It’s probably fair to say that most observers expect artificial intelligence to be embedded into most legacy products and processes over time, with an expectation that successful implementations will produce outcomes such as lower costs; higher profit margins; faster delivery times or higher customer satisfaction.
But it might also happen that AI is applied in ways that create new opportunities--such as business models that were not possible or sustainable before--as well. For some of us, that is the more-interesting opportunity.
Consider the way the shift to digital product creation and delivery possible with the internet created the opportunity for big, industry-leading technology firms supported by advertising revenue models. That was pretty much unthinkable prior to the era of digital media and internet distribution.
So many AI-enabled possibilities might exist in the software business, for example, where previously expensive solutions are cost-reduced or capability-enhanced by AI, making them available to new customers.
Where only enterprises might have been able to afford such tools, AI might enable small business and medium-sized business access to tools including:
AI-powered software that analyzes customer data and automates personalized marketing campaigns with targeted messaging and budget optimization.
AI-powered financial planning software that analyzes user spending habits and recommends personalized budgeting and investment strategies.
AI-powered tutoring software that personalizes learning plans based on student strengths and weaknesses, offering adaptive instruction at a mid-range price point between basic learning apps and expensive human tutoring.
AI-powered software that analyzes medical records and patient data to assist healthcare professionals in diagnosis and treatment planning, potentially making such tools more accessible to smaller clinics and individual practitioners.
Though it might be harder to do so with physical products, the same principle should apply: AI might reduce some barrier to creating products that are more specialized, targeted or positioned someplace between the high-volume, low-cost segment and the low-volume, high-cost market segments in any market.
Prior to the internet era, at least some entrepreneurs had explored creating ad-supported telecom or content services, for example.
That was not a huge leap of imagination, given the ad-supported revenue models of broadcast radio and television; smaller newspapers and magazines. So some explored ad-supported phone calls, for example, none successfully.
Only with the shift to internet-delivered messaging and voice features is it feasible to offer free or ad-supported communications, sometimes with an incremental usage-based fee (for international calls to public switched telephone network devices, for example).
So for some, the interesting possibility is that some previously-unworkable business models might be possible if AI solves some key business problem: value, price, cost, distribution, support or something else crucial to the business model.
And that might happen most often for a broad range of products in any industry that are neither the volume nor the value leader; occupying neither the lower-cost, volume position nor the high-cost, lower-volume, premium position.
It might be conventional wisdom that high-volume, low-price products as well as low-volume, high-price products are “easiest” to create business models around, whereas many products somewhere in the “middle” are more difficult, as selling prices do not provide enough margin for robust customer and product support.
Generally speaking, think of consumer or business subscription products costing $20 to $30 a month, and sold directly using the internet, or high-price enterprise products that are expensive, but also provide enough gross revenue and profit to be sold using a direct sales force.
Then think of all sorts of products that a potential customer has to think about--it is not an impulse buy--and do not represent high sales volumes, and also are priced at levels that require some form of indirect sales channels (channel partners, for example).
The price-value relationship might be part of the problem. Low-cost products can leverage economies of scale to achieve high sales volume and profitability even with low margins. Think of discount retailers or subscription services with minimal features.
That, in turn, can create profit margin issues. Mid-range products can get squeezed on profit margins, having neither the high markups of luxury goods nor the economies of scale of low-cost products.
Luxury or premium products often can command higher prices due to brand recognition, unique features, or exclusive materials and often can be sold to a niche market willing to pay more.
Mid-priced products often lack the compelling features of high-end options or the affordability of low-end ones. They can struggle to attract enough customers at a price point that allows for substantial profit margins.
Customers seeking the best value might be drawn to lower-priced options, while those prioritizing premium features might be willing to pay more.
Study Title/Source | Key Findings | Focus on the Value Gap |
"Blue Ocean Strategy" by W. Chan Kim & Renée Mauborgne | Emphasizes the importance of creating a new value proposition that avoids direct competition. Stuck-in-the-middle products often compete head-on with established players. | Argues that successful businesses either create low-cost, high-volume products or high-value, premium offerings. |
"Winning in the Middle: Avoiding the Commodity Trap" by Michael Porter (Article explores strategies for success in mid-tier markets) | Acknowledges the challenges of mid-tier markets but suggests strategies like differentiation, operational excellence, or niche targeting to succeed. | Offers a more nuanced view, suggesting that success in the middle is possible with careful strategic planning. |
"Pricing Strategy: Setting Price to Maximize Profits" by Nirmalya Kumar (Book explores various pricing strategies) | Discusses the importance of understanding customer value perceptions when setting prices. Mid-priced products risk failing to deliver a value proposition strong enough to justify their price. | Emphasizes that price should be aligned with the perceived value customers receive, which can be challenging for mid-tier products. |
"Value Proposition Design" by Alexander Osterwalder, Yves Pigneur, & Greg Bernarda | Highlights the importance of a clear value proposition that resonates with the target customer segment. | Implies that mid-priced products might struggle to offer a compelling value proposition that stands out from both high-end and low-end options. |
Channel strategy also comes into play. Generally speaking, high-volume, low-price items can be sold directly using the internet or mass market retail. The low-volume, high-price goods are sold using direct sales. The medium-price, medium-volume products use distributors, resellers or agents.
The issue is whether artificial intelligence can change key costs of manufacturing, distribution, support, sales and marketing enough that formerly-difficult models become more feasible. That will matter for the fortunes of all sorts of legacy or startup firms, allowing marginal business models to become more robust.
Perhaps some “one-time sale” products can be changed into subscriptions, increasing lifetime customer value. Perhaps channel sales can be converted into direct online sales, boosting profit margins.
Maybe high after-sale support costs can be sliced. The need for channel partners might be reduced or eliminated by AI that simplifies product manufacturing cost, improves “zero touch” performance so support costs are reduced or allows online direct sales.
Previously Difficult Functions | How AI Could Lower Costs | Use Case |
Personalized Mass Customization | AI can analyze customer data and preferences to design and manufacture products tailored to individual needs, without sacrificing economies of scale. | A clothing company uses AI to personalize fabric choices, cuts, and styles based on customer body scans and preferences. |
Predictive Maintenance in Manufacturing | AI can analyze sensor data from machines to predict potential failures, allowing for preventive maintenance and avoiding costly downtime. | A factory uses AI to monitor equipment vibrations and predict bearing wear, enabling proactive maintenance before breakdowns occur. |
Automated Customer Support | AI-powered chatbots can handle routine customer inquiries, freeing up human agents for complex issues and reducing support costs. | A bank uses AI chatbots to answer basic banking questions 24/7, reducing the need for human customer service representatives for simple inquiries. |
Intelligent Content Delivery Networks (CDNs) | AI can optimize content delivery based on user location, network conditions, and content type, minimizing bandwidth usage and delivery costs. | A streaming service uses AI to personalize content delivery based on viewer location and device capabilities, reducing bandwidth consumption. |
AI-powered Training for Employees | AI can tailor training programs to individual employee needs and learning styles, reducing training time and costs. | A software company uses AI to create personalized training modules for new employees based on their skill gaps and learning pace. |
