What is straight-through processing?

STP can encompass any or all of several automated stages, including initiation, validation, enrichment, routing, fulfillment and settlement. However, STP is not an all-or-nothing proposal; STP is often discussed in terms of STP rate, or the percentage of the process that is automated. Embracing an STP methodology can reduce processing times, which in turn can streamline operational tasks, enable greater scalability and produce a clear, auditable trail of documentation. 

STP is particularly common in financial services, particularly in securities trading, payments, and banking operations. In securities trading, for example, millions of trades each day are analyzed, cleared and executed without human involvement. 

The specifics of an STP transaction can vary, and not all applications will include every step. Let’s look at an example of a retail grocery chain placing an order with its tomato supplier, with the goal of achieving lifecycle STP.

The grocery chain has an automated system that notes that the supply of tomatoes is running low. Based on sales rates and inventory, the grocery chain’s inventory system determines the store needs 10 25-pound boxes of tomatoes.

The system automatically (and in real-time) generates an order and sends the order to the tomato supplier. This order can be sent using standardized business document transmission protocols, such as electronic data interface (EDI).

The tomato supplier receives the order and automatically validates the information therein. Is the pricing correct? Does the supplier have enough inventory itself, and delivery capability? Is the grocery chain up to date on its payments? Are there any existing contracts that need to be checked for accuracy and salience? Does the delivery address match what the supplier already has on file? 

The goal of STP is to answer each of these questions automatically and instantaneously. 

The grocery chain might not include all the information the tomato supplier needs to complete the order; it might simply include quantity, item SKU (stock keeping unit), and delivery address. The tomato supplier will then need to “enrich” that basic order with more information. This could include shipping considerations, such as weight or temperature needs; where the tomatoes are warehoused; where the supplier sources tomatoes for this particular customer and more. 

This data is not necessary for the grocery chain, but is vital for the tomato supplier’s audit trail. 

Routing refers to the process of directing financial transactions through the appropriate clearing channels or payment processing networks. Essentially, determining what payment method will be used and how payment will be delivered.

Methods of electronic payment in an STP system include electronic fund transfers, such as ACH (Automated Clearing House) in the United States or SEPA (Single Euro Payments Area) in Europe. These are commonly used for transactions between banks and other financial institutions, as well as to move money between businesses, consumers and government agencies. They are classic STP methods for things like payroll and vendor payments because files or API calls can be validated, routed, cleared, and settled automatically.

Other methods include card-based payments, instant payment rails such as Real-Time Payments (RTP) and FedNow, and wire transfers.

The tomato supplier can now check warehouse inventory and delivery options and notify the grocery chain with these updates. To achieve 100% straight-through processing, an automatic robotic fulfillment system could be used to collect, pack and ship the tomatoes. But it’s also common to mix manual steps with automated; for example, the supplier can direct its own warehouse employees to pack and ship the cases of tomatoes onto appropriate trucks.

For other STP transactions, fulfillment might simply involve payments. Insurance claims, mortgage payments and financial transactions such as stock trading can all benefit from STP, but do not require a physical fulfillment stage, as there are no physical items involved.

After fulfillment is confirmed, the supplier automatically generates an invoice and sends it to the grocery chain. In turn, the grocery chain automatically receives this invoice and submits payment for accounts receivable via the accepted method (for example, ACH). 

In this entire transaction, humans were involved only in the physical aspects of the tomato purchase: loading, delivering and stocking shelves. The rest of the transaction was completed without the need for human intervention. 

This example is a particularly optimized, automated one. The grocery chain might not have an inventory system that automatically recognizes the need for a new tomato shipment, for example. STP refers to the goal of automating this process. The “STP rate” refers to the percentage of transactions that are processed automatically.

Even if an organization cannot automate the entirety of its processes, it can still benefit from incremental automations and improvements to its straight-through processing system. Key benefits include:

Automation can significantly reduce the risk of human error in data entry and validation. A VAO analysis of data entry in supply chain companies found an error rate of up to 4% for manual data entry, but a maximum of only 1% for STP-friendly automated processes. At scale, this can be a dramatic reduction.

An STP workflow can result in reduced costs. Those reduced costs might come from replacing human labor, reducing the need for paper-based products such as checks, stamps and printing supplies, reducing pressure on customer service representatives through timelier communication and more. A study from Synovus found that the average cost to manually produce one invoice is about USD 10, while those who implemented automation tactics spent just USD 2.81. 

That same VAO study found that the average processing time for a manually produced invoice is roughly 10 to 12 minutes. But that figure snowballs when other elements are factored in, such as document queuing and limited scalability for high-volume periods. The total cycle time, from receipt to completion, was 14.6 days for manual production. With STP methods? About 3 to 5 days.

The automation of individual steps in the transaction process improves and accelerates both that particular step and the process as a whole.

In processes with significant manual intervention, the growth of a business requires proportional increases in staff to handle the new volume of orders. STP can break that cycle. Automated services are often cheaper and easier to scale as order volumes increase. 

STP can increase customer satisfaction by reducing or eliminating many of the pain points in business transactions. For example, notifications, such as shipping updates, are automatically created and sent as soon as they are warranted. These automations reduce processing times and errors, driving more positive customer experiences.  

Automating business transactions and payment systems according to STP goals can help ensure adherence to regulations and minimize security risks. Regulatory compliance can be applied automatically: for example, state tax laws, intellectual property laws and personal data protection laws vary by location but can be automatically applied based on the location of the businesses involved in a transaction.  

STP also enables the implementation of secure transaction protocols, such as EDI, and the incorporation of AI tools that help detect suspicious, and potentially fraudulent, behavior. 

The mere presence of an STP rate, as a metric, shows that while organizations recognize the value of STP as a goal, there are significant obstacles and challenges to complete implementation.

Organizations often rely on legacy systems that can present integration challenges, especially with more modern systems. For example, many older systems weren’t designed to communicate via APIs (application programming interfaces), and such incompatibilities can present barriers to data exchange and integrated workflows. Custom API solutions can be time-consuming and expensive to build, as can migrating to a more modern system.

Automating systems can sometimes require the reworking of established practices that employees are comfortable with. For example, relying on an algorithm to measure, understand and assess risk can seem scary or risky to employees used to making those judgments themselves. 

Implementing an STP system might require time-consuming changes and increased upfront costs in hardware, software and personnel. Teams might need to move data to new platforms, purchase new subscriptions and create custom integrations to connect various data sources and applications. These operations might require the hiring of consultants, additional IT staff and project managers and retraining for existing employees. 

STP is positioned to take advantage of many modern technological advances. Technologies such iPaaS solutions that simplify the integration of applications and the creation of automated workflows are helping organizations attain their straight-through processing goals.

Integration platform as a service, or iPaaS, refers to a suite of self-service, cloud-based tools and solutions used to integrate applications, systems and data sources.

iPaaS products provide a solution to the growing challenge of app, data source and service integration in increasingly complex IT environments. With iPaaS, enterprises can orchestrate integration flows and maximize interoperability between disparate systems by using tools such as pre-built connectors, maps and transformation components.

This centralization and simplified integration makes iPaaS solutions helpful tools for attaining STP goals. In many cases, STP is difficult to achieve because of data exchange limitations: data flows between systems from one step in the process to the next is a fundamental necessity of STP. iPaaS platforms assist developers in building the integrations required to create automated straight-through processing systems.

Artificial intelligence (AI) is already used in many aspects of STP systems. Intelligent document processing, or IDP, relies on AI to read and comprehend handwritten documents, and uses natural language processing to understand them and route the data to the appropriate location. 

AI-powered tools are also used to assess risk, evaluate insurance claims, detect fraud and much more. In banking, AI tools are already used to detect money laundering, match payments to invoices and automatically generate evidence and documentation for auditors.

Blockchain is already used as underlying infrastructure in some fintech applications and presents additional possibilities for future capabilities, such as “T+0” settlement (instant settlement, or near instant settlement).

“T+1” and “T+2” refer to the settlement period for a transaction; in other words, how many business days after a transaction does a settlement occur? T+0 would mean zero days in between transaction and settlement, for instantaneous communication.

In certain designs, blockchain can reduce reliance on traditional intermediaries such as ACH networks and correspondent banks. However regulatory, compliance and integration requirements mean some form of financial intermediaries usually remain involved.

Robotic process automation, or RPA, is not quite AI; it’s more akin to employing a robot to perform a specific task. An AI can take unstructured data, analyze it and adapt to patterns; RPA is more about data entry and other specific tasks. RPA and AI often work together, as a sort of “hands and brain” combination.

In finance, for example, AI can be used to analyze transaction data and assess whether a fraud is taking place. RPA then takes that conclusion and can block transactions for a fraudulent account. This combination is sometimes known as intelligent automation, or IA, and is likely to continue to grow in complexity, ability and popularity in the future.