Data Action Layer December 16, 2021

Automating the Whole Nine Yards of Invoice Processing

by Patrick Vergara

61ba423425c15b4c27b080fe Blog Image 8 p 800

Extract, transform, and load (ETL) has been an essential data integration process for the IT industry since the 1970s. It refers to the three steps that have traditionally been used to integrate data from multiple sources into another, separate data repository. During this process, data is taken (extracted) from a source, converted (transformed) into a format that can be easily analyzed, and stored (loaded) into a data warehouse or other system.

This three-step process is not just reserved for IT. Many departments within an organization—including HR, marketing, sales, and finance—go through similar stages to process new hires, onboard customers, generate reports, and more. For the finance department, in particular, this is a core part of their daily job. So, it comes as no surprise that they’re open to automating this process.

The accounts payable example

Imagine a mid-sized financial services firm seeking to improve its manual accounts payable (AP) process. The business receives more than one thousand invoices every month from dozens of vendors. Most of these invoices are sent to an AP email address, while about 15% are manually uploaded to a shared Microsoft OneDrive repository by the regional finance manager. Next, the accounts payable team must manually open each invoice, locate the necessary data from the invoices—like vendor, amount, date, and terms—and manually key in the details into the company’s QuickBooks system for tracking and processing.

This business would greatly benefit from automating this entire process—especially if they want to eliminate the costly errors that frequently arise during the data entry stage.

Automate with Alkymi and Zapier

The process listed in the example above can be easily automated, leveraging Alkymi Data Inbox’s prebuilt connectors to thousands of enterprise applications powered by Zapier. Let’s take a step-by-step approach to see how this company might go about implementing this solution.

To begin, the financial services team would want to centralize all invoices to enable fast data transformation, especially if they’re stored in different places. Invoices that come in via email can be pulled into Data Inbox by leveraging Alkymi’s native Microsoft Exchange integration. A pre-built connector on the Zapier platform can be activated to collect the invoices uploaded to the Microsoft OneDrive repository and bring them to Data Inbox. In less than 20 minutes, the entire inbound data flow can be automated, and Data Inbox will be set up to receive all of the invoices automatically.

Now that your invoices are in one place, you’re ready to isolate the necessary data points and normalize them for your payment system. A fast way to get started with data extraction is to leverage a pre-made Alkymi Patterns blueprint for invoices. You can leverage it “as is” or make a few tweaks before turning it into automated Patterns. Alkymi Patterns intelligently identifies documents as invoices based on their structure and content, and it extracts the key pieces of information needed to fulfill the accounts payable process. The vendor’s name, amount of invoice, and terms of payment can all be captured automatically. Patterns even supports the processing of handwritten invoices and invoices written in multiple languages.

Next up is the data review. Data Inbox is designed with business users in mind, making it very simple to view the extracted data and full invoice side-by-side. Documents that contain missing or incorrectly formatted data are visually flagged so that employees can review and fix the erroneous fields.

The final step in the business workflow is moving the data—putting it into the accounting system. Once the data has been reviewed and approved, a pre-built connector template from the Zapier platform is used to send the extracted data from Data Inbox to QuickBooks and create an invoice entry and initiate payment.

The impact

Using Alkymi to collect, centralize, extract, and normalize invoice data with Zapier’s app connectors delivers not only an end-to-end solution for accounts payable automation but also for many other processes where data extraction, transformation, and loading is required. In this example, a manual invoice management process that took numerous hours away from employees is now done in minutes— with significantly fewer errors.

Ready to speed up your process? Try Alkymi for free today.

More from the blog

September 22, 2023

A high-performance approach to personalizing an LLM

by Elizabeth Matson

Fine-tuning is not the only way to get relevant, domain-specific responses out of an LLM. Alkymi’s team of expert data scientists explain an alternate route.

September 6, 2023

IDP: find the right document processing solution for your business

by Bethany Walsh

Find out which type of automated document processing solution is right for you: data extraction, an IDP, or a complete business system for unstructured data.

August 29, 2023

Alkymi and Portfolio BI partner to empower alternative asset managers

by Elizabeth Matson

We’re partnering with Portfolio BI, a provider of portfolio analytics and reporting solutions, to bring structured and unstructured data sources together.