4

University of San Francisco Economic Value of Data Research Paper

APPLYING ECONOMIC CONCEPTS TO BIG DATA TO DETERMINE THE FINANCIAL VALUE OF THE ORGANIZATION'S DATA AND ANALYTICS, AND UNDERSTANDING THE RAMIFICATIONS ON THE ORGANIZATION'S FINANCIAL STATEMENTS AND IT OPERATIONS AND BUSINESS STRATEGIES

Abstract

Companies are contemplating the organizational and business challenges of accounting for data as a "corporate asset." Data is now seen as a currency. This research paper deep dives into the economics of data and analytics and defines these analogies.

Mr. Bill Schmarzo,

Executive Fellow, University of San Francisco School of Management

Dr. Mouwafac Sidaoui,

Associate Professor and Chair Department of Business Analytics and Information Systems

The volume, variety, and velocity of data may have changed over the past few years, but one thing hasn't changed—the value of the data to improve operational decision-making and power business strategies. This research paper will explore the following questions with respect to how organizations maximize the economic and financial value of the organization's data and analytics:

• How does an organization identify and prioritize the business use cases upon which to focus its data and analytics initiatives?
• How does an organization determine the economic value of the data that supports the organization's business use cases?
• How does an organization create a framework that facilitates the capture and reuse of the organization's data and analytic assets?
• What is the role of the data lake, data governance, data quality, and other data management disciplines in managing, protecting and enhancing the organization's data and analytic assets?

Introduction

The importance of data has changed over the years. As the volume, variety, and velocity of the data grew over the past few years, the economic value of data has been transformed by the big data phenomenon^{[citation 1]} that has enabled organizations to capture a broader, more granular, and more real-time range of customer, product, operational, and market interactions. Today, business leaders see data as a monetization opportunity, and their organizations are embracing data and analytics as the intellectual capital of the modern organization.

More and more companies are also contemplating the organizational and business challenges of accounting for data as a "corporate asset."

Data as an asset exhibits unusual characteristics when compared to other balance sheet assets. Most assets depreciate with usage. However, data appreciates or gains more value with usage; that is, the more the organization uses the data across more use cases, the more valuable, complete, and accurate the data becomes.

However, there are severe limitations in valuing data in the traditional balance sheet framework. It is important that firms identify a way to account for their data. To address this challenge, this research paper will put forth the following:

1. A framework to facilitate the capture, refinement and sharing of the organization's data and analytic assets, and
2. A process to help organizations prioritize where to invest their precious data and analytic resources.

It is our hope that this research paper will foster new ways for organizations to rethink how they value their data and analytics from an economic and financial perspective. The concepts covered in this research paper will provide a common vocabulary and approach that enables business leaders to collaborate with the IT and Data Science teams on identifying and prioritizing the organization's investments in data and analytics; to create a common collaborative value creation platform.

Creating the Collaborative Value Creation Framework

Data and analytics are powerful assets in which to invest, but organizations struggle to assign these intangible assets their appropriate economic value. Assigning the appropriate value to these digital assets is important if organizations want to maximize their economic impact and optimize organizational investments in data and analytics.

Organizations need a framework—what we will call the collaborative value creation platform—that maximizes the economic value of data and analytic assets across the organization.

Step 1: Prioritizing Business Use Cases

To quantify the value of these intangible data and analytic assets, we need to find a basis point around which the organization can establish the prudent value of the data and analytics. We will use the organization's key business use cases (for example, acquiring more customers, reducing customer churn, improving the quality of care, improving customer satisfaction, reducing cybersecurity risks, reducing maintenance costs) and the financial value of these use cases to establish that prudent value.

Step 2: Role of Analytic Profiles

Even organizations advanced with substantial advanced analytic capabilities suffer from "orphaned analytics"^{[citation 2]}, analytics that address a one-time business need but are not "operationalized" or reused across multiple use cases. The capture, refinement, and reuse of the analytics can be addressed using a framework called an Analytic Profile^{[citation 3]}.

An Analytic Profile consists of metrics, predictive indicators, segments, scores, business rules, and analytic insights that provide a snapshot into the behaviors, preferences, propensities, inclinations, tendencies, interests, associations, and affiliations at the level of the individual entity such as customers, patients, students, athletes, jet engines, cars, and wind turbines.

Analytic Profiles help an organization to prioritize and align data science resources to create actionable insights that can be reused across the organization to optimize key business use cases, reduce cybersecurity risks, uncover new monetization opportunities, and provide a more compelling, more prescriptive customer and partner experience.

Step 3: Role of the Data Lake

A data lake is a data structure that holds large amounts of structured and unstructured data in its native format, that is, no schema is required to load data into the data lake. Unlike a data warehouse where the data is stored in a predefined relational structure, data in the data lake is stored as-is, in its native format. The ability to rapidly ingest, index, and catalog new data sources is critical in supporting the "fail fast / learn faster" data science efforts to identify variables and metrics that are better predictors of performance. As a result, the data lake becomes the organization's "collaborative value creation" platform by facilitating the capture, refinement and reuse of the organization's data and analytic assets across multiple business use cases.

Chipotle Use Case

We will now apply the framework and supporting process outlined in this paper to a real-world organization. We selected a public company so that we could use publicly available data to demonstrate the paper's concepts. The company that we selected was Chipotle Mexican Grill Inc.^{[citation 4]}

Step 1: Identify a Targeted Business Initiative

A Business Initiative is a cross-functional plan or program that is typically 9 to 12 months in duration, with well-defined financial or business metrics. One of the business initiatives highlighted in Chipotle's 2012 annual report was increasing same-store sales.

"Last year we opened 183 restaurants, grew our revenue by 20.3% to $2.73 billion, and saw comparable restaurant sales grow 7.1% for the year. Our restaurant-level margins were among the highest in the industry at 27.1%."

This goal of increasing same-store sales by 7% will be the business initiative upon which we will apply our data valuation framework and processes.

Step 2: Estimate Financial Value of the Business Initiative

First, we need to calculate the financial value of the targeted business initiative. The process of calculating the financial value of the targeted business initiative should be a straightforward financial accounting exercise. Table 4.1 provides an estimate of the financial value of Chipotle's "Increase same-store sales 7%" business initiative."

Targeted Business Initiative
Increase Same-Store Sales by 7%
Chipotle Sales ($000s)	$2,731,224
Number of Stores	1,410
Average Store Sales ($000s)	$1,937
7% Increase in Avg. Store Sales ($000s)	$2,073
Annual Impact ($000s)	$191,186

Table 4.1: Calculate Financial Value of Targeted Business Initiative

Step 3: Identify Supporting Business Use Cases

The next step is to identify the use cases that support the targeted "increase same-store sales" business initiative. We conduct interviews and envisioning exercises to identify the use cases across the different business stakeholders (that is, store operations, procurement, marketing, product development, finance). For Chipotle's "increase same-store sales" initiative, we identified the following use cases:

• Increase store traffic via local events marketing
• Increase store traffic via customer loyalty program
• Increase shopping bag revenue
• Increase corporate catering
• Increase non-corporate catering
• Improve promotional effectiveness
• Improve new product introductions

Step 4: Estimate Financial Value of Each Use Case

Next, we estimate the financial value of each uses case identified in Step 3. To estimate the financial value of each use case, each impacted business function creates a financial scenario for that use case; that is, each use case will have different financial scenarios tied to the number of business functions impacted by the targeted business initiative. For the use case "increase store traffic via local events marketing," we create three financial scenarios for the three impacted business functions of Field Marketing, Store Operations, and Product Development.

• Field Marketing (Scenario #1) created a scenario that estimates the incremental revenue generated from designing localized pamphlets and brochures for local events yielding a financial estimate of $62M.
• Store Operations (Scenario #2) created a scenario for co-branded, holiday events to be executed by local store management (Christmas event with the New York Times) yielding a financial estimate of $54M.

The business stakeholders would then collaborate to evaluate the different scenarios and select the most appropriate scenario (the scenario with a proper match of financial value and execution feasibility). The financial value of the selected scenario would then be used as the basis for the financial value of that use case. In the Chipotle "increase same-store sales via local events marketing" use case, we chose Scenario #1 (Field Marketing) with an estimated financial value of $62M.

After repeating the scenario creation, evaluation, and selection process for each use case, we end up with the following financial value for each use case (see Table 4.2).

Business Initiative: Increase Same-Store Sales by 7% (Estimated Value of Business Initiative: $191M annually)
Use Cases
Data Sources	Increase Store Traffic via Local Events Mktg	Increase Store Traffic via Loyalty Program	Increase Shopping Bag Revenue	Increase Corporate Catering Revenue	Increase Non-Corporate Catering Revenue	Improve New Product Intro Effectiveness	Improve Promotional Effective-ness
Financial Value ($M)	$62.0	$48.0	$26.0	$24.0	$14.0	$18.0	$27.0
Field Marketing ($M)	$62.0	$25.0	$26.0	$8.0	$10.0	$14.0	$18.0
Store Ops ($M)	$55.0	$42.0	$18.0	$24.0	$14.0	$9.0	$16.0
Product Dev ($M)	$45.0	$24.0	$12.0	$8.0	$8.0	$18.0	$22.0
Corp Marketing ($M)	$50.0	$48.0	$22.0	$10	$8.0	$12.0	$27.0
Procurement ($M)	$22.0	$0.0	$12.0	$0.0	$0.0	$8.0	$6.0

Table 4.2: Financial Value of Business Function Scenarios for Each Use Case

Step 5: Estimate the Value of the Supporting Data

The next step is to estimate the value of the supporting data sources. We have each business stakeholder rate the relative value of each data source with respect to each use case; that is, how important is data source #1 to use case #1, how important is data source #2 to use case #1, and so on. One can use a rating scale of 0 to 4, 0 to 10, or 0 to 100, but the finer the granularity of the data rankings, the more precise the data valuation determination.

Next, we calculate the value of each data source vis-à-vis each use case. One can make the formula as sophisticated as required, as long as the business stakeholders understand the rationale for the formula. Finally, the data source values are summed across the use cases to get an aggregated value calculation (see Table 4.3).

Key Business Initiative: Increase Same-Store Sales by 7% (Estimated value of Business Initiative: $191M Annually)
Use Cases
Data Sources	Increase Store Traffic via local events mktg	Increase Store Traffic via Loyalty Program	Increase Shop-ping bag Revenue	Increase Corporate Catering Revenue	Increase Non-Corporate Catering Revenue	Improve New Product Intro Effective-ness	Improve Promotion Effective-ness	Value of Data across all use cases
Financial Value ($M)	$62.0	$48.0	$26.1	$24.0	$14.0	$17.9	$27.2	$219.2
POS Transactions ($M)	$12.4	$16.0	$6.5	$4.8	$2.8	$4.9	$6.8	$54.2
Market Baskets ($M)	$12.4	$16.0	$8.7	$4.8	$2.8	$4.9	$6.8	$56.4
Local Demographics ($M)	$9.3	$4.0	$4.3	$9.6	$5.6	$3.3	$3.4	$39.5
Traffic	$6.2	$4.0	$2.2	$2.4	$1.4	$1.6	$1.7	$19.5
Weather ($M)	$9.3	$4.0	$2.2	$2.4	$1.4	$1.6	$1.7	$22.6
Local Events ($M)	$12.4	$4.0	$2.2	$0.0	$0.0	$1.6	$6.8	$27

Table 4.3: Aggregated Financial Value of Each Data Source across All Use Cases

Step 5: Identify and Capture Analytics

The final step is to use the Analytic Profiles to identify and capture the analytics that support each use case. While the analytic results can take many forms (for example, segments, clusters, business rules, predictive indicators), we will use the analytic concept of scores as a practical way to create analytic insights.

Scores are a rating system that aids in comparisons, performance tracking, and decision-making. Scores are used to predict the likelihood of certain actions or outcomes. For example, the FICO score measures the likelihood of a borrower repaying their loan^{[citation 2]}. Scores are actionable, analytics-based measures that support the key decisions your organization is trying to make.

Identifying and capturing the analytics is a 3-step process:

1. We first list the decisions needed to support the targeted use case ("Increase store traffic via local events marketing").
2. Next, we identify or brainstorm the recommendations that need to be developed to support the decisions. A recommendation is a suggestion or proposal, developed using prescriptive analytics, as to the best course of action.
3. Finally, we identify the scores and the metrics that comprise the score, that support the recommendations and decisions.

For the Chipotle "increase store traffic via local marketing events" use case, we identified three potential scores:

1. Local Economic Potential Score: which measures the economic potential of the area around the store.
2. Local Vitality Score: which measures the amount of activity or "life" around the store.
3. Local Sourcing Potential: which measures the feasibility of getting the necessary organic food items to support the local event marketing.

Table 4.4 brings the Decisions, Recommendations, and Scores process together.

Use Case: Increase Same-Store Sales via Local Event Marketing

Store Staffing	How many people to staff during the local event? What skills to staff for the store as well as promotions? When to staff based upon the local event? How to measure the temporary and permanent staff performance?	Local Economic Potential Score Local demographics Local economic variables Local home values Local unemployment rate Number of university students Local Vitality Score Miles from schools Miles from malls Miles from local sports venues Local sporting events Local entertainment events Other local events (farmers market) Local Sourcing Potential Number of local suppliers Miles from stores Supplier production capacity Supplier quality Supplier reliability Delivery feasibility
Store Inventory	How much food inventory to order for the local event? What would be the minimum quantity of each food item to reorder? What would be the storage plan for surplus inventory?
Local Events Promotions	Which local events to sponsor? How much marketing funds to allocate to the local event? What types of promotions? Special menu pricing for local events?
Supply Chain	Where to acquire additional inventory in case of overrun? Who would be the vendors for reorder procurement? How to set the quality control of the new vendors?

Table 4.4: Mapping Analytic Scores to Recommendations to Decisions

The end result is an Analytic Profile for each Chipotle store that captures the analytic results across all the use cases (see Table 4.5).

Chipotle Store 00134	NCE	Beta	Trend
Local Economic Potential Score 2.1	92	1.85
Local Vitality Score 1.4	67	3.25
Demographic Segments 3.2	82	2.25
Behavioral Segments 3.1	65	1.90
Store Traffic Score 1.0	92	2.89
Store Remodel Score 1.0	55	2.75
Store Loyalty Index 2.0	98	1.35
Store Customer Satisfaction	88	1.74
Vendor Reliability Score	99	1.10
Store Employee Satisfaction Score	78	2.65

Table 4.5: Chipotle Store Analytic Profile

In Table 4.5:

• NCE stands for Norma Curve Equivalent and is a way of standardizing scores into a 0 – 100 scale similar to a percental-rank while preserving the value equal-interval properties of a z-score
• Beta is a measure of the volatility or rapid change in the NCE score

For example, the NCE for the Store Traffic Score for Chipotle Store 00134 is very high (92 out of 100). However, the Beta or volatility of 2.89 means that the Store Traffic Score changes frequently (maybe due to being near a high school where students are only in classes during school time, or near a sporting venue which has fewer sporting events but draws large attendance for those sporting events). In this case, it would be very important for Chipotle Store 00134 to factor in the high school's class schedule or the sporting venues event schedule (and estimated event attendance) into its operational plans.

The analytic results captured in the Analytic Profile are now ready to be refined and shared across other use cases, increasing the economic value of the analytics results and addressing the "orphaned analytics" issue.

Summary

As organizations seek to leverage data and analytics to power their business models and improve operational and strategic decision-making, organizations need to manage and account for data and analytics as corporate assets. Data and analytics will become the primary economic driver in many organizations that seek to optimize key business processes, reduce security and compliance risks, uncover new monetization opportunities, and create a more compelling user experience.

Citations

1. 12 Big Data Definitions: What's Yours: https://www.forbes.com/sites/gilpress/2014/09/03/12-big-data-definitions-whats-yours/#45f849b413ae
2. How to Avoid "Orphaned Analytics: https://www.linkedin.com/pulse/how-avoid-orphaned-analytics-bill-schmarzo?trkInfo=VSRPsearchId%3A790269301484873521738%2CVSRPtargetId%3A6165930841810677760%2CVSRPcmpt%3Aprimary&trk=vsrp_influencer_content_res_name
3. Best Practices for Analytics Profiles: https://infocus.delltechnologies.com/william_schmarzo/best-practices-for-analytics-profiles/
4. Chipotle Mexican Grill: https://en.wikipedia.org/wiki/Chipotle_Mexican_Grill

5. Chipotle Mexican Grill 2012 Annual Report: https://ir.chipotle.com/annual-reports