Project-Based Learning for Universities: Plan a Business Challenge

What you’ll learn

Start with one learning outcome, a bounded business problem and a shared assessment rubric. Run a small pilot, collect feedback, then decide what to expand.

1. Start with the learning outcome

Choose one course and one skill students should demonstrate. For a business analytics course, that might be interpreting sales data and defending a recommendation. For a marketing course, it could be proposing a customer segmentation strategy.

Define the submission before choosing the tools: a short analysis, a chart with an explanation, and a recommendation with limitations. This keeps the project focused on learning rather than producing a polished presentation.

2. Write a business brief students can investigate

Give students a specific question, relevant context, a usable dataset and clear boundaries. For example: which product groups contributed to a sales decline, and what should the company investigate next? Label synthetic or simplified data clearly.

Agree with the company on what can be shared. Use anonymized or synthetic information where appropriate, and make clear that student recommendations are learning outputs requiring review before business use.

3. Share the assessment rubric before work begins

Assess the quality of the question, evidence, reasoning, communication and reflection. Describe what strong work looks like for each criterion. A clear rubric helps students understand expectations and gives reviewers a consistent basis for feedback.

Ask each student to explain their contribution and defend one decision. In team projects, combine the group submission with an individual reflection so a polished team result does not hide gaps in understanding.

4. Make responsible AI use part of the learning

State which AI tools and uses are allowed under your institution’s policy. Ask students to record how they used AI, check calculations and sources, and explain where they rejected or revised a suggestion.

AI literacy includes recognizing uncertainty and protecting confidential information. Students should be able to reproduce important steps and explain their conclusions without relying on an AI-generated answer.

5. Run a small pilot with a feedback checkpoint

An illustrative four-week pilot could use week one for understanding the brief, week two for analysis, week three for peer or expert feedback, and week four for revision and presentation. Adjust the pace to your course timetable and student workload.

Confirm reviewer availability before promising expert feedback. Give reviewers the brief and rubric, and ask for comments students can act on before the final submission.

6. Turn the project into evidence of student skills

Help students create a portfolio summary: the problem, their contribution, the methods used, the evidence behind their recommendation and what changed after feedback. Obtain permission before publishing company material or student work.

Evaluate the pilot using submission quality, student reflection, reviewer feedback and staff workload. These are useful early indicators; a single project does not establish an improvement in graduate employment.

Innovatech helps institutions explore guided business missions, practical AI and data learning, and project feedback. Discuss a pilot around your curriculum, student level and available teaching time before choosing a wider rollout.