Turn Student Work into a Credible Case Study Without Exposing Private Information

Tom — Singapore Digital Business & Systems Consultant

Tom

Digital Business & Systems Consultant · Singapore

7 min read Article date First published
Turn Student Work into a Credible Case Study Without Exposing Private Information

You can turn an internship task, a part-time job improvement or a small pilot into a credible student business case study for your portfolio, but permission comes before polish. Separate your part from the team's, check your agreement and school rules, ask whoever owns the work and remove everything private. Describe results as what happened, not what you caused. If the answer is no, keep the work private or rebuild it as a labelled fictional demo.

Key takeaways

  • Write "I" only for your own part.
  • Check your agreement and school rules, then ask the work's owner in writing.
  • Even with a yes, remove names, customer data, private screenshots, internal figures and unapproved logos.
  • Report what happened and what else changed; do not claim cause.
  • No answer or a no: keep it private, or make a labelled fictional demo.

Separate what you did from what the team did

Real work is rarely yours alone: a supervisor set the task, a colleague pulled the data, the owner decided what to keep. "I mapped the booking steps and drafted a new form; my supervisor chose the final version" beats "I redesigned the booking process". Name the role behind everything you did not do, and check that everyone you mention could agree with the page. My one-page portfolio case-study structure covers the layout; this guide covers what must happen before real work goes in.

Request permission and remove private details

First, re-read what you signed for the internship or job, including any confidentiality or ownership terms, and your school's rules for internships and coursework. I cannot tell you what your agreement says. If unclear, ask your supervisor or internship coordinator before drafting anything public.

Then ask whoever owns the work, usually the business owner, your supervisor or the pilot's client, plus any teammate whose work appears. Ask in writing and say exactly what you would show; agreeing to be described is not agreeing to be named. As with asking a business owner before an introduction, show only what was agreed and treat silence as a no.

"Hi [name], may I include a one-page case study on [the project] in my portfolio, describing only my part? I would not name [the business] unless you are happy for me to, and I would remove customer details, staff names and internal figures. I'll send you the draft to check first. It's completely fine to say no."

What to remove, even with a yes

DetailDefaultShow only if
Customer names, numbers, orders and chatsRemove completely, even if blurredNever; use made-up data
Staff or supplier namesUse a role, such as the supervisorThat person agreed
Screenshots of real systemsRebuild with made-up dataChecked line by line and approved
Sales, costs, prices, marginsRemove or replace, and say soThe owner approved those exact figures
Business name and logoLeave outThe owner agreed to be named

Customers gave their details so the business could serve them, not to appear in your portfolio. Singapore's PDPC advisory guidelines explain that an organisation must obtain a person's consent "before collecting, using or disclosing his personal data for a purpose". They also say the PDPA's Data Protection Provisions "do not apply to business contact information", but that term excludes details "provided by the individual solely for his or her personal purposes". This is not legal advice, and either way a supervisor's work email adds nothing, so leave it out.

Removing the name may not be enough: "a two-outlet optical shop near my campus" beside a CV listing your job there names it anyway. Never copy records to "anonymise later" either; as with mapping bookings from a hospitality shift, rebuild examples from memory with invented details.

Explain results without inventing causation

"After the change, enquiries went up" is an observation. "My change made enquiries go up" claims cause, which a short student project can rarely prove. A promotion, a price change or a festive rush can move the same number in the same month.

OverclaimHonest version
My new form increased enquiries by 30%Enquiries rose in the four weeks after the form went live (the owner's count), but a holiday promotion ran too, so I cannot separate the effects
Saved the team hours every weekNot measured: the team said replies felt quicker, but nobody timed them

An honest version says what you saw, over what period, who counted it, what else changed and what you cannot say. Then add a label from my portfolio guide: Demo, Tested with N people, Used by a real business with permission, or Not measured. If you changed details, say so: "Business not named; figures are made up."

The case-study evidence and permission checklist

If you cannot tick an item, that part stays out.

  • My part: each "I" sentence is something I did; the rest names the role.
  • Rules checked: agreement and school rules re-read; supervisor asked where unsure.
  • Owner's yes: in writing, and clear on naming, figures and images.
  • Team's yes: teammates agreed, or their part is described, not shown.
  • Private details out: no customer, staff or student names, numbers, orders or chats, even blurred.
  • Images rebuilt: made-up data; tabs, notifications and file names checked.
  • Figures and logos: none unless approved for exactly this use.
  • Not identifiable: if unnamed, my description and CV together do not reveal the business.
  • Honest results: what happened, when, who counted it, what else changed, plus a label.
  • Final check: the owner saw the final draft; I saved their reply.
  • A no, or no reply: the work stays private, or becomes a fictional demo labelled from the title down.

Fictional example: one no and one careful yes

(fictional example) Ashwin, a second-year university student, rebuilt a job-tracking spreadsheet during a ten-week internship at a small logistics firm. At weekends he works part-time at a family-run dry-cleaning shop, where he suggested a "ready for collection" WhatsApp template.

The no. He re-read his internship agreement and asked his supervisor. She said no: the spreadsheet holds client names and rates, and the firm prefers its process kept private. Ashwin kept no copies. He built a tracker for an invented courier firm with made-up jobs, titled "Job tracker for a fictional courier firm (demo)".

The careful yes. The shop owner agreed to a description without the shop's name or figures. Ashwin's first draft said "my template cut customer calls". He rewrote it: "The owner noticed fewer 'is it ready?' calls afterwards. Not measured: nobody counted calls before, and the festive rush had also ended." On the final draft, the owner asked him to remove a counter photo; he did.

The upside: two pieces he can show openly, and evidence that he handles information carefully. The effort and uncertainty: the owner took two weeks to reply, a demo carries less weight than real use, and neither piece proves his ideas worked.

Where a website or system fits

You do not need a website to write a case study. If you run a small service and want a site showing permission-cleared case studies beside a clear enquiry form, I build that through website development: custom-coded business websites with contact forms and handover documentation, for an agreed scope. If a case study reveals a repeated workflow problem, such as jobs tracked across WhatsApp and spreadsheets, that is closer to operations digitalisation: a custom web app with role-based access and a central database, also separately scoped. The business owner decides either way.

Sources (checked 31 October 2026)

Frequently Asked Questions

Can I show an internship project publicly?

Only after checking your internship agreement and school rules and getting the organisation's agreement in writing. If they say no, describe your skills generally or build a labelled fictional demo.

How should I label a demo or an unmeasured result?

Put it next to the result: Demo for anything with made-up data and no real users, Not measured for any effect nobody counted. Add what you would measure next.

What if the owner never replies?

Treat silence as a no: keep it private, or rebuild it as a fictional demo using none of their details.

Can I say my change caused the result?

Only if your test ruled out other causes, which short student projects rarely can. Say what happened, when, how it was counted and what else changed.

Has a case study shown that a business needs a clearer website, or would you like a simple site for your own small service that shows past work without exposing anyone? Message me for a no-commitment chat about what it would need. You do not need a website to write a good case study.

No obligation. No sales pitch. Just an honest conversation.

お問い合わせ