From time to time it’s important to revisit the foundation of a concept, especially a burgeoning one. With respect to document processing and how it’s always been done, we feel that now is a pertinent time to challenge the status quo and define what document process outsourcing is and why it’s so powerful. Since we’re dealing with static, paper-based processes, the general approach to automating cumbersome processes is generally the same.
1. Front-end capture
Any meaningful document process outsourcing effort is going to be driven by converting documents to data from the process outset. However these documents arrive, whether physically or digitally (PDF attachments or Tiff faxes) is immaterial...either way they need to get sucked into an advanced OCR process. This means, no template based approach since that will just get tripped up as soon as a vendor shifts their document format around. Rather, initiate with a layout and learning based recognition model and extract all relevant details off of the document. In the event, you can’t pull what you need from the document, boom, it needs to kick to an exception queue for human intervention since requisite data to process is unavailable.
2. Exception handling
Part of good document processing automation is the ability to interface with documents digitally in a manner that parses data into a structured database format. Folders based storage mechanisms aren’t up to snuff because items can easily get mis-filed, lost, and be irretrievable. Therefore, by utilizing status queues and business rules based upon this meta-data, you can force documents that contain exceptions to be handled by processing staff who are familiar with the process nuances and have the means to supply the missing data or who know how to interface with internal or external stakeholders to resolve the exception.
Additionally, assuming all data is present and accounted for to advance a transaction (whether invoice processing, sales order processing, remittance processing, or anything else for that matter), it may be necessary to get an authorized digital approval for large dollar or critical documents. This is achievable through electronic workflow and approval processes, and provides a convenient audit trail for future reference.
4. Back end integration
Now comes the tricky part...in some cases. Obviously the goal of digitizing and automating the data capture effort is to divest oneself of the burden of entering that data into the back end system. By doing so, you can you eliminate between 60-70% of the labor effort required to perform the process. Additionally, because that effort is gone, the process is accordingly expedited, and you pick up the benefit of enhanced visibility, provided you have a decent outlet for reporting and dashboarding. Integration to back end systems can range from direct methods to APIs to ingestible file formats to robotic entry technology. Whatever the means, it is the fundamental pivot point of the process that delivers massive value. Additionally, it’s an equally important step besides the data depositing aspect to match and reconcile data against database tables within the erp. In the case of order processing this could involving product codes, descriptions and pricing, and on invoice processing can include references against vendor master data or PO data tables (as we do in our typical Accounts Payable process automation scenario).
Without these components it’s near impossible to have meaningful insight to the process. This means, you can’t ferret out bottlenecks or use latent process data to affect positive business outcomes. As a case in point in terms of invoice processing, you might miss the capture of an early payment discount by not turning around a payment to a priority vendor in a timely fashion. Alternatively, it may cost you money if you miss a payment and get penalized for late payment because you couldn’t unearth an invoice fast enough. Another impact of visibility is upon ancillary procesess associated with data retrieval including audits or even monthly accrual and forecasting functions. All of these can be positively affected through dynamic visibility, another key facet of document process outsourcing.
All of these are impactful components to a successful document process outsourcing campaign, but the bedrock element to DPO is the connection of cloud based technologies overlaid by efficient and affordable human processing. The key nuance of DPO is that it empowers organizations who have not had the financial wherewithal to invest in automation technologies to do so by leveraging a SaaS infrastructure and transactional pricing model.
We’ll dive more into DPO in the coming weeks when we explore processes that are compatible with this approach and how they can look pre and post automation.