Exploring what straight through processing (STP) looks like in document process automation requires examining actual examples of successful STP. In this article, we go into detail, providing examples of processes—before and after—that make use of STP.
In Part 3 of our STP series, we discussed the major differences between STP involving simple RPA-oriented tasks and STP involving document-based tasks. If you haven’t read that article, you might want to check it out.
For those that are ready, it makes sense how and why the automation of document-based tasks are different. And yet, significant amounts of automation are still achievable; you just have to apply different measurements.
Mortgage Document Classification and Separation Workflow
The processes involved with originating or servicing a loan involves a lot of paperwork – even if that “paper” is now a bunch of digital documents. From proofs of income and assets to other supporting documentation including appraisals, the number of individual documents can number into the hundreds and sometimes the thousands for commercial loans. Common across all of the various scenarios is the ability to identify one document from another.
In modern mortgage origination process, a lender will receive, often by piecemeal, a number of different documents. Upon receipt, the task turns to verification that the document is correctly submitted (it is the document requested) and meets certain criteria such as a pay stub, which is within the last two months. This process often incurs lag due to when the document was submitted, in what manner it was submitted, and how many documents need to be reviewed.
Another related task is the process of separating one document from another. Lenders and servicers like to deal with individual documents so if more than one document is submitted as a single file, that file is “burst” into individual files where each file represents a single document. This process can take quite a bit of time since the person performing the task must first identify the document and then page through the file and locate the last page, and then create an individual file.
Potential for Error
For both tasks, it is almost always a process of first classifying the document, and then separating it into an individual computer file. For both of these tasks there is a significant opportunity to introduce error in addition to delays. For document classification, a credit card statement showing liabilities may incorrectly be identified as a bank statement showing assets.
This introduces unnecessary delays and customer frustration. For document separation, the credit card statement may be correctly identified, but an error could be made such as incorrectly including the last page showing the balance with another document. This affects downstream processes that rely upon verifying total liabilities, which need immediate access to the page where the value resides.
Manual Processes vs. STP in Mortgage Processing
If these tasks are completed manually, there is no STP since there is no component that is not touched by a person. When looking at this same process with automation, we will use both tasks to introduce measurements. For document classification, the measurement can be a simple percentage of documents classified correctly vs. incorrectly.
We use sophisticated statistical models to measure accuracy and establish a “threshold” that governs which documents have a high likelihood of being classified correctly (often at 98% or 99% accuracy), and therefore, can move through with no manual effort involved.
If a single loan service staff can process 100 documents per hour and the total number of documents received is 100,000 in an eight hour time period, this means that the servicer must employ approximately 125 staff to process them inside an 8 hour workday. With automation measured at the document-level success criteria, we can suppose that a well-tuned system can effectively remove 70% of that work. This reduces total staff time per day from 1000 to 300.
With automation measured at the document-level success criteria, we can suppose that a well-tuned system can effectively remove 70% of that work.
If document separation is involved, it can be a more-complex affair since we now must add a new measurement to the % of documents successfully classified. By definition, the system will not produce more than 70% STP for document classification. This is the absolute maximum presuming 100% STP for document separation; however, this is never the case because there will also be error, even with sophisticated systems. The key is that the error is knowable, and therefore, controllable.
Presuming that the system can accurately separate 85% of documents (again measured at 98% or 99% accuracy) this means that 70% x 80% of work, or about 56% of total effort is eliminated and can go straight-through with no manual review. In this scenario, the time to separate documents reduces the throughput of each staff to around 60 per hour. So total daily staff time is reduced from around 1660 to 733. This sound impressive!
Even more impressive is that, through measurable, controllable automation, we are able to sustain a 98% to 99% accuracy rate of that output. Compared with manual processes that can range in accuracy from 93% to 95%, automation effectively reduces error by around 75% (from 5%-7% error to 1% to 2%).
Accounts Payable Workflow
Accounts Receivable / Accounts Payable (AR/AP), which operates on the data itself, is a bit more involved when measuring STP. As we explained in Part 3, we must measure not by the percentage of invoices that have data accurately extracted, but by the percentage of data that is extracted correctly. This takes us to measuring not by document but by data field; there is no such thing as STP at the invoice level, but significant gains in productivity can still be achieved.
Invoice Data Extraction: Accuracy
This is because, with a tuned system, certain data can be accepted as accurate and not require any data review or entry leaving only a relatively small percentage of data to be managed. For invoices, Parascript software can automate data entry of about 85% of the data fields from a these transactional documents. This leaves only 15% of the remaining data on average for manual data entry.
Let’s say an organization deals with 5000 documents per day. If it takes 25 seconds to locate and perform data entry on 15 fields, automation can reduce the number of fields handled manually to about two resulting in a total daily time savings of almost 30 hours. Again, the key measure is the amount of avoided manual tasks, this time measured at field-level data entry, not invoice-level work.
Intelligent capture software is one of the few applications out there that can offer real cost savings along with improved data accuracy. The key to measuring those gains from a straight through processing perspective is to understand the tasks involved so that measurements can be taken at the individual task level and then rolled-up to understand the real ROI.