By Ravindra Babu, Head, Data Science, Sahaj Software
In today’s fast-paced business environment, leveraging advanced technologies is crucial to stay ahead of the competition. One such disruptive technology that could soon become intrinsic to business analysis is generative AI, exemplified by OpenAI’s GPT.
This powerful tool can prove to be transformative for business analysis, revolutionise the way analysts work and deliver unprecedented efficiency. Generative AI can expedite the analysis process by harnessing the power of AI algorithms to quickly analyse problem statements, define high-level features and objectives, and generate epics and user stories.
With AI tools, requirement gathering can be streamlined while enabling structured and organised analysis. By optimising this crucial software engineering practice, one can gain a competitive advantage in the dynamic business landscape. According to Gartner, approximately 75% of organisations are estimated to adopt AI by 2024. AI’s potential is undoubtable and its imprint across various stages of the software development cycle can be a game-changer to drive transformation.
Let’s look at how AI can transform business analysis:
Kickstarting the Analysis Process with AI
Identifying the problem statement and establishing high-level business goals or requirements is the first step in any business analysis process. By utilising OpenAI’s GPT and its chat completion APIs, analysts can engage in interactive conversations that result in rapid idea generation and requirement gathering. This AI-driven approach can significantly accelerate the initial stages of the analysis process, thus providing a solid foundation for subsequent steps.
Generating Epics and Stories with Prompt Engineering
Breaking down complex projects into manageable units is a fundamental component of business analysis. Epics and stories are the foundational elements in this process.
The problem statement can be provided as a prompt to GPT to derive high-level features which can then be transformed into epics, and further broken down into stories.
To ensure consistent style and format in story generation, it is essential to customise the AI model. While the GPT model inherently understands epics and user stories, providing a defined structure tailored to the desired writing style ensures uniformity in the generated output. This structure and format can be defined in JSON format to precisely outline the expectation for epics, stories, and acceptance criteria and ensure that the AI-generated stories align with the analyst’s specific needs.
Integrating User Interface (UI) Layouts and Storyboards: Connecting the Dots
Streamlining complex projects into manageable units lies at the core of business analysis, accomplished through the creation of epics and stories. From there, the focus shifts to UI layouts and storyboards. By combining responses from prompts for stories, UI mockup and acceptance criteria, the process takes shape although some consolidation proves less effective and demands trial and error; for example, the story might have a few missing details such as mandatory fields.
The UI mockup details can be used to generate a UI wireframe using a tool like Figma Wireframe Designer or visily.ai. Based on a prompt, these tools can generate wireframe mockups. These can serve as a great starting point for an analyst to further fill in the blanks by elaborating on the problem statement and features. As language models evolve, these interventions will become less intensive as incorporating more components with better prompts would yield more accurate and comprehensive results.
With all the stories in place, a storyboard can be finally set up bringing this phase to a closure.
Automating the Kickstart for Unprecedented Speed
Imagine each organisation or business analyst having the ability to define a customised structure for epics and stories according to their unique requirements. By leveraging a tool empowered by generative AI, the generation of functional and non-functional user stories, complete with acceptance criteria, can be automated. This will not only save time and effort but also ensure that requirements are well-defined and measurable. With a customised structure and a well-defined problem statement, business analysts can generate user stories with just a click. This approach will enable analysts to focus their energy on problem-solving and collaborating with stakeholders to refine and iterate on the generated output. Such a tool, incorporating prompts and a customised structure, can be seamlessly integrated into various projects, empowering business analysts across different domains.
Considerations and Collaboration
As more and more organisations embrace AI technologies like generative AI, the software development lifecycle as we know it is set to undergo a profound shift. Business analysts need to leverage these sophisticated tools to enhance their entire gamut of responsibilities from identifying high-level business goals to generating stories and epics.
While generative AI tools provide significant benefits, it is crucial to acknowledge their limitations. These tools may not comprehensively analyze every aspect, potentially leading to gaps in the analysis. To ensure accuracy and thoroughness, business analysts and developers must exercise critical thinking and apply their domain expertise to validate and refine the outputs generated by AI models. By combining the strengths of generative AI with human oversight, an efficient and agile business analysis process can become the norm.
