1Bottom line
You can still do well from zero, because Module 1 grades what you learn from real conversations, not how good your first idea is. The real constraints are time and access to people, and both are solvable in a week.
Why this is more doable than it feels
- Module 1 grades learning, not a brilliant idea. The brief calls it a progress update on your learning rather than a final pitch. It asks for initial hypotheses that your research can confirm, partially confirm or reject, and for what changed in your thinking. A first idea that interviews knock down is a legitimate, even strong, result.
- Starting ideas are meant to be rough. Y Combinator's advice is to start from a problem rather than a solution, and not to worry that the first idea be great, because ideas change as you learn (UCI's summary of YC's guidance). Paul Graham's version: look for problems you or people around you actually have, rather than inventing ideas in a brainstorm (How to Get Startup Ideas).
- Fifteen interviews in a week is a normal pace. NSF I-Corps teams are expected to complete at least fifteen customer interviews a week, a hundred over seven weeks (Illinois I-Corps). Your team needs 15 in total, which is roughly three to five each.
- Access beats originality. The Mom Test's advice is to keep slicing your customer group until you know who they are and where to find them (summary). For the next week, the best problem is one whose sufferers you can reach in days, and fellow OU students are the most reachable people you have.
The plan in six lines
- Today: run a 75-minute idea sprint (§3) in class and finish with three problem hypotheses, about people you can reach this week.
- Tonight: message 30–45 people, warm contacts first, and book tomorrow's first interviews (§4).
- Thu–Sun: complete 15 interviews, about three a day across the team. Log each one the same day. After interviews 5–6, hold a 20-minute debrief and narrow to one lead concept (§5).
- Mon Sep 28: use the class work session to synthesize what you heard and build the slides.
- Tue Sep 29: rehearse twice on a timer, then upload (Boomer). Sooner teams have two extra days.
- Checkpoints: at least 4 interviews by Thursday night, 8 by Friday night, 11 by Saturday night. If you are three or more behind, tell the professor the next morning, with a plan (§6).
2Today, Wed Sep 23
Today's class is a team work session on ideation exercises, and the syllabus says the professor is available to guide teams during these sessions. Office hours are also today, 3–5 PM in person (PH 3050G). Use both. It's the best-resourced day you have.
| When | Do this | You should end with |
|---|---|---|
| Now, before class | Group chat, 15 minutes. Everyone writes 8–10 frictions from the last month (school, a job, housing, getting around, hobbies, family). Reopen your Aug 31 pitch slides for each person's "venture interests" and "deal-breakers." Pick a note-taker. | A shared list of about 30 frictions, plus each person's deal-breakers. |
| In class | Run the 75-minute idea sprint (§3) alongside the professor's ideation exercises. Ask the five questions below. | Three problem hypotheses, each with a named customer group and 15 reachable people. |
| Right after class | Set up a shared Excel log (copy the template to OneDrive or Teams) and a Planner board with buckets To contact, Scheduled, Done, Synthesize. Send outreach (§4). | At least 30 messages sent and at least 3 interviews booked for tomorrow. |
| 3:00–5:00 PM | Office hours with Professor Dwyer (PH 3050G; Zoom by appointment otherwise). Bring the three hypotheses and your schedule. Ask for a reality check. | Confirmation that your top hypotheses can be interviewed this week. |
| Tonight | 10-minute check-in: count messages sent and interviews booked; assign tomorrow's slots. | A named person for every booked interview. |
Five questions for the professor today
- "Here are our three problem hypotheses. Which can we realistically interview about this week?"
- "Do interviews with classmates or students in the other section count if they fit our target group?"
- "Do our interviews need any consent wording or human-subjects check? Results stay in class."
- "Which presentation group are we in, and is the upload due before our slot or Oct 1 at 11:59 PM?"
- "If we're a few interviews short on presentation day, how would you like us to handle it?"
Asking early isn't a confession. The syllabus itself invites students to say as soon as possible if something will be late, so a plan can be agreed.
Five roles (everyone also interviews and presents)
- Hypothesis lead: keeps the hypotheses and the script current.
- Pipeline lead: owns the outreach tracker (contacted, scheduled, done).
- Log keeper: enters every interview in the Excel log the same day and checks the columns.
- Synthesis lead: runs the debrief after interview 5–6 and builds the themes.
- Presentation lead: builds the slides and runs timed rehearsals.
If someone goes quiet, use the "Consequences" ladder in your signed team contract: a private reminder, then a group discussion, then the professor.
Today's checklist saved in this browser only
3The 75-minute idea sprint
The goal isn't the best idea. It's three testable problem hypotheses about people you can reach this week. Paul Graham's test for a good source is that the problem is one the founders themselves have or see, and YC's tip is to keep a running list of problems you and the people around you run into (Graham, UCI summary of YC).
- Silent dump (10 min). Each person writes 8–10 frictions from the last 30 days on separate lines: things that took too long, cost too much, needed too many steps, or that you avoid because they're a hassle. Prompts: What did I complain about this week? What do I pay for that I dislike? What workaround do I use?
- Share and cluster (10 min). Read them out, group similar ones, and name each group's customer as specifically as you can (who exactly?).
- Filter for access (10 min). Keep only groups where you can name 15 reachable people within five days. This is the "who and where" test from The Mom Test. If you can't say where to find them, slice the group smaller until you can.
- Filter for deal-breakers and sensitivity (10 min). Drop anything a teammate ruled out in their pitch (for example an app, a physical product, or a service). Also drop topics that are sensitive to interview about (see §8).
- Score the survivors (15 min). Use the matrix below.
- Write three hypotheses (10 min). Use the template below.
- Line up interviews 1–5 (10 min). Decide who talks to whom, finalize the script (§5), and send the first messages before you leave the room.
Score the candidates
Score each 1–5 (5 is best). The weights put access first, because with one week the problem you can actually interview about beats the problem that sounds best. Tick the box to exclude a deal-breaker.
- Interview access (×3): 5 = you can name 15+ reachable people; 3 = about 8–10; 1 = you'd have to find strangers you can't locate.
- Pain (×2): 5 = people complain unprompted, at least weekly; 3 = occasional annoyance; 1 = "would be nice."
- Team fit (×2): 5 = someone lives this or has the skills; 1 = nobody cares.
- Already spend time or money (×2): 5 = people already pay or waste real time on workarounds; 1 = nobody does anything about it.
- Prototype-able (×1): 5 = a rough mockup or landing page in two weeks; 1 = heavy engineering or regulation.
On a tie, take the higher access score. The weights are my suggestion, not the professor's.
| Candidate problem (one line) | Access ×3 | Pain ×2 | Team fit ×2 | Spend ×2 | Prototype ×1 | Deal-breaker | Score /50 |
|---|
Top three scoring rows are highlighted. Your entries are saved in this browser only.
Write each hypothesis in one block
Template
Example from your own class (Class 3.3 exercise)
Problem statement: Restaurant servers (who) have too little time to thoroughly clean tables during busy shifts (what).
If… then…: If servers use a faster cleaning tool, then tables will be kept cleaner during busy shifts.
Assumptions worth testing: it happens every busy shift (frequency); servers already improvise a fix or complain about it (intensity); you can reach 15 servers this week (access); someone would change a routine or pay to fix it (willingness).
If the room goes quiet: seed problem spaces
These are prompts for your own frustrations, not recommendations. Pick problems your team actually feels. "Sensitivity" matters because some topics are harder to interview about (§8), and with one week you want the low ones.
| Space | Why it's worth an interview | Evidence | Sensitivity |
|---|---|---|---|
| Commuter parking and getting to class | Every commuting student has stories, and they're easy to find. | OU sells commuter permits (not valid overnight, 2–7 AM) and plans commuter-parking changes for fall 2026 (OU Daily, OU Parking). That shows a live topic, not proof of pain. | Low |
| Running a student organization | Officers are a large, reachable customer group (events, membership, money). | OU lists over 500 registered student organizations (OU RSO). | Low |
| Finding off-campus housing or roommates | Almost every upperclassman has done it. | OU hosts an off-campus housing listing site (offcampushousing.ou.edu), so this is a live activity. The pain itself is untested; ask. | Low |
| Balancing class with a job or family duties | Scheduling, shift swaps, and time planning are concrete, describable problems. | 50% of students say balancing academics with financial or family responsibilities stresses them (Student Voice, Aug 2025, 5,065 students; Inside Higher Ed). | Low–medium |
| Planning for surprise costs and finding student aid | Real, and people often don't know where help exists. | Same survey: 36% say an unexpected expense of $1,000 or less could threaten their enrollment, and 64% don't know whether their school offers emergency aid. Interview about how they find and plan, not about personal finances. | Medium |
| Study habits and time management | Everyone has a routine and a workaround. | A small 2025 survey (200 undergraduates) found 45% expected anxiety and stress to be their top challenge and 33% balancing school and work (Study.com). Treat it as a hint. Stick to study routines; avoid mental-health questions. | Medium |
| Small businesses near campus | Owners and managers are a distinct customer group you can walk up to. | None gathered; an untested seed from your context. | Low |
| Game-day logistics | A big, recurring local event with many moving parts. | None gathered; an untested seed from your context. | Low |
Optional: an AI brainstorming prompt allowed with a disclosure
Module 1 allows AI used responsibly with a short "Use of Artificial Intelligence" note in Canvas. AI is good for widening a list of problems. It can't tell you whether a problem is real. Only interviews can.
Prompt
4The interview sprint: 15 in about a week
Recruiting is a numbers game. A practitioner guide on customer-interview recruiting suggests contacting two to three times as many people as the interviews you need, and keeping to two or three interviews a day per person so quality holds (Wynter). Your course rules add: real-time conversations only, aim for about 15 minutes, and log each one.
4.1 Your pace
4.2 Where to find people (a "who and where" list)
| Where | Who you reach | How | Watch out for |
|---|---|---|---|
| Your own contacts | Friends of friends, roommates and floor mates, classmates from other courses, coworkers, club and intramural teammates. | Text or DM (message A). Ask each person to name two others who fit. | Close friends tend to be polite. Keep them to about a third of your 15, and still ask about specific past events. |
| Student organizations | Members of OU's 500+ registered student organizations. | Browse the RSO page or OU Engage, then message the president or an officer (message B) for two or three members. | Ask, don't spam. Use each group's own channels. |
| Online groups | Class GroupMes, Discords, Facebook groups for your segment. | Ask the admin for permission before posting, then keep the ask short. | A link to a form is not an interview. Follow each group's rules. |
| In person on campus | People at the Union, library, rec center, coffee shops. | Ask for 10–15 minutes now, or book a time. | Under 5 minutes is "questionable" under your guidelines. No 60-second intercepts. |
| Local businesses and professionals | Owners, managers, employees. | Walk in or email (message C). Offer to come to them. | Keep it to 15 minutes and expect business-hours limits. |
| Referrals from interviews | Anyone an interviewee knows who fits. | Ask at the end of every interview (message D). | This is what makes days three to five easier. |
4.3 Copy-ready outreach messages
Keep them short, say how long it takes, and say it isn't a sales call. Bracketed parts are yours to fill in. Practitioner guides recommend a human tone, an explicit time ask, and no lengthy explanations (Wynter, ProductPlan).
A. Warm text or DM
B. Student organization officer
C. Local business or professional
D. Referral ask (end of every interview)
E. Reminder, and the consent line
4.4 Checkpoints
Cumulative interviews completed by the end of each day. If you're three or more behind at any checkpoint, escalate the next morning (§6). Rows fade once the date passes.
| By end of | Boomer (Sep 30) | Sooner (Oct 2) | Also that day |
|---|---|---|---|
| Wed Sep 23 | 0 | 0 | Three hypotheses, roles, 30+ messages sent, 3+ booked for Thursday. |
| Thu Sep 24 | 4 | 3 | Log every interview the same day; adjust the script. |
| Fri Sep 25 | 8 | 6 | Debrief after interviews 5–6 and pick a lead hypothesis (§5). |
| Sat Sep 26 | 11 | 9 | Keep the referral chain going. |
| Sun Sep 27 | 14 | 11 | Draft the slides from your themes. |
| Mon Sep 28 (class work session) | 15 | 13 | Synthesize and build the deck in class. |
| Tue Sep 29 | — | 15 | Boomer: two timed rehearsals and upload. Sooner: finish interviews. |
| Wed Sep 30 | Present | — | Sooner: watch the Boomer presentations, tighten your deck. |
| Thu Oct 1 | — | — | Canvas due date 11:59 PM. Sooner: upload tonight. |
| Fri Oct 2 | — | Present |
4.5 Keep the machine running
- One log, updated the same day. Use the professor's Excel template and the same columns (interview number, date, length, interviewer, interviewee, customer or expert, interviewee information, summary, major takeaways, other notes). Notes written the next day are thinner and can cost "poorly documented" deductions.
- A Planner board (the tool the course taught) with buckets To contact, Scheduled, Done, Synthesize. Drag each person through it.
- A 10-minute check-in every night at a fixed time: messages sent, interviews done, tomorrow's names.
- Same core questions every time (three or four), so you can count patterns later ("7 of 11 said…").
- Interviews, not surveys. A survey doesn't count as an interview under your guidelines, though it can count as additional work. Don't let it replace conversations.
5Interviewing when you don't have an idea yet, and narrowing fast
You can interview before you have a solution. Start from a situation, not a product. Practitioners who teach problem interviews recommend asking people to tell the story of the last time something happened rather than asking them to list their biggest challenges, which tends to produce generic complaints (Ash Maurya). That is also the heart of The Mom Test, so it fits your course.
5.1 An exploratory script (about 15 minutes)
- Open (1–2 min). "Thanks for making time. I'm a student at OU researching how people handle [situation]. I'm not selling anything and there are no right answers. It'll take about 15 minutes. Is it okay if I take notes?"
- Their context (2 min). "Tell me a little about [their role or routine around the situation]."
- The last time (6–8 min). "Tell me about the last time you [did the activity or hit the situation]."
- What happened next?
- What made it slow, annoying or expensive?
- What did you do about it? How long did that take, and what did it cost?
- Have you tried anything else? What happened?
- What do you use now, and what don't you like about it?
- Follow the feeling (2–3 min). When something sounds frustrating or surprising, say "tell me more about that."
- Priority and spend (1–2 min). "Where does this rank among the things you're dealing with right now?" "Do you or people you know pay for anything to handle this?"
- Close (2 min). "Is there anything I should have asked?" "Who else deals with this? Could you introduce me?" "Could I follow up if I have another question?"
Avoid
- "What are your biggest challenges with [topic]?" It invites generic complaints.
- "Would you use…?" or "How much would you pay?"
- Describing your idea in the first 10 minutes.
- Counting a compliment or a hypothetical as evidence.
Optional concept check (last 2 minutes, once you have a lead hypothesis)
"One direction we're exploring is [one sentence]. Based on what you told me, what would have to be true for that to be worth trying? Would you be willing to [look at a mockup / talk again / introduce me to someone]?" Record what they agree to do, not what they say they like.
5.2 What counts as evidence
Strong signals
- A specific recent story: when, where, what they did.
- Visible emotion: annoyance, embarrassment, urgency.
- An existing workaround, or time or money already spent.
- The same problem from three or more people, independently.
- A next step: an introduction, a follow-up, a trial.
Weak signals
- Compliments ("great idea").
- Hypotheticals ("I would definitely use that").
- "Someone should build…" with no story.
- Generic claims ("I always…", "I usually…") with no example.
5.3 Narrowing from three to one
- Interviews 1–5: explore. Spread them across your two or three hypotheses. Ask the same core questions each time.
- After interview 5–6: debrief for 20 minutes. Write what you heard three or more times. Keep the hypothesis with the most strong signals as your lead concept. Note briefly why the others were weaker; the Module 1 brief asks you to explain exactly that.
- Interviews 6–15: go deeper on the lead. Ask about frequency, cost, current workarounds and next steps.
- If nothing has strong signals in at least three interviews, pivot the problem or the customer group and test the new version in your remaining interviews. Class 1.2's three decisions apply: persist, pivot or pause. A documented pivot is a good story for the presentation, not a failure. (The "three interviews" threshold is my rule of thumb.)
6If you still fall short
Plan to hit 15, and know the cost of missing. The assignment deducts 5 points for each required interview not completed, and may deduct partial points for interviews that are incomplete or poorly documented. Not submitting by the start of class is 0%, so presenting on time with an honest count is always better than missing your slot.
6.1 What a shortfall costs
| Interviews completed | Interview-rule deduction |
|---|---|
| 15 | 0 |
| 13 | −10 |
| 12 | −15 |
| 10 | −25 |
| 8 | −35 |
| 5 | −50 |
| 0 | −75 |
Out of 100, and only from that one rule. Presentation-quality and rubric-level deductions are separate.
6.2 Escalation ladder
| Trigger | Action |
|---|---|
| Today | Ask the five process questions (§2). This isn't a confession. |
| 3+ behind at any checkpoint | Email the professor the next morning with numbers and a plan (template below). Offer a 10-minute chat. |
| Mon Sep 28 work session | Bring the log so far and ask how to handle a remaining gap. |
| Day before you present | Freeze what you have. Build an honest "N of 15" slide with a catch-up plan. |
6.3 Email to the professor (send only if you trigger it)
Template
Advice on emailing instructors is consistent: tell them early, be specific and honest, propose a plan rather than an excuse, and offer to meet (College Transitions). Don't ask for an extension in the first message unless you truly need one. The professor's syllabus invites early notice.
6.4 What softens a shortfall, and what doesn't
Do
- Protect quality: fewer, well-documented interviews beat many thin ones.
- Submit on time and present, whatever the count.
- Make your Outstanding Questions and catch-up plan specific.
- Report true counts ("7 of 11") everywhere.
Never
- Invent or embellish an interview, or count the same person twice.
- Count a survey or a one-minute chat as an interview.
- Pad the log with people who aren't potential customers.
- Skip your slot.
7Presenting when the evidence is thin
The brief rewards honest learning. Use the same five sections, keep the numbers true, and lean on what changed in your thinking. Everyone still needs a speaking role, and the talk should run 8–10 minutes.
| Brief section | If evidence is thin | Sample wording |
|---|---|---|
| 1. Problem and initial hypotheses | Show the dated hypotheses you wrote on Sep 23. Being early-stage is the point of this section. | "On September 23 we hypothesized that [customer group] struggle with [problem] because…" |
| 2. Proposed solution | Describe a direction, not a finished product. Name the value proposition and the possible revenue streams. | "Our best current direction is… It would be faster or cheaper than [today's workaround] because…" |
| 3. Interview insights | Use your true count. Give two or three stories and at least three patterns with counts. | "In 11 interviews, 7 people described… and 4 told us…" |
| 4. Industry context | Keep it high-level with 3–5 sourced facts (the free sources are in §6.6 of your earlier course-analysis report). | "This is a [type] industry dominated by…" |
| 5. Moving forward | Give an honest confidence score, mark each hypothesis confirmed, partially confirmed or rejected, and list three unknowns with how you'll answer them. If you pivoted, say so. | "We're at 5 out of 10. What raised it: … What holds it back: … Our next three questions are…" |
Make two things obvious: what changed between your Sep 23 hypothesis and your conclusion, and that your numbers match your log. Put the interview count on a slide.
8Guardrails
Real conversations only
The syllabus lists penalties from a zero on the assignment up to expulsion for serious integrity offenses, and your presentation includes Q&A. Each interview in the log carries a name and details, so every one has to be real.
AI: what's fair game
Module 1 allows AI used responsibly, with a brief "Use of Artificial Intelligence" statement. Good uses: widening the list of problems, drafting outreach messages and the script, turning your own notes into themes, and formatting slides. Not fine: generating interview content or "findings." Say what you used and how, and check facts yourself.
Consent and privacy
Say what the conversation is for, ask before taking notes or recording, and keep interviewee names inside the class. The log needs names for the professor; don't share it or any recordings publicly.
Human-subjects (IRB) check
Universities draw the line differently. One example, SIU, treats class-assignment interviews as not needing IRB review only when the project is for skill development, has under 25 participants, avoids protected groups and sensitive topics, keeps results in class, offers no incentives, and destroys data afterward (SIU guidance). I couldn't find OU's own rule, so ask the professor today (question 3), avoid minors, and don't offer cash incentives.
Sensitive topics
That same guidance lists mental health, sexual practices, substance use, illegal behavior, and financial or employment risk as sensitive. With one week and no time for approvals, pick a low-sensitivity problem (see the seed table) and steer conversations away from health and personal finances.
Outreach etiquette
Don't pressure anyone, don't spam group chats or organizations, and follow each group's and platform's rules. A "no" or no reply is an answer; move to the next person and keep your 2–3× buffer.
9Sources and limits
From your folder
Module 1 brief and Interview Guidelines (interview rules, deductions, log columns), Syllabus v1.2 (late-work invitation, integrity and AI policy, team work sessions, office hours), Class 1.2 (persist, pivot or pause), Class 3.3 (problem statement and "If… then…" exercise), and the Team Contract Template (Consequences section).
From the web
Finding an idea
- Paul Graham: How to Get Startup Ideas
- Y Combinator Startup Library: How to get startup ideas (the page loaded without its article text for me, so I used the UCI summary below)
- UCI ANTrepreneur Center: How startup ideas really form (lessons from YC)
Customer discovery and recruiting
- Ash Maurya: the 30-minute customer discovery
- Summary of The Mom Test (customer slicing, warm introductions)
- Illinois I-Corps: National I-Corps (15+ interviews a week)
- Wynter: recruiting customers for interviews
- ProductPlan: customer interview email templates
Problem-space evidence
- Inside Higher Ed: Student Voice survey on surprise costs
- Study.com: college students' 2025 predictions (small survey)
- OU Daily: OU parking changes · OU Parking: park on campus
- OU Registered Student Organizations
OU help and process
- Tom Love Innovation Hub / Startup OU
- College Transitions: emailing a professor
- SIU: IRB guidance for class projects