How do we identify and define a compelling problem to solve?

1. Your "PROBLEM" checklist

We've achieved market fit: 

  • We have a number of engaged users of my MVP 

  • We have enough users that we are sure that our service/product creates value for a specific sub-market

  • Our users are repeatedly engaging with my product and are satisfied  

  • We have a revenue model in mind, with some evidence of its effectiveness

We've achieved product-founder fit: 

  • We are still excited by the problem we are solving

  • We enjoy the day to day work 

  • Our skills are well suited to the final product/services we are working on

  • We have access to networks, advisors and colleagues that are suited to help us with this idea

My organization is resourced to scale:

  • We have sufficient funds to manage working capital needs

  • We have the right board and advisors to help with scaling


2. How do we find a problem to solve?

Find a problem by first finding someone specific you are passionate about helping.


Example:  “I would love to help fellow first-generation college students because I was a first-gen college student myself.”

Then, interview this person to understand what this person is desperate to solve in their life.


Example: “As a high school teacher, it takes too much time and effort to prepare for parent-teacher conferences.”


The most successful education innovators find someone they care about and something this person is desperate to solve. If you aren’t helping someone whose life you want to improve, you’re unlikely to continue caring when the going gets tough.  It's easier to start with someone - or something - you already know. 

A Fellow shares his experience:

In my case, I am passionate about helping first-gen college students secure their first jobs because I personally struggled with the job search as a first-gen college student with little parental guidance. Once you’ve identified a specific person, talk to them face-to-face for an hour to understand what struggles they face on a typical day. By the time you speak to ~10 people, you’ll start seeing patterns in what these people tend to complain about.


A common mistake many entrepreneurs make is to identify an opportunity or medium instead of a problem. For example, there are many AR/VR ed-tech ideas fail because the founders were so caught up with using AR/VR that they didn’t think at all about what problem they were solving. AR/VR is but a tool—it’s not a problem. Using surveys too soon is another common mistake that I’ve personally made. At the early stages, you want to listen to people speak in freeform since they’re likely to tell you things you’ll never learn through a narrowly-scoped multiple-choice survey.

Questions to ask yourself:

  • Whom do we care about helping?

  • What pain is this person desperately trying to eliminate in their life?


3. How do we best define and focus our problem?

A well-defined problem has a specific quantifiable pain that a particular person feels like a massive headache. A poorly-defined problem has a broad scope and is aimed at a wide audience who might not even notice if the issue is fixed.

A Fellow shares his experience:

"I generally know when a problem is poorly-defined when I feel like I’m hearing the “Voice of God”:  “Students need to be equipped with 21st-century skills!” or “Teachers need more ways to collaborate!” I say “Voice of God” because only some higher power can proclaim that people “need” something. No student wakes up in the morning thinking, “Gee, I could really use some 21st-century skills today!” Similarly, no teacher loses sleep over a lack of ways to collaborate. Students do, however, think, “I have no idea how to do this math problem,” and teachers think, “I am so overwhelmed with the amount of grading I need to do that I hardly have time to cook dinner.” Well-defined problems require painkillers—if you don’t pop that aspirin, you’re bound to feel terrible. Poorly-defined problems, on the other hand, are like popping vitamins—you might be better off if you take your vitamin C, but you’re unlikely to notice if you don’t.


As someone who loves to read industry reports and news articles, I personally found myself often defining my problems broadly, with statements like, “we need to close the ‘skills gap’” and “college students don’t find career services helpful.” In retrospect, this was a huge waste of my time because even though these are big societal issues, they aren’t specific pains that can be turned into solutions a particular person would find compelling. It wasn’t until I started focusing on specific pains in the job search such as “students can’t access career services after hours” or “students don’t know what they want” that I started making more progress. Yes, you should absolutely be passionate about the macro problem, whether it’s the skills gap or economic inequality. However, you also need to be comfortable with the fact that magic wands don’t exist—and that it simply isn’t possible to expect your solution to solve all of life’s problems on day one. You may get to it eventually, but start with a specific pain point first."

Questions to ask yourself:

  • Does our problem affect a particular person?

  • Does our problem cause a massive headache for this person?

  • Will this person feel immediate pain relief if the problem is solved—or will they not even notice that anything happened?


4. How do we know if we are the right people to solve this problem?

If you are a founder, you should (a) be passionate about the problem, (b) be willing learn enough about the problem to have a better understanding about what’s going on than anyone else, (c) offer at least one skill to the table that is critical to solving the problem.


If you are a joiner, you (a) be passionate about the problem, and (b) should bring at least one skill to the table that is critical to solving the problem.


If you are a founder: You don’t need to have dropped out of school to help schools solve its student retention problem, but you should speak to every stakeholder who is involved with student retention so that you understand the problem better than everyone else—including the very people who surround the problem. Only with superior understanding can you come up with a different idea that can withstand the test of competitors and incumbents. Assuming you’ve already identified a problem to solve, you’ll need to list out all the stakeholders who touch the problem and then interview them. You don’t need a Ph.D. on the topic, but you should have an understanding + a skill that is useful for solving the problem (so continue to read the joiner section).   


If you are a joiner: If you are not as familiar with student retention as the founder(s) are, you should (a) be interested in the problem, and (b) possess a skill that can help solve the problem. Skills can be hard skills (e.g. coding, data analysis, graphic design, video production) or soft skills (sales, interviewing, partnership development). You don’t need to be the world’s best front-end web developer to contribute, but you should bring a skill that is necessary given where the venture is today. For example, you might be an expert at machine learning, but if the team is still trying to understand the problem, then you should resist the urge to start coding just yet. Put your machine learning skills off to the side for now and think about what other skills you possess that are more relevant for the team’s needs today (for example, your interviewing skills or cold emailing skills).

You don’t need to be the world’s foremost expert on everything related to your problem, but you should be an expert in at least one piece of the puzzle to know how to execute at the early stages. Don’t be discouraged by the fact that you are not an expert. This can actually be a strength! It’s often outsiders who bring a fresh perspective. YOU can be that person! In the end, the most important factor is your personal passion. As we discuss in “How do I find a problem to solve?” the emotional ups and downs can be challenging. The only way you’ll persist when the going gets tough is if you can personally relate to the problem.


As it relates to content experience, a well-rounded team will have both content experience that comes from both personal and professional experiences.  This allows you to understand the problem from a potential user’s perspective while also having an understanding of how a business approaches the problem.

Questions to ask yourself:

If you are a founder:​

  • ​What do I know better than anyone else? 

  • What’s one skill I have that can help solve this problem? 

  • Am I passionate about this problem?


5. How do we validate whether a problem actually exists?

Take two steps:


1) See how others are describing the problem online and note any patterns you see 

Google any terms related to your topic and open all the relevant links you see. Pay attention to forum threads (especially Quora and Reddit if you’re investigating a younger audience), reports, blogs, op-eds, YouTube videos, and Google’s search term suggestions. Make a list of (a) key terms people use to describe the problem, (b) existing solutions, and (c) any patterns you see.

Every time you stumble upon a new key term, repeat your Google search with that term. This will help you broaden your search to corners of the Internet you may not have had the vocabulary to search for.


Every time you stumble upon an existing solution, Google for that solution with two variations:

  • [The name of the solution you found] + “reviews” (and ask yourself, “What aren’t they doing well?”)

  • [The name of the solution you found] + “competitors” / “alternatives” (and ask yourself, “What are they not doing?”)


E.g. “college student job search,” “how to find a job,” “career services suck,” and “career services unhelpful” (without the quotes).


2) Interview people who feel the pain, then have them explain the problem back to you in their own words 


Find and sit down with the target person you’re trying to serve for an hour and consider asking them the following questions:

  • “Tell me about a typical day [related to whatever topic you’re interested in investigating].”

  • “What are some of the most frustrating things about doing [whatever you’re investigating]?”

  • “Have you ever come across [whatever solution you found in your Google search]? What are your impressions?”

  • “How would you describe the problem we’re discussing?”


The goal is to ask open-ended questions to let your interviewee reveal their real thoughts to you. Withhold all judgments and solutions for now; asking leading questions like “Wouldn’t it be great if we had [your solution] will only lead to biased answers.

Questions to ask yourself:

  • Can we find people concerned about the problem?

  • Can we get a group of people to articulate the problem back to us in their own words?


6. How do we know if people care this problem?

People care enough about your problem if it is among the major pain points to getting something done.


Example: Within the context of applying to colleges, it is a major pain to (1) know which colleges and programs are right for you, (2) compare colleges based on how much it will cost, and (3) know what your chances are of gaining admission. In comparison to these 3 pains, scheduling a college tour itinerary is a relatively minor pain.


People are busy. If the pain point you are solving is #13 on someone’s list of pain points, it’s unlikely your solution will ever be used; your user will be too busy dealing with pain points #1-3 and won’t have time for pain point #4, let alone pain point #13. 


A Fellow shares his experience:

"To determine where the problem  want to solve sits on my target user’s priority list, I first ask my target user, “What are the biggest frustrations when it comes to [doing X]?” Then, I let them rant. If you don’t hear the problem you are trying to solve among this person’s top list, this is likely a sign that you’ve identified a relatively minor pain that doesn’t actually matter. Of course, different people will find different parts of a certain task or difficult, so it’s important to talk to more than just one person. I’ve personally found that talking to ~6 people is enough for me to start observing patterns. First, I let each person rant independently. Then, I write out each pain point I hear on index cards and then have a different set of people sort them from “most painful” to “least painful.” Typically, what emerges on top will be the biggest, most commonly-felt pain point.


Don’t be discouraged if your interviewees end up saying something along the lines of “Oh, everything’s fine.” From what I’ve seen, people generally have a difficult time articulating their pain points. I suspect it’s because once we encounter the same pothole multiple times, we adapt and end up subconsciously driving around the pothole without ever giving it a second thought. If this happens to you, ask, “Walk me through how you go about doing X.” You might find yourself being able to spot a pain point your interviewee has already gotten used to.


From what I’ve seen, entrepreneurs make two mistakes at this stage: (1) They are so married to a particular problem that they completely ignore what their interviewees say. So, instead of listening to the other person’s top pain points, they end up listening only for whether the other person mentions their specific pain point. (2) They are so married to a particular solution that they completely ignore whether their solution is even solving a big pain point. Remember: the purpose of speaking to people is to make sure you identify the biggest problem. Only with the biggest problem can you build a solution people will actively seek out. You’d rather get the harsh feedback now than later, after having investing months building a solution no one cares about."

Questions to ask yourself:

  • Is our problem among the top 3 pains / frustrations to accomplishing something?


7. This space feels crowded. Do we need a new idea?

Try each solution yourself to understand where each solutions strengths and weaknesses. As you do so, ask yourself, (1) “What pain point(s) is/are each incumbent player addressing, specifically?” (2) “What pain point(s) is/are each incumbent player overlooking, specifically?” and “In what specific ways are incumbent players executing poorly?”


Example: All existing grading tools are focused on helping teachers grade assignments on the computer. No existing solutions help teachers actually do their grading more quickly / efficiently. Existing grading tools also do not permit teachers to copy and paste the same comment between students and require that teachers grade only when they are connected to the web.


Just because there are a lot of players in a given space doesn’t mean that every problem has been solved—and it definitely doesn’t mean that incumbents are executing their solution correctly. Often, a given problem space only looks crowded to outside observers who haven’t tried any of the solutions and who aren’t familiar with user’s pain points. You may have heard that Google beat Yahoo and a slew of other search engines despite being a new entrant into a crowded market. It happens all the time. The key is to identify the missing gaps left behind by the incumbents and execute better where others have fallen short.


To understand where the remaining gaps are, put yourself in the shoes of a user and/or buyer and give each solution a try.

  • What pain point is it solving—and not solving (and perhaps even creating)?

  • What is it focusing on—and not focusing on?

  • In what ways is it easy to use the product or service—and in what ways is it difficult to use?


Then, create a table. Across the top, list out each solution. Along the side, turn each strength into a row. All the gaps are spaces where a potential new solution can do a better job. You won’t need to do everything better—but whatever idea you end up creating should be meaningfully better on the dimensions that matter such that existing buyers/users are willing to switch over.

Questions to ask yourself:

  • What pain point is each incumbent player addressing, specifically?

  • What pain point is each incumbent player overlooking?


8. How do we find stakeholders who care about this problem?

Make a list: (1) Who feels the pain? (2) Who’s reputation is on the line? (3) Who loses money? (4) Who is already trying to solve this problem? (5) Who has to deal with unnecessary stress? (6) Who’s mission involves having this problem solved?


Example: Every time a kid shows up to class without their homework completed, the teacher can’t move ahead with their teaching¹ and will later have to explain to the administration why this kid is falling behind.⁵ The more a student falls behind, the more the school’s reputation (which falls on the principal) takes a hit.² Every time a student drops out, the school district loses funding from the state.³ Organizations such as the Carnegie Foundation⁶ are investing to increase student graduation rates. The student¹ himself or herself might eventually regret their decision to drop out of school, but probably only at a later stage of life. While some parents⁵ might be worried, some might also be too busy to care, while others might not see the full picture at home.


To build this list of stakeholders, it’s helpful to Google for the problem (“homework completion”/“the importance of homework”/“teacher homework”—try any combination of keywords) and to skim every website on the first page of hits. This is where it makes sense to have as macro a view as possible and to imagine your problem as a big board game with many little pieces. This will save you time in the long-run vs. cherry-picking a single stakeholder, going too deep, and needing to backtrack later after realizing that they really don’t care about your problem. Surprisingly, it actually isn’t all that helpful to talk to people at this stage. Our guess is that it’s difficult for people within the system to see the bigger picture.


Once you list out the all the stakeholders touched by your problem, next you want to rank them by order of how much they care. Once you’ve identified the person who cares the most about your problem, talk to them and understand their daily life. What are their top 3/5/10 biggest daily headaches? Where does your problem sit on their headache list? The goal is to (1) align yourself with someone who cares the most, and (2) make sure that the problem you’ve identified is one of their top headaches.


Often, we find entrepreneurs struggling because they try to tailor/sell their solution to a stakeholder who doesn’t care enough. this is perhaps because (A) they found the wrong person, and/or (B) they found a problem that is only a minor issue and that isn’t actually a big enough pain for people to change their behavior around. For example, I’ve seen many entrepreneurs try to build better ways to give teachers feedback, only to fail because many of the teachers they approached didn’t think they needed feedback and the school districts they were selling to didn’t feel like their feedback process needed improvement.

Questions to ask yourself:

  • Have we identified someone?

  • Who feels the pain? 

  • Who’s reputation is on the line? 

  • Who loses money? 

  • Who is already trying to solve this problem? 

  • Who has to deal with unnecessary stress? 

  • Who’s mission involves having this problem solved?


Resources at Harvard

  1. Speak with specific Harvard administrators who feel the problem you are trying to solve
    If you are solving a problem that touches a Harvard department (e.g. Alumni Relations, Career Services, Library), look no further than our own staff! 

  2. Book office hours with Entrepreneurs in Residence at the Harvard Innovation Lab
    If you have a legal, product, marketing, or really any other startup-related questions, book time with an expert here.

  3. Book office hours with Harvard professors who are experts in your industry
    If you search for your topic (e.g. “parental involvement”) + “Harvard Professor,” to find a related faculty member. Professors can help you understand the macro picture and what research has to say about the problem. If you know of someone who’s taken a given faculty member’s course, ask for for an intro to increase your chances of getting a response.

  4. Reach out to Harvard alumni who have worked in your industry
    Google for your school's “alumni network” and you’ll find a link to your school’s alumni database. Search for alums who’ve previously worked for competitors you’ve surfaced and reach out to ask for quick chat about their experiences. 

  5. Find Harvard students who have worked in your industry before
    Go on LinkedIn or Startuptree and search! Contact them and offer to meet up to chat about their experiences.

  6. Join student clubs that are related to your industry and be on the lookout for events with industry leader
    Google your school (e.g. “Harvard College”) + “student organizations” and browse the list. 

Know of more resources?