These goals can be achieved with transaction profitability analysis (TPA).

Transaction Profitability Analysis

Your business is engineered to transform raw inputs, deliver products and services, and garner fees. Firms revolve around customers buying, using, returning, complaining, inquiring, or needing service on some product through some channel at some location at some point in time.

Your corporate P&L is a highly summarized result of millions of these transactions. Managing a firm toward improved financial results is like training an artist; it can only be done at the level of each brush stroke of each transaction. To understand and manage the traditional corporate P&L, you must have P&Ls for each transaction from which the corporate P&L is derived.

TPA traces the impact of each customer-company transaction through every operational function of the business, links the costs in those functions with the revenue derived from the transaction, and classifies that transaction according to the customer, product, channel, and location that generated it. TPA allows the myriad transactions to be summarized along any line — customer, product, channel, time-period, or any combination. TPA provides seamless multidimensional drill-down on profitability from the total company to the finest transaction detail, where operational decisions are executed.

TPA is integrated operational and financial data; TPA is operationally relevant financial information.

Three Prerequisites

Three specifics prevent finance from being operationally relevant:

1. Granularity: The categories are so broad that they combine things that are dissimilar;

2. Precision: The assumptions regarding cost causality are so broad that they do not differentiate between things that have different operational implications; and

3. Timeliness: The information is too old.

Granularity

There are many ways of categorizing data, but three generate most of the conflict between finance and operations reporting: product, geography, and customer segment. For instance, the marketing department of a consumer firm spends millions of dollars promoting a product to a certain demographic in a certain city; let’s say unlimited-user wireless instant messaging among Chinese-speaking consumers in Seattle. If finance attempts to analyze those expenses at the level of wireless, consumer, Washington, there is going to be a granularity-disconnect between finance and operations. Operations wants to know if the promotion generated positive profits, and finance can only answer that question when its data is at the level of unlimited instant messaging, Chinese, and Seattle. Finance can only answer that question when it tracks revenues, operational metrics, and costs at the same level of detail that operational managers use day to day.

Precision

The operational requirement for precision is related to the requirement for granularity. If granularity is about categories, precision is about measurements or metrics. Granularity is the filing system; precision is the number of unique observations filed under each category. Clearly, measuring on one metric — say, revenue — for each customer-product-location combination is not sufficient for operations managers; they need to see a wide variety of metrics on Chinese-speaking customers on unlimited-use instant messaging in Seattle (see Figure 1).

This information is relevant to operational managers because it affects their spending, and that makes these metrics the only operationally acceptable basis for assigning costs to customers, products, and locations.

For example, consider how the wireless company would analyze the costs of a customer support call center. It is really no wonder that operations managers reject finance’s assignment of customer support call center costs to products or customers based upon revenue or some algorithm. The costs are clearly related very closely to an observable transaction: the duration of each customer’s call to the support center. But, when finance uses TPA to assign call center costs to customer-product-locations based upon the duration of support center calls, finance becomes operationally relevant. At the same time, direct mail costs need to be assigned to customer-product-locations based upon the number of direct mail pieces sent to each. It is not a matter of finding the one magic metric that works for all processes. A wide variety of metrics must be gathered within each customer-product-location in order to provide the flexibility and precision required across all the major cost processes of the firm.

Timeliness

If finance gets all its information from the accounting systems, the accounting close cycle dictates finance’s decision cycle. This means finance can suggest course corrections monthly — and even then finance is informed by data 30 to 45 days old. That is ancient history for operations managers.

Operations managers need feedback quickly. The shelf life of decisions is measured in weeks rather than months. In the absence of actionable profitability signals, operations managers use intuition and proxy metrics. But in complex firms the interdependencies between departments create a web of effects that is impossible to manage intuitively.

To ensure that every operational decision is focused on maximizing profitability, finance must be able to demonstrate the effects of operating decisions in as little as three to five days. Consider the following example: Marketing and product management launch a new discount offer for an existing product. This offer is only available via the Web site, and only for customers who sign up for electronic billing. Is the offer working? To answer that question promptly, finance must use TPA.

With TPA, finance would be able to immediately isolate customers who responded. The cost of the direct mail campaign promoting the offer would be captured. There would be a noticeable absence of calls from these customers to the sales center, as they entered their own orders via the Web. Unfortunately, these customers make many mistakes in their orders, and TPA would immediately show that Web orders were increasing the work volume at the order-correction center. When installation technicians visit these customers, finance would immediately see that the duration of the visits was much longer than typical, perhaps because the technician was the customers’ first human interaction.

Since TPA shows the financial implications of each of these transactions, it allows a P&L to be constructed every day — without waiting for the accounting close. This allows finance to focus operational attention on processes where efficiency is not consistent with the initiative’s business case, and it may lead to withdrawing or repricing the offer if operations managers determine that efficiencies will not improve as the program matures.

It is important to note that these are not independent analyses; this is an integrated analysis of revenue and cost impact from customers responding to a particular operational offer. With all these effects together in a decision-specific P&L within a few days of the decision, finance brings new and improved information to the operational discussion — and becomes relevant.

Figure 1: Sample of Operationally Relevant Metrics

TPA: Evolution or Revolution?

TPA is the predictable next step in the evolution of financial information technology over the last 20 years. It began as CFOs realized that the general ledger simply did not provide actionable insight into the company’s profit machinery. The general ledger was designed to provide general auditability.

Activity-based costing (ABC) used periodic analyses to assess the costs of business processes that run across organizational boundaries. ABC improved the firm’s horizontal insight across processes, but it could not adapt quickly with the business and it was not very operationally granular.

The first attempt to bridge the granularity gap between financial and operational information came with customer profitability analysis (CPA). CPA used the unit cost factors from ABC, but it applied them at the customer level to provide a point-in-time rating of the profitability of each customer — or at least of each customer type — thereby improving the firm’s vertical insight. CPA was actionable by operations, but being dependent upon ABC, it suffered from some of the same difficulties: untimeliness of observations and difficulty of updating. In addition, CPA rarely provided detailed insight into “why” a customer garnered a certain profitability rating — or whether there was anything the firm could address to change the picture.

For reasons to be discussed in the next section, TPA exploits recent advances in technology, which allow TPA to simultaneously deliver the breadth of ABC, the depth of CPA, plus timeliness and responsiveness that neither of these predecessors delivered. By analyzing near-real-time operational data at the transaction level, TPA affords insight into the impact of each transaction on each process, customer, product, channel, and/or location at each point in time.

Not only is TPA the next step in financial analysis evolution, it is also the final logical step. When there is no finer operational data available, there is no lower level of attribution and causality to understand. Clearly, the analytical expertise to exploit the TPA approach will continue to evolve within firms using the approach, but the technology and data structures will have already fully matured. This affords CFOs a high degree of confidence that, while they will have to continue to make investments in analytical skills and practices to exploit the TPA approach, those investments will not be rendered obsolete by a need to move to a higher level of precision.

Figure 2: Silo Analysis and Reporting

How Does It Work?

Consider an insurance firm, in which four departments pull reports on different metrics relevant to their individual responsibilities (see Figure 2).

Where today each department within the firm pulls its operational information independently of the other departments, TPA pulls all of those data feeds together into a unified database.

For instance, TPA provides the sales department the ability to see premium production by sales agency by policy, the claims and call center contacts associated with each of those policies, as well as the integrated profits generated by each of those policies (see Figure 3).

Clearly, it would be possible to integrate all this information manually in a single analysis without using TPA. Such manual ad hoc analytical work is rampant within non-TPA firms. However, the effort required for a single analysis means that a very small portion of the firm’s operating footprint is investigated in a thorough and integrated fashion. By automating the data integration and costing procedures that traditionally require time-intensive special projects, TPA ensures that finance can analyze the entire breadth and depth of the company.

Notice that TPA pulls together four types of information for the insurance firm:

1. Operational metrics: These are data on the events being managed in each department — data that are pulled directly from the production operations support systems used by the operating departments.

2. Accounted costs: These are expenses that are already tracked by customer, product, and location within the existing accounting and control systems. A good example is commission expense, which is tracked to the customer order and agent level in the compensation system.

3. Measured costs: These are expenses that the accounting systems do not track by customer, product, and location, but which are directly a function of an operational metric that is observable at that level of precision. A good example is call center costs (discussed below).

4. Corporate allocated costs: These are expenses that sustain the business or some very large segment of it; they are not directly related to an operational metric or to any single customer, product, or geography.

One of the big payoffs from TPA comes in the area of transaction assignments, which relates to the last two types of information (see Figure 3). Because of the rich detail of operational data gathered, TPA has unprecedented capability to measure costs that traditionally were treated as corporate allocated costs. This allows TPA to minimize the portion of type 4 costs far better than traditional ABC.

A great example in the case of the insurance firm is the cost of operating customer service call centers. Traditionally, these costs would have been allocated to products as general corporate overhead, using revenue or a periodic time and motion study. Tracking the accounting cost of calls handled on each policy in the ledger would be prohibitively expensive. But, the operational support systems already track the most precious call center resource, agent talk time, by call and policy. TPA simply leverages the operational information to assign actual call center operations costs to each policy based upon the actual talk time incurred in support of that policy during that month. With TPA, profitability reports adapt immediately and automatically to actual operational results. Assertions regarding the call center costs driven by a product are supported by detail down to the level of every call. As a result, operations managers get timely profitability information that is synchronized with their recent operational experience.

Figure 3: Integrated Analysis and Reporting

Why Is TPA Surfacing Now?

CFOs who have been around the block more than once will recall at least one previous effort to employ something like TPA, an effort that invariably failed. So, it is reasonable to ask why firms are succeeding with TPA now. The answer is found in the underlying causes of past failures.

The first cause of failure in early experiments with TPA-like projects was immature technology.

These advances in technology have dramatically reduced the cost and difficulty of implementing TPA. Most companies can deploy a best-in-class capability in a few months, at a cost that is offset by systems/labor efficiencies in the current financial and operational reporting processes.

The second cause of early failures of TPA-like projects was that operational groups did not need the information. There has been a sea change in this area over the last decade, making TPA far more viable today, as there is operational demand for the information. As more firms have deployed sophisticated operations control systems and the associated business practices, their operations managers have developed the capability to operationalize TPA’s profitability information. At the same time, increasing levels of competitiveness in many industries are pushing those operations managers to find every available means of improving margins, an objective for which TPA’s information is invaluable.

Challenges in Adopting TPA

It’s no surprise that the major challenges posed by breakthrough information systems projects like TPA, are technology and people.

Technology

TPA can radically simplify the analytical process, but to do that, some very sophisticated technology is required. The first technical issue is whether to build the platform from scratch, integrate components from several vendors, or buy a turnkey suite.

Keep in mind that a TPA implementation requires software for: receiving, inventorying, and scrubbing data feeds from many operations support systems; integrating those feeds into a common cube record; assigning costs from the general ledger to appropriate transactions; storing the results; and providing user reporting and analysis on the results. One piece without the others will prevent you from achieving your objectives.

People

The typical finance department spends most of its time pulling data and preparing reports. TPA automates the process of pulling the data, standardizes the language and labeling schemes used, and allows virtually limitless flexibility in analytical reporting and mining. Accordingly, there is a significant change in roles and required capabilities within organizations that implement TPA.

Implications of TPA

Properly implemented, TPA will change your firm — and finance’s role within it. As competition increases the necessity of prospering in a tight-margin environment, TPA provides greater certainty of action than analyses based upon averages. As coordination between marketing and operations becomes the next step in wringing inefficiency out of the value chain, TPA’s 360-degree view of customer transactions provides the key to making coordinated, profit-focused moves. Expect to reap the following benefits from moving to TPA:

While the firm benefits from TPA, so does the finance organization. Daily updated profit information means finance gets a seat at the table for operating discussions. Human resource reallocation means finance gets a skills upgrade, allowing the organization to drive thought leadership in strategically important areas like strategic planning, pricing, and capital allocation.

Conclusion

Every customer transaction creates or destroys shareholder value. Increasing shareholder value requires that finance be able to understand the profit mechanics of transactions. By generating a P&L for each customer-product-location transaction, TPA sheds new light on operationally actionable profit improvement opportunities.

By integrating operational and financial information from the summarized level to the transactional level, TPA improves cycle time on management financial reporting, makes operationally relevant financial information ubiquitous throughout the company, and links operational financials to reported financials for integrated accountability.

TPA allows tighter coordination of operations policies and financial policies, which is essential to raising the influence of finance to correspond with the CFO’s heightened accountability.