Rossum Integration (1): Accurate Document Data Capture is Essential to Your ERP

Rossum modernizes and simplifies document processing and management. We designed the platform to increase the efficiency and effectiveness of your company’s business processes. While the platform’s benefits are clear when used on its own, it truly shines when you integrate it into your company’s enterprise resource planning (ERP) system. In the coming weeks, several of our partners will share their Rossum integration experiences, and demonstrate the value it has delivered to their customers – value that it can also deliver to you.

Before we get to their stories, it’s important to understand just how vital accurate data capture is to ERP systems.

ERP is data-driven

If your ERP system is the heart of your organization’s processes, then data is its lifeblood. The quantity of data it requires to flow is staggering; the accuracy of that data determines how well the ERP serves your organization. Timing is also a key factor in your ERP’s usefulness – many business processes, such as procurement, delivery, and accounts payable (AP), depend on rapid, or even real-time, document processing.

enterprise information system

Because your ERP system is modular, each of your business units is inputting data from a variety of sources, both internal and external. This contributes to the complexity of ERP solutions, which require specialists to ensure they’re set up to properly serve the organizations that purchase them.

ERP systems are expensive 

The centralization of business processes helps ensure smooth cross-functional workflows, which is crucial to enterprises that have an extensive network of offices or branches, or that consist of a group of companies. This capability comes with a hefty price tag, as recent contracts with the US Army and the city of Oxnard, California demonstrate.

When you invest in an ERP solution you need data entry that matches the high standards you expect of your costly ERP system. If, for instance, you replace a decades-old legacy system to help your organization go paperless, but use time-consuming manual or traditional OCR data capture methods, it’s going to be a while before you see a return on investment (ROI) – if you see one at all. 

GIGO can ruin companies – just ask Target Canada

Making matters potentially worse, inaccurate data capture will inevitably lead to more than a classic garbage-in-garbage-out scenario – garbage going in can potentially grow worse as it passes through your costly ERP system, potentially robbing you of any hope of ROI.

In 2015, Target Canada became the poster child of manual data entry and validation disaster. While the GIGO in its ERP system was one of several factors that led to the American retail giant’s spectacular failure north of the US border, ultimately racking up losses of USD $2.1 billion in just two years, it is arguably the most significant. As detailed in Joe Castaldo’s fascinating feature article in Candian Business magazine, The Last Days of Target:

“… merchandisers were under pressure to enter information for roughly 75,000 different products into SAP according to a rigid implementation schedule. Getting the details from suppliers largely fell on the young merchandising assistants. In the industry, information from vendors is notoriously unreliable, but merchandising assistants were often not experienced enough to challenge vendors on the accuracy of the product information they provided. ‘There was never any talk about accuracy,’ says a former employee. ‘You had these people we hired, straight out of school, pressured to do this insane amount of data entry, and nobody told them it had to be right.’ Worse, the company hadn’t built a safety net into SAP at this point; the system couldn’t notify users about data entry errors. The investigative team estimated information in the system was accurate about 30% of the time. In the US, it’s between 98% and 99%.”

An excess of data entry errors played a critical role in the death of Target Canada.

Of course, it’s easy for us to look back on this and say Target Canada could have saved themselves this particular grief by integrating a document processing solution with machine learning into their SAP instance to extract product data accurately. This would have helped solve purchasing errors that resulted in empty shelves and disappointed customers. But, as the article makes clear, as significant as it may have been to Target Canada’s downfall, slow and inaccurate data entry was not the only problem the retailer had to contend with.

Automated invoice processing

At the moment, AP teams are the primary Rossum users. Their typical workflow, from invoice receipt to vendor payment, is a four-step process:

  1. Receive, scan (when necessary), and store invoices
  2. Capture invoice data, human validation when necessary
  3. Export invoice data to ERP, accounting, or document management system
  4. Approve and pay invoices
Automated data capture is a vital part of the AP process.

Of these, the second step is where Rossum’s machine learning comes into play. With its out-of-the box capabilities to read and understand thousands of invoice formats, the platform can process a huge quantity of invoices in minutes. Members of your AP team can validate the data just as quickly; during this stage, Rossum learns new information based on additions or corrections for subsequent invoices.

Rossum can do so much more

In addition to extracting data from invoices, we can train custom Rossum AI models to capture whatever type of information that your ERP system requires. The platform can learn to capture data from product information documents (cough… Target Canada… cough), delivery notes, bills of lading, or any other paper or digital document. The only conditions are that they have Latin characters, tables must be in grid format with uniform columns.

AI for GIDGO

We haven’t really got an antonym for GIGO… Perhaps GIDGO – Good Input Delivers Good Output could do (if you come up with, or know of, something better, please share it with us!). Anyway, advances in AI technology are helping forward-thinking companies ensure accurate data entry into their ERP systems. Powered by its unique neural network Rossum’s easy-to-integrate data capture solution has enabled several companies and ERP consultants to improve operational efficiencies quickly and easily. The platform’s assured GIDGO has also proven to deliver fast ROI.

Rossum integration stories

Over the next few weeks, you can look forward to the experiences and insights that several of our partners have been kind enough to share with us. You’ll see the ease with which companies and ERP consultants have integrated Rossum into ERP systems, and discover how they’ve benefited from doing so.

Leave comment