"We need to add AI to our customer experience."

This increasingly common directive is driving companies across industries to implement AI solutions, often without first asking the essential question: "What customer problems will this actually solve?"

Consider what happens when this crucial step is skipped: You need to change your seat before a flight, so you try your airline’s AI chatbot—only to get stuck in an endless loop of irrelevant FAQs with no option to reach a real person. Instead of making the process smoother, the AI creates friction, leaving you frustrated and stuck in the middle seat for six hours.

# The AI Implementation Gap in Customer Experience

AI is often marketed as the ultimate fix for modern business challenges, promising to automate tasks, craft personalized user journeys, and “revolutionize” how businesses engage with customers. With AI projected to grow to a $826.7 billion market by 2030 and 91% of companies optimistic about its role in customer engagement, the pressure to implement quickly is intense. But rushing to deploy customer-facing AI without intentional design and planning creates experiences that frustrate rather than delight.

The data tells a clear story:

  • 45% of businesses struggle to maintain personalization
  • 40% face accuracy issues
  • 32% wrestle with integration challenges

When done wrong, AI missteps can erode customer trust and inflate costs, turning AI into a liability instead of an asset. The most successful AI implementations begin not with technology selection but with comprehensive customer research that identifies genuine pain points, clarifies user expectations, and reveals the opportunities where AI can truly enhance experiences.

This blog explores the hidden costs of reactive AI adoption, key lessons from flawed launches, red flags to avoid, and essential factors to help you determine whether AI is truly enhancing the customer experience or just adding noise.

Related Content: Center People and Processes in Your Next AI Implementation

# The Hidden Costs of Jumping into AI Too Soon

AI adoption is accelerating, and according to Gartner’s Top Strategic Technology Trends for 2025, 70% of customer experience leaders believe AI bots are effective for personalization. With stats like that, it’s easy to assume implementing AI will lead to happier customers. However, rushing to adopt AI without a clear plan can lead to wasted resources and frustrated customers.

Organizations often fall into two traps:

  • The FOMO effect: Fear of falling behind competitors leads to premature AI adoption
  • Unclear ROI: Companies expect AI to improve customer experience but don’t define (or know) how to measure success

AI should make things easier, but it does the opposite when rushed or poorly implemented. Instead of improving experiences and enhancing value, it adds complexity, frustrates users, and delivers minimal benefit.

The real risk isn’t just wasted investment but losing customers’ trust. AI integration should amplify a strong user experience by making processes more efficient, personalized, and scalable. By addressing fundamentals first, companies can avoid costly mistakes and set the stage for AI to truly enhance the user experience and deliver value.

# Wendy’s AI Drive-Thru: When Automation Slows Things Down

Even the biggest companies get AI wrong. In early 2024, Wendy’s introduced AI-powered drive-thrus to streamline service, but instead of speeding up orders, it frustrated customers. The AI misunderstood accents, misheard background noise as orders, and forced customers to repeat themselves, leading to delays instead of efficiency.

One viral incident showed a customer ordering a simple burger, only for the AI to keep adding unwanted toppings and extra drinks. Repeating the order multiple times defeated the purpose of efficiency and led to longer wait times.

What Went Wrong?

  • Lack of contextual understanding – AI struggled with real-world speech patterns and environmental noise.
  • Lacked user-centric design – The system prioritized automation over customer needs, offering no easy way to correct errors.
  • Increased frustration, not efficiency – Instead of enhancing service, the AI slowed it down, adding friction where speed mattered most.

This case reinforces a critical lesson: AI should simplify and improve customer interactions, not add frustration. When automation is implemented without a clear understanding of user needs, it can erode trust, slow down experiences, and create barriers instead of removing them. Wendy’s experience highlights why AI must be tested in real-world conditions, designed with human-centered principles, and always leave room for human intervention when needed.

# Implementing AI Through an Experience Design Lens

To ensure AI delivers maximum impact and value, organizations should take an experience design approach. Rather than jumping straight to technology solutions, experience design offers a clear, iterative process for solving complex problems and creating user-centered AI solutions. It emphasizes alternating between divergence—exploring possibilities—and convergence—narrowing focus—to stay innovative while tightly aligned to real user needs.

# Three Design Principles for AI Implementation

  1. Define the Problem Space Clearly
    Start by identifying true problem statements and uncovering root causes. This avoids the common pitfall of solving symptoms rather than addressing core issues—a mistake that often leads to wasted AI investments.
  2. Thoughtfully Explore Potential Solutions
    Explore a range of opportunities without being constrained by existing technologies, processes, or organizational siloes. Align ideas with broader business priorities to ensure AI solutions drive strategic value, not just novelty.
  3. Co-Design with Users and Gather Early Feedback
    Engage users early and often to validate assumptions, learn quickly, and refine solutions before scaling. This promotes buy-in and ensures the AI experiences created are intuitive, helpful, and truly meet user needs.

By focusing on these points, organizations can avoid common traps: rushing to implement AI as a catch-all solution, investing in the wrong areas, or designing in isolation from actual users. An experience design approach helps AI initiatives stay human-centered, delivering real business value while enhancing customer experience.

ChatGPT: A Success Story in User-Centered AI

In contrast to Meta’s misstep, ChatGPT’s explosive growth to over 200 million active users in 2024 (doubling year-over-year) demonstrates the power of aligning AI capabilities with user needs and expectations.

OpenAI’s approach exemplifies user-centered AI development:

  • Continuous iteration: Early feedback led to improvements in accessibility, usability, and data protection.
  • User-driven enhancements: Surveys, forums, and usability studies informed updates that simplified workflows for all skill levels.

This user-centered approach resulted in over 1.6 billion monthly visits and propelled OpenAI’s valuation to $86 billion, proving that AI succeeds when it prioritizes real user needs.

# AI Readiness Checklist: Is Your Organization Prepared?

AI works best when layered onto a solid, well-researched system. Adding AI to flawed processes is like putting a high-performance engine in a car with flat tires—it has potential, but it won’t perform.

Before implementing customer-facing AI, assess your organization’s readiness across these five dimensions.

  1. Optimized Workflows: AI amplifies good processes but magnifies problems in bad ones. Are your current customer journeys efficient, streamlined, and working well without AI?
  2. Clear, Measurable Goals: Do you have a specific problem AI is solving, or is it being implemented without a clear purpose?
  3. Proven ROI Potential: Have you evaluated the cost-benefit tradeoff to ensure AI delivers tangible value (e.g., efficiency, revenue, or customer satisfaction)?
  4. Scalable Infrastructure: Can your systems support AI growth without lag, slowdowns, or instability?
  5. Strong Data Governance: Is your data clean, reliable, unbiased, and secure data, or could poor data quality compromise AI’s effectiveness?

When these factors are in place, AI can enhance workflows, personalize user experiences, and unlock real value.

Related Content: AI Governance: How an AI Council Could Make or Break Your Strategy

# Building for the Future: A Strategic UX-Driven Approach

Adapting to the fast pace of AI evolution requires infrastructure and measurement strategies designed to prioritize user experience (UX) while maintaining flexibility. Organizations that focus on creating adaptable, user-centered systems can navigate technological shifts while enhancing usability and satisfaction.

# Designing AI That Adapts to Users: Flexible Architecture for UX

Flexibility isn’t just about technology, it’s about empowering users with consistent, reliable, and intuitive experiences. Infrastructure must be designed to ensure the AI adapts to user needs as they evolve. That requires an architecture built for adaptability.

  • Modular: Allows seamless updates or integration of new features without disrupting the user experience. For example, adding a new AI capability should feel natural and intuitive for users rather than requiring them to relearn workflows.
  • Scalable: Ensures that as user interactions grow, the system performs efficiently, maintaining speed and responsiveness even under increased demand.
  • Adaptable: Supports the integration of emerging capabilities, ensuring the system evolves with user expectations and stays aligned with their needs.
  • Resilient: Maintains a seamless experience for users, even during system upgrades or technological disruptions. Downtime or instability erodes trust and user satisfaction.

This kind of architecture ensures the technology evolves alongside its users, making the system not only future-ready but user-friendly throughout its lifecycle.

# Prioritizing Metrics That Reflect UX Success

Traditional business metrics often fail to capture the nuances of user interactions with AI. Measuring AI success isn’t just about adoption rates but rather whether AI is actually enhancing user experience. Instead, businesses should focus on metrics that provide actionable insights into the user experience:

  • Task Completion Rates: Assess whether the AI is helping users efficiently achieve their goals, reducing friction in their workflows
  • Time-to-Value: Evaluate how quickly users benefit from AI features, indicating whether the tool simplifies or complicates their tasks.
  • Direct User Engagement: Track how users interact with AI, revealing the features they value most and highlighting opportunities for improvement.
  • Cost per Successful Interaction: Balance financial efficiency with user satisfaction, ensuring the investment in AI delivers tangible value to the user.

With flexible architecture and user-focused metrics, businesses can create AI systems that are not only adaptable to change but also continually improve the user experience. This approach ensures AI remains a tool that empowers and delights users, even as technology evolves.

# Conclusion: Making AI a Strategic Advantage

AI has the potential to enhance user experiences, but success depends on thoughtful, user-centered implementation. AI should amplify what works, not patch what’s broken.

Key Takeaways:

  1. Solve Real Problems: Start with user pain points, not trends. AI should be a response to real needs, not an experiment in novelty.
  2. Adopt Strategically: Align AI initiatives with measurable goals and test rigorously before scaling.
  3. Design for Scalability: Build systems that evolve alongside user needs.
  4. Measure What Matters: Use user-centric metrics to evaluate AI’s true impact.

AI’s greatest impact comes not from automation alone, but from how well it understands and supports human needs. When businesses approach AI with clarity, preparation, and user insight, it can deliver immense value, transforming workflows and transforming customer experiences.

Need a Smarter AI Strategy? Let’s Build It Together

AI works best when it’s intentional, user-centered, and built on a strong foundation. If you're thinking about AI adoption but aren't sure where to start, Propeller can help you:

  • Assess UX readiness to ensure AI enhances strong foundations.
  • Develop strategic roadmaps for seamless implementation and measurable outcomes.
  • Build feedback loops to refine AI systems based on real-world user insights.
  • Identify both quick wins and scalable opportunities for long-term impact.
  • Measure success through actionable metrics to optimize AI investments.

AI should enhance the way you work, not add complexity. Let’s build a strategy that makes AI intuitive, scalable, and truly valuable for your users.

Curious how to design AI that actually helps users? Learn more about our Human-Centered Design Workshop.