Identifying data-driven use cases with a value driver tree

A known best practice is to align your data strategy with the overall business strategy by focusing on use cases. By working with various clients across sectors, we’ve learned that one of the most powerful ways to do so is through a value driver tree.

What is a value driver tree?

A value driver tree is a framework for identifying and prioritizing the key drivers of value within an organization. It consists of different branches representing the various components that contribute to value creation. These branches can include revenues, costs and other key performance indicators that are specific to the sector and organization itself.

Figure 1: Sample driver tree (source: Supercharged Finance).

The revenue branch can be subdivided into both the number of customers and the average revenues per customer. By analyzing these subcategories, an organization can identify specific use cases that drive revenue growth such as increasing customer acquisition or improving the average transaction value. Take for example precision marketing, where companies focus investments in their target customers, routinely results in higher return on marketing investments.

Similarly, the cost branch can be subdivided into fixed costs such as overhead, fixed marketing costs, property and equipment cost. This can include variable costs such as the cost to produce products. By mapping use cases to these subcategories, an organization can identify opportunities to reduce costs, improving production efficiency or streamlining marketing efforts.

Standard sector driver trees

Driver trees can vary dramatically across sectors. Key revenue drivers for retail companies may include store-level and product basket insights meanwhile a telco company instead may focus on network optimization and product bundling.

Yet it turns out that driver trees are often consistent within sectors. Of course, each company is unique, but the different ways companies can increase revenues, cut costs and drive value creation are practically identical.

Consider the below standard driver tree. On the left side, you can see common drivers related to revenues, costs and asset efficiency. On the right, there are typical divisions or departments that many companies have (although they might be labelled something differently).

Figure 2: Standard value driver tree.

Assessing impact and prioritizing use cases

With a general driver tree in place, the corporate divisions are a good place to begin exploring data-driven use cases. Take marketing for example:

Figure 3: Marketing impact in a driver tree

A big advantage of such an approach is the actionability. Corporate divisions and core underlying processes typically have clear ownership. It is therefore straightforward to identify key stakeholders and validate the relevant use cases.

Let’s assume that for a given CPG company the following use cases have been identified:

• Personalization: Using data on customer behavior and preferences to personalize marketing messages and offers.
• Targeting: Analyzing customer data to identify specific segments that are most likely to respond to particular marketing messages or campaigns.
• Customer Lifetime Value (CLV): Using data on customer behavior and purchasing history to predict the lifetime value of a customer, and then using that information to optimize marketing strategies.
• Cross-selling and upselling: Analyzing customer data to identify opportunities for cross-selling and upselling, and then delivering targeted messaging and offers to encourage additional purchases.
• A/B testing: Testing different marketing messages, offers, and creative using data-driven insights to determine which approach is most effective.
• Precision marketing: Using data to target individual customers or small groups of customers with highly personalized and targeted messaging and offers.
• Marketing mix modeling: Analyzing the impact of different marketing channels and tactics on customer behavior and sales, and using that information to optimize marketing strategies.
• Predictive analytics: Using data to make predictions about future customer behavior and preferences, and then using those insights to optimize marketing strategies and campaigns.
• Social media monitoring: Using data from social media platforms to monitor brand sentiment and engagement, then using those insights to inform marketing strategies and tactics.

With this list in hand, a step further can be taken to verify what possible top-line and bottom-line impact could be. In CPG, for example, boosting cross-selling and up-selling capabilities routinely leads to a 1 to 3% uptick in revenues (and more than 5% in unique cases with low existing maturity).

Diving deeper, the existing maturity can be assessed as well as what data types (or sources) are (or should be) used. All of this leads to an overview below:

Figure 4: Organizational assessment of data-driven marketing use cases

Picking use cases that (can) have a demonstrable substantial impact but low current maturity provide an excellent starting point for data modernization or transformation. Common data sources can be reviewed from a data product enablement perspective. In the example above, Data Source 2 is required for eight of the nine use cases and three of the four use cases of low maturity. Conceivably, there must be a way to verify why maturity is low and how these use cases can be better enabled through uplifting a single data product.

Telling the story

As you continue to work through your use cases and document them and their impact, you’ll have an opportunity to update your driver tree as seen below:

Figure 5: Impact of data-driven use cases on total enterprise value

You can use a view like this to quantify the consolidated value that’s created through specific use cases. More importantly, it allows you to tell the story of the data organization and explain the value it adds. This is critically important given the struggles data leaders like Chief Data Officers have faced in the past, trying to explain the return on invest in their respective organizations and explaining their historically short tenures of less than two and a half years on average.

Conclusion

Using a driver tree to identify, prioritize and enable data-driven use cases can transform the way organizations think about and invest in data innovation. It also helps by informing business case development and tracking value creation post-implementation. We’ve seen evidence of revamped data modernization journeys across sectors like CPG, retail, hospitality, transportation, insurance and banking. No matter what industry you are in — if you are a data leader — a value driver tree might be the right tool for you.

References:

• How to Define Value Drivers, Supercharged Finance.
• A complete guide to data-driven customer acquisition, Ivy Liu.
• Evaluation of precision marketing effectiveness of community e-commerce–An AISAS based model, Jun, Li, Yanzhou, Gonzalez, Weiyi, Litao, and Zhang.
• Why Do Chief Data Officers Have Such Short Tenures?, Harvard Business Review.
• What’s the big deal about Data Products?, Medium.
• Data Products as a Lever to Quantify Data Value, Medium.