Industry patterns and examples

In late November, ahead of its December meeting, the board asked a question Andre had expected: "What are our competitors doing?" Two large corporate veterinary chains had opened clinics in the region over the past three years, and one advertised "AI-powered appointment booking" on billboards. Some board members worried Cascade Ridge was falling behind. Others thought the billboard was mostly marketing. Andre needed a way to answer that was neither panic nor complacency.

This lesson covers the third task of Domain 2: positioning AI for competitive advantage. It looks at how to assess a competitive landscape, how AI can transform business models, which advantages last, and how to set the level of AI investment to suit an industry's maturity and competitive dynamics. It also surveys common AI patterns across industries, because many good ideas travel from one sector to another.

Assessing the competitive landscape

Noor suggested a structured look rather than reacting to advertising. A competitive landscape assessment for AI asks what competitors, new entrants and suppliers are doing with AI, and what that means for customers' expectations. Andre gathered information from public sources: competitors' websites and apps, job postings, trade press, vendor case studies, conference talks, and conversations with clients who had used other clinics.

The assessment changed the board's framing. The billboard was not a reason to copy the competitor's chatbot. It was evidence that client expectations for convenient self-service were rising, and that capabilities such as note drafting were about to become standard features of practice software. That second point mattered: there was little advantage in building something the group's own software vendor was about to provide.

Parity, advantage and differentiation

Strategists separate AI uses into three kinds, because each calls for a different approach:

• Parity uses keep the organization level with competitors and with customer expectations. Online booking with reminders, and answers to routine questions, were becoming parity at Cascade Ridge. The goal is to meet the standard at reasonable cost, usually by buying.
• Advantage uses make the organization measurably better than competitors at something customers value. Filling more appointment slots for sick animals, using the group's own booking history, could be one: it combines proprietary data with a redesigned process.
• Differentiation through trust is less visible but powerful. A group known for using AI carefully, explaining it honestly and keeping clinical decisions with veterinarians may win clients who are wary of impersonal, automated care.

Andre's recommendation followed from this split: meet parity cheaply, invest properly in the few advantage uses, and protect the group's reputation for trustworthy care in everything.

Which advantages last

The exam guide asks how AI creates sustainable competitive advantages and operational improvements. An advantage is sustainable if competitors cannot easily copy it. Noor summarized the main sources:

By contrast, an advantage based only on having access to a particular model or tool rarely lasts, because competitors can buy the same thing. Operational improvements can still be very valuable even when they are not unique: lower costs and less staff burnout matter whether or not a competitor achieves the same.

Transforming business models

Most AI initiatives improve existing operations. A few change the business model itself: how the organization creates, delivers and captures value. The exam guide asks you to identify such opportunities. Examples across industries include:

• From product to service. Equipment makers selling uptime guarantees instead of machines, using AI to predict failures before they happen.
• From one-off transactions to continuous relationships. Retailers using personalization to turn occasional buyers into subscribers.
• From reactive to preventive. Insurers and health providers using risk prediction to offer prevention programs rather than only paying for problems.
• New products from existing data. Organizations offering anonymized benchmarks or insights to partners, where law and consent allow.
• Self-service at scale. Professional services firms packaging expertise into AI-assisted tools that serve smaller clients economically.

At Cascade Ridge, the board discussed one transformation idea: expanding its wellness plans, which clients pay for monthly, with AI-assisted reminders for preventive care tailored to each pet's age, breed and history. It would move part of the business from treating problems toward preventing them, with steadier revenue. Dr. Kaur supported it on one condition: care recommendations would follow the group's clinical protocols, with the AI only deciding when and how to remind, not what care a pet needed. The board put it on the roadmap as an exploration for 2028.

Industry patterns

Many AI patterns appear across industries in slightly different forms. Knowing them helps a strategist recognize opportunities and avoid reinventing solutions.

Several patterns recur in every row: document processing, customer communication, forecasting and knowledge search. These are often good first initiatives for a mid-sized organization, because they are well understood, supported by managed services, and measurable.

How much to invest

The exam guide asks you to determine appropriate AI investment levels based on industry maturity and competitive dynamics. There is no universal right amount. Noor offered a simple way to think about it, using two questions: how mature is AI use in this industry, and how strongly does it affect competition?

Veterinary care in 2026 was moving: client-facing convenience and documentation tools were spreading quickly, but few groups had redesigned their operations around AI. Cascade Ridge's assessment placed it as a fast follower: invest in data foundations, buy for parity, and back two or three advantage uses with proper measurement. Being a fast follower is not timid. It means learning from others' mistakes and spending where the group's own data gives it a real edge.

Ratios like this are rough sense checks, not targets. An organization that spends the "right" share of revenue on poorly chosen initiatives creates no value. The investment level should follow from the portfolio of business cases, then be checked against posture and size to make sure it is neither reckless nor so small that nothing meaningful can be done.

The cost of waiting

Lorraine pressed the other side of the question: what does it cost to wait? Noor named three costs. First, client expectations rise whether or not the group acts, so parity gaps grow. Second, the data and skills needed for advantage uses take years to build; starting late means starting behind. Third, staff notice when competitors' tools make work easier, which matters in a market where technicians can choose where to work. None of these costs appears in a single year's accounts, which is why they are easy to ignore.

Learning from other industries

Some of the most useful ideas for Cascade Ridge came from outside veterinary care. Airlines and hotels had long used no-show prediction and overbooking; the group would never overbook appointments, but the airline practice of offering waitlisted customers a seat when a cancellation is likely translated directly into its waitlist design. Retailers' demand forecasting suggested how to forecast pharmacy stock by product and location. Banks' experience with document processing showed the value of a human check on extracted values before they enter a record.

Borrowing ideas works best when the strategist asks what is different about the new setting. Airlines can accept occasional overbooked passengers with compensation; a clinic cannot turn away a sick animal. Retailers can tolerate a stock-out of a sweater; a clinic cannot run out of an emergency drug. Each borrowed pattern was adapted with these differences in mind, which is the real skill behind learning from other industries.

For Andre, the main lesson of the session was proportion. The board did not need to match every competitor's AI feature, and it did not need to ignore them. It needed to know which uses were becoming parity, buy those sensibly, and put its real effort where its own data and processes could make it better than the competition.