Automated assessment tool
Lean Canvas Business Plan
October - November 2023 - CARDIFF UNIVERSITY
CMT313 - Group Project
The first phase of a full software development lifecycle project — where we took an idea from zero to a structured business case, applying the same lean thinking used by real engineering teams to ship products.
The Problem- Teaching staff at universities face a significant overhead managing large-scale assessments — distributing work, tracking submissions, and providing timely feedback at scale. Our team identified this as a solvable software problem and used the Lean Canvas framework to define the business case for an automated assessment tool.
My Contribution- While this was a group project, I owned three specific areas:
Competitor Research — investigated existing assessment tools on the market to identify gaps and inform our unique positioning
Unique Value Proposition — defined what would make our tool meaningfully different from competitors, grounding it in the pain points we identified in current systems
Cost Structure — mapped out the financial model, considering development, hosting, and operational costs
Legal Risk Assessment — identified key legal challenges in the risk register, including data protection (GDPR), student privacy, and liability considerations around automated grading
The Approach- We applied lean thinking to strip back to what mattered most — the core problem, the users (teaching staff and students), and the minimum viable solution. Alongside the Lean Canvas we built a risk register covering technical, legal, and operational risks to ensure we were thinking critically about the full project lifecycle from day one.
As part of this module we also attended a workshop with Red Hat on Agile methodologies and software sprints — which directly shaped how we approached iterative development in the later phases of the project.
Lean Canvas
Risk Assessment-
Challenges- Defining a truly unique value proposition was harder than expected — the edtech market is crowded and many tools already claim to solve assessment problems. It pushed me to go deeper in my competitor research and think more precisely about which users we were solving for and what specific pain point we were addressing, rather than making broad claims.
On the legal side, automated grading raises genuine questions around fairness, bias, and student rights — areas I hadn't considered before. Researching GDPR compliance and academic integrity obligations was a valuable introduction to thinking about software not just technically but ethically and legally.
Key Takeaways
Competitor research isn't just about listing rivals — it's about finding the gap your product can credibly own
Legal and ethical considerations aren't an afterthought; they shape what you can actually build
Lean Canvas is a genuinely useful tool for cutting through ambiguity at the start of a project — I've applied the same structured thinking to scoping personal projects since
Next Steps
The next phase of this project was to explore and define the requirements for our automated assessment tool's minimum viable product (MVP). This involved:
Top-Level Use Case Diagrams: Collaboratively creating a diagram to highlight the key features and services of the system, ensuring a user-focused perspective.
User Stories and Acceptance Criteria: Developing detailed user stories for each key feature to specify functionality and validate requirements effectively. These were crafted from the perspective of educators and students to ensure alignment with their needs.
Non-Functional Requirements: Documenting essential software product quality requirements, such as system performance, scalability, security, and accessibility, to ensure the solution meets broader operational standards.
