Determining the Economic Value of Data
This post introduces a data economic valuation process that uses an organization’s key business initiatives as this basis for establishing prudent value.
Data Economic Valuation Example
Let’s walk through an example to highlight how this process works. We’re going to start by using the publicly available information for a bank we will call ACME Bank. From ACME Bank’s annual report, we can determine that the bank is trying to “increase the number of products per household.” Their 2010 annual report states the following:
“This year, we crossed a major cross-sell threshold. Our banking households in the western U.S. now have an average of 6.14 products with us. For our retail households in the east, it’s 5.11 products and growing. Across all 39 of our Community Banking states and the District of Columbia, we now average 5.70 products per banking household (5.47 a year ago).”
So based upon the above information, ACME Bank wants to grow the number of products held per household from 5.70 to 6.20. So let’s make that our targeted business initiative:
Increase number of products per household from 5.70 to 6.20 over next 12 months
Step 1: Determine Financial Value of Targeted Business Initiative
So what is the potential range of value to the bank in increasing the average number of products held per household from 5.70 to 6.20? The bank’s annual report didn’t spell out the value of the initiative, so we’re going to perform some rough calculations based upon data that is available in the annual report.
Doing some rough calculations using numbers that we were able to glean out of the annual report, we estimate that each product held per household is worth $31.33 annually (see table below).
So if we could increase the number of products held per household from 5.70 to 6.20, it would be roughly worth $1.1B to the bank per year (see table below).
That seems like an incredible number, so let’s cut it by 90% just to be conservative. That puts the value of this initiative at $110M.
Step 2: Identify the Decisions That Drive the Targeted Business Initiative
The next step in the process is to identify the high-level decisions that need to be made to drive the targeted business initiative. Below are some of the decisions that ACME Bank would need to make to support the “Increase number of products per household” business initiative:
- Improve cross-sell profiling
- Improve customer segmentation
- Improve targeting prioritizing
- Improve offer effectiveness
- Improve re-targeting effectiveness
- Improve close effectiveness
- Reduce time-to-close
- Improve customer satisfaction
Step 3: Quantify the Value of Individual Decisions
Next we need to assign a rough order of financial value to each of the decisions that support the targeted business initiative. We can conduct a brainstorming session with the key business stakeholders (i.e., those business users who either impact or are impacted by the targeted business initiative) to assign a rough order of value to each decision in light of the overall targeted business initiative.
We could then discuss the results and allow the different stakeholders to state their case for the rough order of value. Then everyone could vote one more time. The result of the brainstorming process would then end up looking like Table 1.
Some notes about Table 1:
- ACME Bank’s business initiative of “increase number of products held by household from 5.70 to 6.20” equates to a 9% increase in the number of the bank’s products held by household over the next 12 months. So I have arbitrarily targeted a 10% increase/improvement for each of the decisions to create a baseline for the conversation. 10% may be too aggressive for a 9 to 12 month timeframe and you may want to amp that down to something around 3% to 5%.
- A facilitated brainstorming session using techniques such as weights, ranks and votes with the key business stakeholders can yield the best case, worst case and most likely scenario numbers.
- The aggregation of the dollar-valuations for each decision will not sum to the total value of the targeted business initiative. There is just too much correlation and interplay between the decisions for that to happen.
I have to admit that the process of assigning a “rough order of value” to each decision is not a hardened process. However, the process does have the benefit of forcing the conversation between IT and the Business
Step 4: Assess Value of Data Sources to Each Decision
Next, we need to determine a rough order of value for each data source with respect to how important each data source is to supporting the respective decisions. We will use Harvey Balls (with a value from 0 to 4) to value each of the data sources. I like using the Harvey Balls as it allows even the causal user to visually ascertain which data sources are likely the most important. If one wanted more precision, then using a scale from 0 to 10, or 0 to 100, might be more advantageous. However for this exercise, we’ll just stick with the Harvey Balls (see Figure 2).
Key stakeholder interviews (to get the initial value approximations) and then a facilitated workshop driving collaboration across key business stakeholders can yield the Harvey Ball rankings (0 to 4) that appear in Figure 2.
Next, we can create a formula that calculates relative economic value of each data source vis-à-vis the decisions. One can make the formula as sophisticated as you want, as long as the business stakeholders can clearly understand the rationale for the formula. Below is the formula that I used:
It is a very simple formula. But if explaining the formula loses the interest of the business leaders, then they will have little confidence in the results of this exercise. Consequently, err on the side of keeping the formula simple versus making it overly complicated.
Step 5: Aggregate Economic Value for Each Data Source
Finally, the economic value of each data source can then be summed across the decisions to get a rough order assessment on just how valuable each data source could be (see Figure 3).
Organizations have an opportunity to use data to improve their decision-making. While that’s something that most companies have been doing with data for the past couple of decades, there is the opportunity to take decision-making to the next level of granularity and actionability. Access to more-timely, more complete and more accurate data can enable organizations to tease out more significant, material and actionable insights about their customers, products and operations in order to make “better” decisions.
The advantage of this data economic valuation process includes:
- By starting with a key business initiative, you have established the financial basis for “prudent value” that we can use as the basis for ascertaining the economic value of the supporting data sources
- You are forced through a process of identifying the different decisions necessary to support the targeted business initiative, and to associate a rough order magnitude of value to improving the effectiveness or outcomes from those decisions
- Forces the business users to contemplate and rank the perceived value of each data source vis-à-vis the decision that they are trying to optimize
- Finally, the valuation formula puts you in a position to attach reasonable financial value to the different data sources that can ultimately prioritize data acquisition, cleansing, transformation and enrichment activities
Ideally, one would want to take this exercise to the next level and add a process for determining the cost of acquiring each of the data sources. The cost would need to consider not only the cost to acquire the data, but also the cost to clean it up, align it, transform it and enrich it. Maybe that’s a topic for my Big Data MBA class to explore.
Not all data is created equal, that is, some data is more important than other data in supporting the decisions that support the organization’s key business initiative. Consequently it’s important to have a rough estimate as to what data is most important in order to guide your “data as an asset” management strategy.
Original. Reposted with permission.
- Internet of Things: “Connected” Does Not Equal “Smart”
- The “Thinking” Part of “Thinking Like A Data Scientist”
- Simplifying the Internet of Things Conversation