Evaluating AI contribution to value creation and capture: a tentative framework
There are different ways in which AI adoption can positively add to a firm´s strategy but still significant ambiguity in understanding and evaluating their mutual relationships.
How do AI creates value?
To answer this question it´s useful to start from the definition of value capture as “the benefits the user gets from a firm´s offering (product and/or service) expressed in terms of willingness to pay (WTP) minus the cost of resources incurred by the firm to deliver that offering” (INSEAD).
AI contributes to this equation in a two types of ways, as illustrated below.
This very simple framework can help frame an AI implementation in terms of either a) increased willingness to pay or/and b)cost reduction.
Increased Willingness to Pay (WTP) is connected among other things to the provision of a better experience to a customer/user/employee. This in turn is achieved through two basic macro-tactics: either improving the quality of an existing offering (a) or create an entirely new offering (b). A media company we worked with managed to achieve a substantial improvement in terms of the quality of their content - which was expressed by an uptick in online views. The company didn´t create any new type of offering, but captured a more readership by ameliorating what they were already serving to their readership. In contrast, a pharma company who uses AI to discover new drugs is an example of novel offering creation.
It´s worth noting that in both these examples, the AI implementations are also linked to the creation of efficiencies that allow to reduce the costs of existing processes - or the potential cost of new ones. So there is a double effect on value creation, as AI simultaneously contributes to WTP and to cost decrease.
Limitations with measurement of AI impact
While the contribution of AI to cost decrease is more easily quantifiable (fx, number of hours saved, FTEs´ cost x hour, estimated return from higher-value activities performed thanks to time savings), calculating the impact on AI on WTP remains arduous, due to both the inherently qualitative nature of this dimension and the uncertainty of the benefits from AI implementations.
If we take for example the case of novel offering creation through AI, the adoption of the technology is just one factor in a complex fabric of value interdependencies - external partnerships, patenting, forming alliances, and using entrepreneurial orientation and dynamic capabilities to quickly shift the necessary resources in order to seize opportunities (Dyduch et alii, 2023). Therefore, it can often difficult to isolate the singular dollar value contribution of AI to a firm´s value creation when this value is represented by a novel product or offering within an existing business model, like in the aforementioned examples of the media company.
A different scenario would be that of companies whose business model is sort of built around the AI. In the case of Synthesia (see below) value creation can ber roughly calculated by netting out costs from subscription revenues.
The business model illustrated here constitutes a good example of value-creation and value-capture dimensions alignment, which previous research identifies as a pre-condition for commercially viable AI business models (Åstrom, et alii).
However, the same studies suggest, this kind of scenario tend to apply to AI providers. AI´s impact on value creation for AI providers´own customers is quite elusive to unpack and further research is needed on this direction.
Types of Value Creation enabled by AI
The following is a non exhaustive mapping of opportunities for value creation within the enterprise context. Each type matches different AI capabilities and differs as to what extent it allows to be quantified.
For Generative system the following capabilities are considered:
1)summarisation 2)content generation (text or other media) 3)question&answer (conversational) 4)extraction(of information)
Whilst for non general machine learning,
6)input classification 7)forecasting 8)recommendations.
Type: AI-powered assistant (for external users) AI capability:Q&A value creation: improved CX and satisfaction through improvement of existing services (for example when the conversational experience serves as a better communication or customer experience channel) or, novel offering, when the conversational experience unlocks new ways for accessing and creating know-how (for example a chat with data room kind of a experience) that were not previously in place.measurement: qualitative for the first case, though can be inferred through quant metrics like resolution time, call lengths and similar);only partially to quantify in the second case, through measuring time of information retrieval.
Type: Classificator + recommendations (ex, CV rating system for HR) AI capabilities: Classification, Recommendation, possible summarisation Value creation: improved processes and efficiencies (leading to lessen costs), but also improvement on the quality of the outputs (in the HR example, a better way to classify and predict CV could lead to more qualified hires or increased talent retention).
Type: code/text/other media generators AIcapabilities : content generation Value creation:novel offering (creative concepts ot new drugs discovery) but also improved efficiency ( like mail generation, visuals marketing contents).
[This non-exhaustive listis continuously updating]
Still overlooked: value capture through AI
The essence of conducting any organisation's activities is the implementation of strategic actions that will enable value creation and value capture (VCVC), resulting in the effective achievement of set objectives.
While they can help create substantial value, AI technology capabilities are not enough—companies need to understand how the technology can be commercialized through appropriate AI business model innovation. (Astrom et alii).
The ways AI can contribute to value capture seem to have received less attention in academic literature.
Some important areas are:
Dynamic Pricing Models: AI systems analyze market conditions and consumer behavior to optimize pricing strategies, enhancing revenue capture through real-time adjustments.
Predictive Analytics: Businesses utilize AI to forecast customer needs and behaviors, allowing for tailored offerings that improve customer satisfaction and retention, ultimately leading to increased profits.
Performance-Based Pricing: AI enables companies to link pricing directly to the performance of their services, ensuring that customers pay based on value received, which can enhance perceived fairness and loyalty.
Outcome-Based Contracts: AI facilitates the creation of contracts that tie payment to specific outcomes or results, allowing for more flexible and customer-aligned pricing structures.
AI-Driven Market Insights: Companies leverage AI to gain insights into market trends and consumer preferences, helping them refine their offerings and capture more value from their products and services.
An interesting topic of research into these categories may shed light on how value capture strategies through AI complement and reinforce valie creation in the AI driven enterprise.