What do you do if stakeholders in your AI project have unrealistic expectations and demands?

It's all about expectations management. For some reason people are okay with other people making mistakes (a clerk giving you wrong change, a support specialist not knowing how to resolve a problem) but when the machine (AI) is performing the task they expect 100% accuracy. The solution is often to determine if a certain case will or won't be solved properly by AI and in the case of high risk of failure you delegate certain case to the employee flow rather than the AI agents.

Start with the education. Often the unrealistic expectations come from lack of understanding of AI capabilities. Explaining them in simple terms, often drawing analogy of human capabilities help. Show case studies. Show what others are doing. Encourage them to go beyond marketing literatures, which are often misleading, and focus on true implementations. Bring in external voices. To infuse more confidence, having somebody educated and from external organisation helps. Having a discussion with analysts (e.g. Gartner, Forrester, HfS) can bring in perspective which are often missed within the organization.

Because of the AI "hype", stakeholder expectations are often "inflated". Explain early and clearly what are the limitations and constraints of the technology. Furthermore, you should avoid scope creep. Keep stakeholders engaged by involving them in the decision making progress.

If stakeholders in your AI project have unrealistic expectations and demands, first, set up a meeting to openly discuss their expectations. Clearly communicate the capabilities and limitations of AI, using data and examples to provide a realistic picture. Educate them about the time, data, and resources required for AI projects, and the importance of setting achievable goals. Offer alternative solutions or compromises that align with what is technically and practically feasible. Regularly update stakeholders on progress, challenges, and adjustments to manage their expectations continuously. Establishing a transparent, educational, and communicative approach helps align stakeholder expectations with the project's realistic outcomes.

- Address unrealistic expectations and demands of stakeholders early on in the project - Connect stakeholders to the customer and focus on a narrow use case with achievable business goals - Ensure a diverse team representing different disciplines (product management, UX design, engineering, data science) - Work together in lockstep to ensure project feasibility and alignment with overall goals - Have a clear understanding of data requirements and limitations - Prototype the vision to avoid potential failures - Have the right people driving the project - Take a pragmatic approach to data availability and usage - Focus on high-value use cases rather than trying to solve all data issues before starting the project

I often turn to my strategy of setting boundaries, but with a twist that involves clarity and a dash of creativity. For example, during a particularly ambitious project to integrate AI into our legal document analysis process, expectations were soaring beyond the stratosphere. I decided to host a "Future Gazing" session where we mapped out the project's goals and timelines on a timeline that resembled a journey through the cosmos. Each "planet" represented a major milestone, and the "asteroids" were the challenges we might face. This visual and engaging approach helped everyone understand the realistic scope and scale of the project, grounding expectations in the realm of the achievable, without resorting to corporate speak or buzzwords.

Educate on AI Capabilities and Limitations To align expectations, educating stakeholders about the capabilities and limitations of AI technology is crucial. Providing examples of successful AI projects, while also highlighting the challenges encountered and the iterative nature of AI development, can help stakeholders understand what is feasible. This education should also cover the data requirements, processing times, and the importance of quality input for effective AI outcomes. Such understanding fosters realistic expectations from the outset.

Imagine a shop captain navigating through treacherous waters, where every decision counts. In AI projects, setting boundaries with stakeholders is like charting a course for that ship. ?? You need to clarify what is achievable within the project's scope, timeline, and resource constraints, much like a captain would explain the capabilities of their ship and crew, and the impact of weather conditions on their journey. ???? This clarity from the start acts as a compass guiding the project through the high seas of ambitious expectations towards the shores of achievable outcomes, ensuring that all abroad are prepared for the voyage ahead, aware of what the journey can realistically offer.

Define the right scope for your AI projects. Do not throw AI technologies at every process in your organization at once. Find those processes that will benefit the most from an AI enhancement. You define where AI in company provides benefits within the ethical and cultural framework of your organization. Get the processes and the data right first and then commit to clear goals. Define KPIs for your projects that make it easer later on to follow up on the success of your AI integration.

Begin by defining project scope, objectives, and deliverables, clearly communicating what can and cannot be achieved with AI technology. Educate stakeholders on the capabilities and constraints of AI, emphasizing the importance of aligning expectations with these realities. Encourage open dialogue and collaboration while firmly asserting boundaries to prevent scope creep or unrealistic demands. By proactively managing expectations and boundaries, stakeholders gain a clear understanding of project parameters, fostering a more productive and successful AI implementation process.

Here is a To-Do List from my personal experience to communicate your progress effectively: 1) Weekly Updates: Schedule weekly update meetings or emails ?? to share the latest project milestones and hurdles. 2) Dashboard Access: Provide stakeholders with access to a live project dashboard for real-time progress tracking ??. 3) Success Stories: Share short, impactful stories of project successes ?? and how they were achieved. 4) Challenge Briefs: Briefly outline challenges encountered, solutions considered, and the outcomes of those solutions ??. 5) Q&A Sessions: Hold regular Q&A sessions to address stakeholder questions and concerns directly ??.

Communicate Progress Regular communication about the project's progress is vital. This includes both successes and challenges faced along the way. Keeping stakeholders informed helps manage their expectations and fosters trust. When they understand the complexities involved in developing and deploying AI solutions, they are more likely to have a realistic view of the project's trajectory and potential outcomes.

Regular communication about the project's progress is key to managing stakeholder expectations. I implement a structured update system, providing weekly or bi-weekly reports that highlight achievements, challenges, and next steps. This continuous stream of information keeps stakeholders informed and engaged, reducing the space for unrealistic expectations to form. It also opens the door for timely feedback, allowing for adjustments that keep the project aligned with stakeholder needs and real-world constraints.

In navigating the complex waters of AI project management, I've found that keeping stakeholders informed without overwhelming them requires a delicate balance. To tackle this, I adopted a strategy where I distill our progress into digestible, engaging updates. For instance, in a project aimed at enhancing our legal analytics platform, rather than deluge stakeholders with technical details and data points, I chose to share concise, story-like updates. I framed our challenges and achievements within the context of a journey, highlighting key milestones as chapters in our ongoing saga, without resorting to overused metaphors or jargon.

Regular communication about the project's progress is vital for ensuring transparency, alignment, and stakeholder satisfaction. It involves providing frequent updates on milestones, achievements, challenges, and any changes to the project timeline or scope. This ongoing dialogue fosters trust and collaboration among team members and stakeholders, allowing for timely feedback and adjustments as needed. Additionally, it helps to manage expectations effectively by keeping stakeholders informed and engaged throughout the project lifecycle. By maintaining open lines of communication, potential issues can be addressed proactively, leading to smoother project execution and successful outcomes.

Education and Communication Educating stakeholders about the AI project's potential limitations and challenges is crucial. Hold informational sessions to discuss AI concepts, development processes, and realistic project timelines. Transparent communication about potential setbacks and the technical complexities involved can adjust stakeholders' expectations to be more in line with what is achievable. This approach ensures that everyone is on the same page regarding the project’s scope and potential outcomes.

I engaged stakeholders from the ideation phase by conducting workshops inviting input on potential applications of Gen AI. This participatory approach leverages the diverse expertise and insights of the stakeholders and ensures they have a personal stake in the project's outcomes. Their active contribution and regular feedback become invaluable assets, driving the project forward with a shared vision. This continuous engagement keeps stakeholders informed of progress and empowers them to steer the project in response to evolving challenges and opportunities.

This can be transformative, leading to more successful outcomes and increased stakeholder buy-in. You can leverage their expertise, insights, and perspectives, which can enrich the project's development and implementation. Collaborating with stakeholders fosters a sense of ownership and commitment, as they feel invested in the project's success. Plus this facilitates alignment of project goals with organisational objectives, ensuring that the AI solution meets the needs and expectations of key stakeholders. This collaborative approach promotes transparency, fosters trust, and enhances communication, ultimately leading to more effective decision-making and a higher likelihood of achieving desired outcomes.

Properly involving stakeholders in the project is truly a game-changer. It's important to understand that in most cases, stakeholders possess business knowledge, and you possess the technological expertise. Even if you have 'the best team' in the world, you won’t be able to build the 'right' project 'on time' without properly balancing these two aspects together. Continuous collaboration and emphasizing the team aspect will help you identify goals properly, be aligned together, and, as a result, build more reasonable expectations.

Actively involving stakeholders in the development process can significantly mitigate unrealistic demands. Seek their input, address their concerns, and make them feel heard. When stakeholders understand the intricacies involved, they're more likely to have reasonable expectations. This, again, is no different from good practices of any software project.

I've learned the power of offering alternatives without falling into the trap of jargon or overused strategies. On a project aimed at developing a new client intake system using AI, I encountered resistance due to concerns over its complexity and implementation time. Instead of pushing back with technical rebuttals, I presented a menu of options, each tailored to different phases of readiness and resource availability, akin to choosing dishes from a diverse culinary spread. This approach allowed stakeholders to visualize the project's trajectory through practical, tangible alternatives, akin to selecting ingredients for a custom meal, ensuring their concerns were addressed with flexibility and creativity.

When stakeholders' expectations soar skyward, construct a village before the castle. Present quick-win projects as the foundation, demonstrating AI's value in tangible terms. Suggest skunkworks projects as inventive outposts, running parallel to the main endeavor. These serve as proof of concept, marrying ambition with pragmatism. It’s about shifting from "all or nothing" to "brick by brick," subtly guiding expectations to align with technological feasibility. This approach doesn't just temper aspirations; it transforms them into a series of achievable milestones, laying a realistic pathway to the grand vision, proving step by step, that innovation is built on solid ground.

If stakeholders' demands are unrealistic, offer alternative solutions that can still address their underlying needs. Present different options that are more feasible and align with the project's objectives.

Sometimes (a lot of times), AI is not the solution. The problem can be solved with "traditional" software. Clearly outline the trade-offs and compromises required. Collaboratively explore solutions that align with the project's constraints while still delivering value. Flexibility and a problem-solving mindset are invaluable in such situations.

This presents an opportunity to demonstrate adaptability and creativity in problem-solving. By offering alternative solutions that are more achievable, you showcase flexibility and a proactive approach to addressing challenges. This not only helps manage stakeholders' expectations more effectively but also fosters a collaborative environment where constructive dialogue and innovation thrive. Additionally, presenting alternative solutions demonstrates your commitment to finding practical and realistic ways to meet stakeholders' needs and objectives. Ultimately, this approach can lead to greater stakeholder satisfaction, as they appreciate your efforts to navigate obstacles and deliver valuable outcomes despite initial setbacks.

Be open to adjustments and adaptations throughout the project. As new information arises or circumstances change, be willing to modify the project plan or adjust expectations accordingly. Flexibility can help manage stakeholders' evolving demands.

AI is a rapidly evolving field, and stakeholder expectations may shift as new possibilities emerge. Remain agile and adaptable throughout the project's lifecycle. Regularly reassess priorities, timelines, and resource allocations. Embracing change is crucial for meeting evolving expectations while maintaining project integrity.

It is important to remember that managing stakeholders' expectations is an ongoing process. Regular communication, proactive education, and a collaborative approach are key. Additionally, documenting decisions and agreements can serve as a reference point to ensure everyone is on the same page.

When stakeholders have unrealistic expectations and demands in an AI project, it's vital to address these challenges through clear communication and education. Firstly, organise a meeting to discuss the project's scope, limitations, and realistic timelines. Use data and case studies to illustrate the complexities and potential challenges. It's beneficial to set up regular check-ins to keep stakeholders informed on progress and any emerging issues. Emphasise the importance of flexibility, as AI projects often require adjustments based on iterative testing and feedback. By managing expectations through transparent, informed conversations, you can align stakeholder demands with the project's capabilities and progress.