Building a data and AI strategy that drives real business value

A three-step approach to building your data and AI strategy

by Andreas Odenkirchen, Dr. Matthias Schlemmer, Dr. Maik Hesse, Agnes Heringer and Justus Rödel

When it comes to turning data into value, you probably already know your company needs to connect the dots along the data value chain – from data collection and data management to Analytics and AI, and finally data monetization. We discussed the importance of this concept in our last article, and this imperative is becoming increasingly clear across sectors. But the next logical question, of course, is how?

How can your company build a data and AI strategy that drives real business value?

To start, you might consider our three-step strategic approach. Based on our vast experience working with industry-leading organizations on their own data-driven transformations, here is what we’ve seen work. First, clearly define your aspiration (A) for data and AI. Second, underpin it with a qualified and integrated portfolio of business-oriented use cases (B). And finally, align your company’s data & AI capabilities (C) to effectively support your aspiration and use cases.

By following this approach, your company will increase its opportunities to become a truly data-driven enterprise, which in turn will yield real business benefits, i.e. growth, efficiency compliance and de-risking. Let’s take a more detailed look at this approach in practice.

How to get it right

A) Determine your aspiration for data and AI.

It all starts with your business strategy: Why does your company exist today and why will it exist in 5 years? What will you be offering, in which markets, and to which clients? How do you expect your revenues and costs to develop? And how will external factors like regulatory requirements, emerging technologies, and ethical principles influence your business?

Your data and AI strategy needs to clearly support your business strategy, while also anticipating a degree of agility as things change and evolve. You need to be able to explain the contribution of data and AI to your company’s vision, strategic objectives, and financial targets. Otherwise, you are unlikely to receive the level of investments you will need to create a significant business impact with data and AI.

So what could be your company’s aspiration for data and AI? We propose exploring different scenarios and ways-to-play – with some being more defensive (i.e., focused on managing risks) and others being more offensive (i.e., focusing on data-driven revenues or cost reduction). Then evaluate your options based on how well they support your business strategy, fit to your corporate culture, and leverage your strength and competitive differentiators. You might start with a generic scenario, and gradually define the nuances that will ultimately lead to your specific way-to-play along the offense-defense spectrum. Here are some examples of aspirations that you might consider starting with:

1. We want to be a trusted data processor: By ensuring trust in handling our clients’ and business partners’ data, we will strive to secure or strengthen our market position.
2. We want to be a data-driven business optimizer: We will strive to increase the effectiveness and efficiency of our business operations using data and AI, thus driving growth and profitability.
3. We want to be a data ecosystem builder: Leveraging our business network, we can provide, orchestrate, and participate in a platform for new data-driven business models.
4. We want to be a data monetizer: We can assemble and sell data products (i.e., services and solutions) that have a high value for our business ecosystem and will generate new revenue streams.

If you’re uncertain which aspiration to choose, select two or three at a time as working hypotheses to kick off the discussion. This will help you better understand and flesh out your way-to-play and to assess required investment and change activities.

B) Identify and qualify data and AI use cases.

Once you have defined your specific way-to-play (hypotheses), you need to back them up with a set of high-value data and AI use cases. These use cases can be identified anywhere along your company’s value chain and from your product and services portfolio. You can also seek inspiration from startups, research institutes, or solution vendors, as long as you focus on the business impact of the solutions. 

Based on our experience, the challenge is usually not identifying data and AI use cases. Instead, it is rather scoping them right, determining their business value, and evaluating their feasibility in face of existing capabilities. Therefore, you need to involve the relevant business, market, and technical experts, as well as finance experts to define and validate the assumptions for your business case calculations. You should also take into account effects (such as potential synergies) from combining use cases that can jointly deliver higher value, have similar requirements, or even build on each other. For example, if you implement a predictive maintenance use case for a production line, you typically need to implement a condition monitoring use case first. And you may then also want to implement a predictive quality use case that leverages similar production line data but taps additional business value potential.

C) Derive and align required data and AI capabilities.

Your prioritized use cases may have different requirements for data and AI capabilities, such as specific data modeling skills, tools, or algorithms. Even so, you should try to build a coherent set of capabilities that create synergies and can be operated efficiently, so to best leverage your investments into data and AI. In other words, you need to think about data and AI capabilities holistically. We suggest using our framework as a reference to help you structure your discussions around required capabilities. This framework is composed of the six key building blocks and capabilities that you need to realize data and AI use cases. Establishing any of these capabilities requires investments in the right set of technology, people, processes, and supporting data culture.

To ensure that you’ve defined the right capabilities, we recommend validating your capabilities against the defined use cases after each design iteration. This will ensure that your capabilities match your use cases’ requirements and allow your strategy to yield its full value potential.

Conclusion

Having properly defined your strategic aspiration, underpinned by a set of qualified high-value use cases, and the corresponding data and AI capabilities, you have strategically aligned your data value chain. And you have created the foundation for a solid investment case: Bringing together the expected value of the use cases with the implementation and operating cost of the target data and AI capabilities, you have a full perspective on your investments and expected returns. This, in turn, should help you facilitate discussions with the leadership team to get their sign off, and it should also help you stay on track during the implementation phase.

Stay tuned for more articles on data and AI strategy. In the meantime, we’d love to hear from you! Please share your own perspective in the comments section, or feel free to contact us directly at andreas.odenkirchen@pwc.com.