On this page
Customer discovery at college fests starts when you stop treating a crowded campus as an audience and start treating it as a field site. By the end of a fest day, you should know which problem people describe in their own words, who feels it often enough to act, and what evidence you need next. A stall can produce hundreds of conversations; only a disciplined process turns them into a startup decision.
Set a discovery goal before the fest
A fest is loud, time-bound, and full of distractions. If you arrive with a vague aim such as “getting feedback,” you will leave with compliments, social follows, and no usable evidence. Set one learning goal for the day: understand a problem, test a target segment, compare current alternatives, or assess willingness to change behaviour.
Do not try to validate your entire startup in one event. A student founder building a roommate expense tool, for example, may need to learn whether students struggle more with tracking expenses, collecting money after a group purchase, or avoiding awkward follow-ups. Those are different problems, with different users and different products.
Write your learning goal as a sentence that can be proven wrong: “We believe hostel students who organise group purchases face repeated payment collection delays.” That statement gives your team a clear job. You are looking for frequency, context, existing workarounds, and consequences—not approval for your idea.
At Nebula, we treat validation as work that reduces a specific uncertainty. Our venture-building process moves from idea and market questions toward product, fit, funding, and scale. A fest belongs at the early end of that work: it can expose a real problem quickly, but it cannot replace repeated conversations with a defined customer group.
Choose the right people at the fest
College fests bring together students, organisers, sponsors, food vendors, faculty, alumni, and visitors. They are not one customer segment. If you interview whoever stops at your booth, your notes will mix people with different needs and produce a false pattern.
Define the person you need to speak with before your team reaches the venue. Use observable criteria. Instead of “college students,” recruit “students who live in hostels and order food with friends at least once a week,” or “cultural-fest organisers who managed vendor coordination this semester.” Your screener should take less than thirty seconds.
| What you need to learn | Who to recruit | Simple screener |
|---|---|---|
| Recurring personal problem | People who recently faced it | “When did this last happen?” |
| Operational workflow | Organisers, club leads, vendors | “Did you handle this during the fest?” |
| Payment behaviour | People who made the purchase | “Were you the one who paid?” |
Use a few screening questions, then politely end interviews with people outside the segment. A clean set of 20 relevant conversations is more useful than 100 random opinions. If your product serves both students and organisers, assign separate interviewers and keep the notes in separate sheets. Never merge their responses because both groups happened to attend the same event.
Ask about behaviour, not your solution
Your first question should not describe your product. “Would you use an app that solves this?” invites a quick yes because the respondent wants to be helpful, curious, or polite. That answer tells you almost nothing about whether they will change what they do today.
Start with a recent incident. Ask them to walk you through what happened, in order. The detail is where useful evidence sits: who was involved, what triggered the problem, which tool they used, how much time it took, what went wrong, and what they did after that.
- “Tell me about the last time you had to deal with this.”
- “What did you do first?”
- “What did you use to manage it?”
- “What was difficult or irritating about that process?”
- “What happened when the issue was not solved?”
- “Have you tried anything else? Why did you stop?”
Ask one question at a time and wait. A pause often produces the most specific answer. Avoid filling it by explaining your concept. If someone mentions a workaround—a WhatsApp group, a spreadsheet, a friend who handles the task—ask to see it when appropriate. A real workaround is stronger evidence than a stated preference because it shows the person has already spent effort to cope with the problem.
End with a recruiting question: “Who else has dealt with this recently?” The best fest interview may lead you to a later conversation with the person who has deeper experience and clearer stakes.
Build a booth that earns conversations
A discovery booth should make the right person stop without turning the interaction into a product pitch. Put the problem context on the sign, not your feature list. “Do group expenses become awkward after the trip?” is a better opening than “Try our expense-splitting app.” The first invites a story; the second asks for a verdict.
Give every team member a defined role. One person opens, one conducts the interview, one records notes, and one manages the queue. When the same founder tries to do all four, they either lose answers while writing or cut the interview short because another group is waiting.
Field rule: Do not reward people for saying they like your idea. If you offer a game, snack, or small prize, make it available for completing the interview, regardless of their response. Your incentive should pay for attention, not agreement.
Keep the interaction short enough for a fest setting, but do not chase speed at the cost of detail. If a respondent has a relevant story, ask whether they can spare a few more minutes or schedule a follow-up. Record consent before taking photos, audio, or contact details. For student founders, that discipline matters: trust is part of your early operating standard.
If you need a sharper validation plan after the fest, our Startup School is built to help founders move from assumptions to investor-ready work.
Capture evidence while it is fresh
Memory is a poor research system. At a busy fest, five interviews can blur into one unless you capture the same details every time. Use a shared form or sheet with fixed fields. Do not leave note-taking to the end of the day, when your team will remember the loudest conversation rather than the most representative one.
| Field | What to record |
|---|---|
| Segment | Role, context, and screening criteria |
| Recent event | What happened and when it happened |
| Current method | Tools, people, or workarounds used today |
| Cost of the problem | Time, money, stress, missed outcome, or conflict |
| Exact quote | The respondent’s words, without interpretation |
| Follow-up status | Contact permission and next conversation |
Separate facts from your interpretation. “She used three WhatsApp groups to coordinate volunteers” is a fact. “She needs a volunteer-management platform” is your hypothesis. The distinction prevents a familiar mistake: turning a single detail into a product requirement before you have checked whether the pattern repeats.
Tag each interview after it ends: strong signal, weak signal, outside segment, or follow-up needed. The tags are not final conclusions. They help your team review the day without losing track of which conversations deserve further work.
Separate signal from fest noise
Fests create a bias toward novelty. People may enjoy your booth, praise your prototype, and still never use it after they leave. Treat excitement as a reason to investigate, not as validation. The evidence that matters is repeated behaviour from people who match your segment.
Run a thirty-minute debrief immediately after each day. Review notes one interview at a time before discussing product ideas. Ask what was actually said, which problem appeared repeatedly, and where answers contradicted your original belief. Make a simple count, but keep the underlying stories visible.
- Strong signal: a recent, recurring problem with a clear workaround and a visible cost.
- Promising signal: the problem is real, but frequency or urgency remains unclear.
- Weak signal: praise for the concept without a relevant past experience.
- Noise: answers from people outside the segment or answers shaped by your pitch.
Look for disconfirming evidence on purpose. If half your target users say the problem occurs only once a year, your original market may be too weak even if the other half sound enthusiastic. If organisers already solve the issue with a process they trust, your next task is to understand why that process persists.
Do not announce that you have product-market fit because a fest went well. You have earned a better hypothesis and a list of people to investigate further.
Turn fest interviews into a validation plan
The real work begins after the stalls close. Within 24 hours, clean your notes, group recurring themes, and write down the three strongest assumptions still open. Then choose the smallest next test that can reduce one of them. You may need follow-up interviews, a manual service, a landing page, a prototype test, or a conversation with the person who controls payment.
Convert every discovery finding into a decision. If students describe the problem but do not act on it, do not build yet; test urgency. If one segment reports frequent pain while another does not, narrow your early customer. If people already pay for a workaround, investigate pricing and the buyer. Your next step should follow the evidence, even when it sends you away from the feature you wanted to build.
Keep a research log that records the assumption, evidence, decision, and next test. This becomes useful when you speak with co-founders, mentors, and early investors. A founder who can explain what they learned, what changed, and why they changed it signals operating discipline.
Nebula is a Tamil Nadu-based venture builder building for India. We work alongside founders across validation, product, fundraising, and go-to-market because discovery only matters when it changes execution. If you are a student founder ready to turn field evidence into a fundable plan, Apply for Nebula 1.0.
Your fest booth is not the startup. It is a controlled chance to find out whether a startup should exist, for whom, and what you need to prove next. Run the conversations with that standard.
Enjoyed this? Get the next one in your inbox.
Fundraising guides and validation frameworks, every two weeks. No spam.
Frequently asked questions
How many interviews should a student founder conduct at a college fest?
Prioritise relevant interviews over volume. Stop when your team has enough detailed conversations to see repeated patterns, then schedule follow-ups with the strongest matches.
Should we demo our product during customer discovery at a fest?
Start by learning about the person’s current behaviour without showing the product. Show a prototype later only when you have a specific assumption about usability, value, or willingness to change.
What should we do after customer discovery at a college fest?
Review the notes within 24 hours, group repeated problems and workarounds, identify open assumptions, and run the smallest next test that can reduce the most important uncertainty.
Ready to build your startup?
We work with a small number of founders each year — mentorship, fundraising support, and a co-founder network included.
Start a conversation
Talk to the founder directly. We reply within two working days.
Applying to Nebula 1.0? Apply here →