The Future-Ready Business Analyst: Why Generative AI Matters
How Business Analysts can accelerate digital transformation with generative AI and large language models, applying prompt engineering across project phases.
Introduction
Digital Transformation emerges as the North Star for forward-thinking enterprises. It isn't just about embracing the latest digital advancements but strategically integrating them to foster innovation and growth. Business Analysts play a pivotal role in the digital shift, turning ambitious visions into actionable, strategic steps. Adopting Generative AI is a strategic and competitive move that Business Analysts can leverage to accelerate a company's Digital Transformation strategy.
The Emergence of Generative AI
In 2022, the global artificial intelligence (AI) market's value reached an astonishing $136.55 billion. Although significant, the market value merely foreshadows what is to come. The GrandViewResearch predicts this figure will skyrocket to $1,811.75 billion by 2030, reflecting a compound annual growth rate (CAGR) of 37.3% from 2023 onwards.
Among these statistics, one fact is crucial: The McKinsey Global Survey of 2023 reveals 1 in 3 organizations is not only considering AI but has integrated it into their business operations. The shift towards AI adoption represents a change affecting technology and how companies approach their business strategy.
As the driving force behind these changes, Generative AI is proficient at creating new content, such as text-to-text and text-to-image, through associations based on its training data. Prime examples of Generative AI are Large Language Models (LLMs) like GPT-4 and Llama 2.
Accessing the capabilities of either of these models is now remarkably straightforward. Companies can integrate these language models into their business processes and experiment with them as their APIs are well-documented. With digitalization being the center of modern business strategies, LLMs are becoming a cornerstone for driving innovation.
Leveraging Large Language Models, we have observed the rise of Prompt Engineering, a practice we will delve into next.
Prompt Engineering: A Deeper Insight
Prompt Engineering is a part of the Natural Language Processing (NLP) domain, a subset of AI. NLP helps machines understand human language, while Prompt Engineering specializes in designing and refining prompts to elicit a desired output from a Large Language Model. Prompt Engineering comprises three distinct phases.
Problem Formulation
Problem Formulation marks the first step in utilizing an LLM’s ability for your company. Begin by identifying the business use case you aim to address and deconstruct it into small independent pieces. From there, define the prompt engineering scope of work by outlining the task's boundaries. At this stage, you can move on to designing your set of prompts.
Prompt Design
Prompt Design is where you lay out the roadmap for your Language Model. You must assess if your prompt contains the necessary details and guidelines the model must follow to achieve your desired result. The designed prompt must be easy to maintain over time. An easily maintainable prompt can be:
"Summarize the author's key points succinctly and ignore non-essentials. Return the main response only. Remove pre-text and post-text."
This prompt guides the model to summarize the text's main points and puts constraints on what it should ignore.
Prompt Engineering
Finally, Prompt Engineering is when you take in an operational prompt for ongoing optimization. Rather than "set it and forget it", prompts require regular inspections and adaptation. Iteratively refine the prompt's structure and context based on end-user feedback and the model's performance.
Let's delve into the ones that Business Analysts can use immediately to gain tangible benefits, separated into phase-specific applications.
Prompt Engineering Applications for Business Analysts
As a Business Analyst, you may be included in the project's Definition Phase. LLMs can create the initial skeleton for the Business Model Canvas when you possess limited information. Through proper prompt design, you can also gain insights into shaping the Product's vision and avoid additional rework in the future. You can also design targeted questionnaires for user interviews, accelerating data collection and insights gathering.
As for the Design and Planning Phase, you can use LLMs to create detailed Empathy and Scenario Mappings from the gathered user behavioral patterns. From there on, you can start compiling the initial Business Requirements and Requirements Architecture documentation by having Language Models analyze your work and suggest optimizations.
In the Development and Testing Phase, you can use your compiled documentation to create user stories. The LLMs can use the requirements and output constraints to generate story drafts. Another use during the Development Phase is that models can help with SQL queries or code explanations so you can be sure that everything is on track.
The Road Ahead
AI is here to stay, and you can leverage it to your advantage by using prompt engineering to optimize your work efficiency and speed up project delivery.
By delving deeper into Prompt Engineering and embracing it, you can secure your place in the AI future of Digital Transformation. Become the Business Analyst who contributes to the AI-driven Digital Transformation and strategically navigates it forward.
