Model what you bill: A new approach to commercial banking pricing

Commercial banks invest significant time and resources in repricing and revenue optimization initiatives. Yet many still rely on spreadsheets to model changes that their production pricing and billing systems already calculate every day. This disconnect can lead to inaccurate forecasts, operational inefficiencies, and missed revenue opportunities.

Early in my career, I was brought in as a consultant to work on an annual commercial services repricing event at a major bank. As someone new to banking, I welcomed the opportunity to work on the analytics for such an important project. It was a major initiative with significant potential impact on the bank’s future earnings. It seemed like a great opportunity to build my professional skills. As I quickly discovered, it also taught me to build resilience under pressure.

The bank had a flexible, powerful commercial pricing and billing engine, with which I was familiar. However, the repricing event was essentially an effort to recreate the pricing capabilities of that engine in various spreadsheets to forecast the effect of proposed changes.

The challenge with annual repricing

We spent months analyzing the current pricing models, inputting changes into spreadsheets, and checking the anticipated results. It was a lot of long nights, and a lot of pure guesswork. Management eventually signed off on a set of changes, which were then manually put into production by the Operations team from the same spreadsheets we’d used for analysis.

A few months later, I was brought back to analyze the impact of the changes in anticipation of the next annual repricing event. The forecasts from our previous effort had been off by tens of millions of dollars. The spreadsheets did not account for the impact of existing exception pricing, complex pricing (minimums, maximums, tiers, flat billing arrangements), relationship pricing, bundles, account waivers, percentage discounts, and changes in earnings credit rates.

The old hands at the bank told me to just quantify the differences in actual versus forecasted as best I could, and to get ready to do it all again in a few months.

Why forecasting falls short

As a numbers guy, the process made absolutely no sense. We were running scenarios, potentially affecting hundreds of thousands of commercial accounts, using spreadsheets we knew could not actually reproduce the full breadth of functionality the pricing and billing engine was capable of. Our previous results had been completely inaccurate. The next round of results would also be completely inaccurate but, as I was repeatedly told, this was the “best we could do.”

Over the course of my career, as non-interest income became a growing component of overall bank revenue, the tolerance for inaccurate forecasting steadily decreased. That led to me working on multiple projects to improve the effectiveness of repricing events. Lots of time and effort were poured into creating tools that better mirrored the calculations native to the pricing and billing engines, with better approval processes and with more automated implementation of approved changes—these were costly, resource-intensive projects.

None of those projects ever yielded the desired results. Eventually, I came to realize that not only had those efforts failed, but that they would always fail. The basic premise of each was fatally flawed because we assumed that it was somehow possible to effectively emulate the functionality of an enterprise pricing and billing engine outside of that system.

Furthermore, even if it were possible to do so, why should a bank spend piles of money replicating what an in-house system was already doing? The real breakthrough came when I stopped asking how to better replicate the pricing engine and started asking why we were doing it at all.

A different way to think about modeling

I concluded that the only way commercial banks could accurately and cost-effectively model proposed scenarios (at the enterprise or customer level) was to run those scenarios on their actual pricing and billing engines.

As a software developer, that realization completely reframed my approach to my own application, Infor™ Complete Billing System. Instead of trying to rebuild core system functionality, my focus shifted entirely to building scenario management tools, analytics, pro-formas, workflows, customer disclosures, and automated implementation methods that worked with modeling instances of Infor Complete Billing System that my bank partners were already running in production.

The results were revolutionary. Gone was the need to try and reproduce complex calculations in some kind of subsystem. Gone was the need for months of trading spreadsheets back and forth, relying on analysts to apply manual review of proposed changes. Gone was the need for bank commercial customers to rely on generic disclosures of pricing changes—that may or may not even apply to them. The same system that produces their monthly invoices can now produce an accurate pro-forma that highlights the actual changes impacting their banking relationships. Manual implementation was no longer necessary. What the bank modeled, and approved, was what the bank implemented in a straight-through process.

Perhaps, most importantly, banks were no longer limited by the capabilities of their modeling subsystem. Need to run scenarios based on implementing new types of hybrid (interest-paying) accounts? Want to check the impact of possible key rate changes in the coming months? Or the impact on major customers of implementing new methods of earnings credit sharing and credit carryforward? If the bank’s pricing and billing system can do it, then the bank can model it.

Closing the analytics gap

In retrospect, using what you have instead of building something new seems an obvious conclusion. Unfortunately, commercial banking has taken entirely too long to get here. My partner banks are no longer in the business of trying to rebuild a technology platform they already have in production. Instead, their focus is on implementing tools and procedures that maximize their existing technology investments. The results are appreciated by employees, shareholders, and the corporate customers of the bank.

When banks model directly in their system of record, they can achieve:

• More accurate forecasting
• Reduced manual effort
• Better client-specific pro formas
• Faster implementation of approved pricing changes
• Greater confidence in revenue projections

Closing the final gap with Infor Complete Billing System

With the paradigm adjusted to use the same system to run modeling statement calculations, the last missing piece was analytics: a way to show the delta between the modeling and production systems, and to provide a visual, easily navigable metaphor that relationship managers can use to tune the model with confidence.

From there, banks can produce pro formas and client communications to introduce the coming changes tailored specifically to each account. To bring the entire modeling process together into a cohesive end-to-end solution, Infor has developed the Bank Modeling Tool as a key new module of Infor Complete Billing System.

See the approach in practice

Read how a leading commercial bank used Infor Complete Billing System Bank Modeling Tool to optimize pricing across 1.5 million accounts and improve profitability.