Generative AI for Business Analysis

YouTube search... ...Google search

It's essential to recognize the transformative potential of Generative Artificial Intelligence (AI), a rapidly advancing technology that's revolutionizing various fields, including Business Analysis. Looking forward, it's exciting to envision how requirements elicitation could be accomplished in the not-so-distant future. For instance, suppose your team is tasked with developing a new capability, such as internet infrastructure on the moon, to meet the needs of your organization's customers. In that case, you could leverage various types of generative AI to extract requirements and ensure that the resulting infrastructure aligns with stakeholder needs.

• First, the team selects from the organization’s service library an app to create create a chatbot that can interact with stakeholders and answer questions about the proposed internet infrastructure. This chatbot is trained to understand the technical language used in the industry, making it easier for stakeholders to communicate their requirements.
• The team uses voice generative AI to create a sample voice for the chatbot. They test different voices and accents to see which one is the most effective at conveying information to stakeholders. The selected voice is then integrated into the chatbot.
• The team uses using generative AI to create visualizations, simulations, and other representations of a proposed solution such as the team uses video generative AI to create a simulated video of how the internet infrastructure would work on the moon. This video includes different scenarios, such as how astronauts would access the internet and how data would be transmitted. The video is shared with stakeholders to help them visualize the proposed infrastructure and provide feedback.

Source: NASA

Solution Envisioning: The utilization of diverse forms of generative AI has enabled our requirements team to develop a highly precise and comprehensive understanding of the requirements and performance metrics for the lunar internet infrastructure. This innovative approach has provided our stakeholders with a visual representation of potential scenarios and outcomes, allowing them to identify any possible issues and opportunities prior to implementing the solution.

Generative AI has facilitated our stakeholders to explore and experiment with different courses of action, thereby enabling them to select the most effective solution that caters to their specific needs. This has resulted in greater innovation, as stakeholders have been able to evaluate ideas and concepts that may have been overlooked in the absence of generative AI. Moreover, this approach has significantly reduced the risk of errors, oversights, and misunderstandings during the process, leading to a more effective and efficient solution. The advent of generative AI marks a paradigm shift in the field of business analysis. It has enabled us to uncover hidden patterns and devise innovative solutions that were beyond our imagination before. We stand on the cusp of a revolutionary transformation in the way we conduct business analysis.

---

Generative AI is revolutionizing the way we approach requirements gathering in business analysis by uncovering hidden patterns and generating novel solutions that were previously unimaginable

---

Same Skill, New Techniques: In order to achieve optimal outcomes, it is important for analysts to continuously enhance their skills in information extraction. This involves identifying relevant data from the environment and extracting vital information to drive informed decision-making. Analysts are experts in working with stakeholders to synthesize gathered information and identify patterns and trends that can address business needs and promote success.

However, with the increasing complexity of modern environments, analysts require additional tools to supplement their refined skills. Generative AI presents promising advancements that can complement and enhance the critical thinking and expertise of analysts. By incorporating generative AI as an adjunct tool, analysts can perform higher-level tasks such as information synthesis, requirement interpretation, and solution development more efficiently.

Generative AI chatbot capabilities such as OpenAI's ChatGPT, Perplexity's Perplexity, Microsoft's BingAI, and Google's Bard can provide analysts with real-time access to relevant information, suggest possible solutions, and help refine their analysis. The incorporation of Generative AI into the work of analysts highlights its potential as an invaluable aid that can seamlessly integrate their skills into its use.

By working in tandem with generative AI chatbots or using a "pair analysis" approach, analysts can collaborate and solve complex problems more efficiently. This approach ensures that multiple perspectives and ideas are considered, leading to more innovative and effective solutions. Ultimately, by combining the power of generative AI with human expertise, business analysts can uncover valuable insights, refine their analysis, and develop more effective solutions, ultimately leading to improved business outcomes.

Prompt Engineering:

Analysts use their information prompting skills to extract stakeholder needs and create effective prompts that enable better utilization of generative AI. To ensure optimal AI results, analysts need to craft concise and informative prompts that provide context, instructions, expectations, and output examples. These prompts enable analysts to get feedback, clarify requirements, validate assumptions, and resolve issues. When providing context, analysts can use generative content such as summaries, outlines, tables, charts, diagrams, etc. By following the prompts listed here, analysts can quickly and efficiently generate contextually relevant content that enhances the stakeholders' understanding of the data and the AI's outputs.

Your organizations may have standardized templates or artifacts that aid in this process, and analysts can leverage AI to generate content. The use of AI can enhance the conversation's uniqueness and interest, avoiding a rigid script and creating a remarkable experience. If standardized templates are not available, utilizing a structured prompt such as "Outline: {text}" can facilitate the discussion where {text} represents the subject matter or project. Continuously eliciting more details from the AI for each point on the outline can be accomplished.

---

Prompt: Outline: {text of background information}

... for quick demonstration... Outline: lunar internet

---

Engaging in conversations with generative AI, it is crucial to provide comprehensive context about the topic being discussed. This context should cover the history, current state, problem statement, goals, objectives, known future state, stakeholders, and success factors. It should also address the why, who, what, and when of the topic, as this information will enable the AI to understand the conversation's scope and purpose. By providing context upfront, the AI can better understand the subject and provide more accurate, relevant, and personalized responses. It can also lead to an efficient conversation by eliminating the need for the AI to ask clarifying questions or seek additional information. This will reduce ambiguity and confusion, enabling the AI to address the specific topic at hand promptly and effectively.

---

Prompt: List requirements by type: {text of background information}

... for quick demonstration... List requirements by type: lunar internet

Prompt: more

Prompt: Table that has the following columns id, requirement, Key Performance Indicator (KPIs), type

Prompt: more

Prompt: more

---

After the background information or topic has been introduced, shorter prompts can be used since the AI can comprehend the conversation by analyzing previous dialogs. This allows for more specific inquiries or prompts that build upon previous topics. For example, while discussing a project, one can ask "What are the possible challenges of this project?" or "How can we mitigate the risks associated with this project?" As the conversation progresses, it may not be necessary to repeat the initial prompt if the AI is already discussing the same topic. This aids in the natural flow of the conversation, allowing for better focus on the subject matter. Below are prompts analyst find useful:

• Background: {text}
  • History: {text}
  • Current State: {text}
• Strength: {text} or Have: {text}
• Weaknesses: {text}
  • Challenges: {text} or Problem: {text}
  • Limitations: {text}
  • Root Cause: {text}
• Opportunities: {text}
  • Goals: {text} and Objectives: {text}
  • Future State: {text}
  • Benefits: {text}
• Threats: {text} and Risks: {text}
  • Mitigate risk or Mitigate risk: {text}

• Scenarios and outcomes Use the following in prompts: Atomic, Complete, Consistent, Concise, Feasible, Unambiguous, Testable, Prioritized, and Understandable. Good user stories should be focused on a single feature, contain all necessary information, be consistent with project goals, be short and to the point, be technically feasible, be clear and easy to understand, have clear acceptance criteria, be prioritized based on importance, and be written in language that all team members can understand. User story requirements should be identified with IDs and traceable to higher-level project requirements.
  • Provide atomic user stories and having one "so that" in the answer: {text of requirement}
  • Provide acceptance criteria: {text of user story}
  • Provide functional requirements: {text of user story} using the acceptance criteria: {text of acceptance criteria}
  • Provide non-functional requirements: {text of user story} using the acceptance criteria: {text of acceptance criteria}
  • Provide transitional requirements: {text of user story} using the acceptance criteria: {text of acceptance criteria}
  • Provide table with the following columns "functional requirement", "non-functional requirement", "Description" based on the following transitional requirements: {text of transitional requirements}

• Analysis --
  • Stakeholders: {text}
  • Critical Success Factors: {text}
  • Paradigm shift with: {text}
  • Possibility of: {text}
  • Potential of: {text}
  • Provide use case: {text}
  • Process flow: {text} or Steps {text}
  • Identify pattern: {sequence of numbers}
  • What is the trend: {sequence of numbers}
  • Compare and contrast: {text} and {text}
  • Table comparing and contrasting: {text} and {text}
  • Pros and cons: {text}
  • Table of pros and cons: {text}
  • Estimate: {text}
  • Outline of the key findings: {text}

• General prompts --
  • Thoughts on: {text}
  • Describe: {text}
  • Example of: {text}
  • Rephrase: {text}
  • Rephrase using executive-level: {text}
  • Rephrase using 7th grade level: {text}
  • Simplify: {text} or Summarize: {text}
  • Summarize in 3 sentences: {text}
  • Summarize {URL} example Summarize: https://time.com/2803417/moon-internet
  • More details: {text}
  • List {text} example data generation list: 5 UHF frequencies
  • Table: {text}
  • Table that has the following columns: {text of columns} for {text}

Opportunistic Strategy: As organizational environments increase in complexity, analysts are implementing opportunistic strategies and acquiring new skills, such as prompt engineering. The emergence of user-friendly generative AI tools enables analysts to progress from problem decomposition to solution synthesis, creating fresh possibilities for problem-solving. By amalgamating generative AI with a process that involves deconstructing customer problems and transitioning from requirements to architectures, designs, and implementations, analysts can proactively impose structure on the disorder within an organization. This methodology empowers businesses to augment their proficiency and employ analytical techniques to scrutinize customer challenges, resulting in solutions that surpass the mere sum of their components. By adopting this approach, businesses and analysts can position themselves for success and take their organizations to the next level.

---

• Linked – How Everything Is Connected to Everything Else and What It Means for Business, Science, and Everyday Life | Albert-László Barabási
• Prompt Engineering (PE)
• Generative AI
• Business Strategy/Consulting
• Requirements Management
• Prompts for assessing AI projects ... Checklists