Tracking the Imported invoice amount and Imported sales tax amount values

The Imported invoice amount and Imported sales tax amount values for vendor invoices can be provided in the vendor invoices import file. Typically, these values are from an invoice that was scanned by an outside provider and included in the import file. As the invoice is processed in Accounts payable, the system calculates values based on the invoice data. The invoice can be posted only if the imported values match the calculated values. Matching values ensure that the invoice accurately reflects the amount that is due to the vendor. If your organization allows imported invoices to be submitted to the workflow system automatically, you can optionally require that the imported totals match the calculated totals before the invoice can be submitted to the workflow system.

Vendor invoice automation - Resume automation processing for multiple invoices

When an imported invoice isn’t submitted successfully to workflow through the automated process, the system will remove it from further automated processing. An accounts payable clerk can review and edit the invoice before the automated process resubmits it to workflow. When a failure reason can be resolved by the same fix for multiple invoices, you can restart the automated process on the Resume automated invoice processing page.

Coments are closed

Most Viewed Posts

  • Download Google Drive Stream Mac
  • Merge Multiple Cells In Excel Without Losing Data
  • Merge Excel Sheets Into One New Sheet

Scroll to top