Every project starts with the solution. Someone has an idea, thinks it is good, gathers people and starts building. The problem, which should have come first, gets left for later, and sometimes for never. This article is about reversing that order.In this article, I show you how to separate a real pain point from one you have merely inferred, how to find out who will actually use, decide on and pay for your project, and a five-question script that works in twenty minutes of conversation. Along the way, I use two cases: an app for condominium buildings that missed the mark and the reason Lecursos.com exists in the first place.Seven months focused on the wrong problemA four-person team spent seven months building an app for condominium buildings. The function room booking screen was flawless: a colorful calendar, notification confirmations. At launch, two buildings installed it. Within three months, usage had fallen to almost zero.A late conversation with a building manager revealed the information that would have changed everything. Booking the function room was never a problem, because it happens four times a year and the caretaker writes it down in a notebook. What consumed that manager's evenings was collecting overdue payments without making enemies in the elevator.Seven months of competent work solved a quarterly inconvenience while leaving a weekly pain point untouched. This is the most expensive mistake a project can make, and it almost never looks like a mistake: it looks like productivity. The team was busy, delivering and refining.Why does this happen so often, even to experienced people? Because solutions are fun and problems are boring. And for a less acknowledged reason: until the problem is defined, anything you do looks like progress. Once it is defined, there is a criterion, and that criterion may condemn the idea you already wanted to build.Real pain and imagined painURL_DA_CAPA_NO_CARDS_REALM{Imagined pain has an abstract verb and an undefined subject. Real pain has a person, frequency, cost and a workaround.}Imagined pain is what you have inferred. It appears with an abstract verb and an undefined subject: "people struggle to manage their finances," "students are not engaged," "customers want more convenience." Statements like these are comfortable precisely because no specific fact can challenge them.Real pain has four things that imagined pain does not:- A specific person, with a name or a clearly defined role.- A frequency: every Tuesday, or every time the month closes.- A cost that can be estimated in hours, reais or stress.- A workaround: something the person already does to get around the problem, even if it works poorly.The workaround is the most reliable of the four signs. No one builds a workaround for a problem that does not bother them. When the building manager has been writing bookings in a notebook for years, they are telling you it does not hurt. When they send individual messages to four delinquent residents every week, rewriting each one so as not to sound aggressive, they are shouting where it hurts. The information was available all along. All that was missing was the question.One note on intensity: major but rare pains do not change anyone's behavior, because people solve them in the moment and forget about them. Small, constant pains build habits, and habit is what makes someone adopt a new solution. Better projects tend to emerge from the latter.Who feels it, who decides and who paysOne quiet mistake is treating three roles that are almost always different as if they belonged to the same person. A school management system is used by the administrative office, chosen by the school leadership and paid for by the school owner, and each assesses it according to their own criteria. At an event, the same separation has different names: attendees participate, coordinators authorize it and sponsors fund it.A project that convinces only the audience never gets off the ground. A project that convinces only the sponsor takes place in front of an empty room. The practical question is short: who feels it, who decides, who pays? If they are three different people, the project needs three different arguments.In your current project, are the person who feels the problem, the person who decides and the person who pays the same person?The questions that are worthless"Would you use an app like this?" "Would you attend an event about this?" "Would you pay for this service?" All of them ask someone to predict their own behavior in a hypothetical scenario. Human beings are terrible at that and polite by default. The answer is almost always yes, and that yes is worthless.The questions that matter look to the past, which has already happened and cannot be invented in the middle of a conversation:- How do you do this today? It forces them to describe the real process, including the patches.- When was the last time this got in your way? If the person cannot remember, the frequency is low and the pain is minor.- What have you already tried to solve it? It reveals whether effort, time or money was invested before you showed up.- How much does this cost you per month? In hours or reais; the estimate can be rough.- Who else in your situation struggles with this? It opens the door to the next interview and tests whether the problem belongs to one person or a group.Notice that none of them mentions your idea. In a good discovery interview, the interviewee talks eighty percent of the time and you present the solution only at the end. Presenting it early contaminates everything: from that point on, the person comments on your proposal instead of describing their life.The same conversation, done two waysIn the version that does not work, the very first sentence gives away the app: "We are developing an app for building managers. Do you think it would be useful?" The manager thinks it is great, would use the function room booking feature and would pay thirty reais if it is good. Three positive answers, no information. The team leaves thinking it has validated the idea.In the version that works, the first question is, "Tell me about your last week as a building manager. What took up the most time?" And the answer contains everything that matters:“I send private messages, one by one. I write and delete them about three times so I do not sound rude. It takes an hour, easily. Every month. And when I run into the person in the elevator, there is that awkward atmosphere. I tried asking the management company, but they only send the bill. The embarrassment is still mine.”- A building manager, when no one had mentioned the appMonthly frequency, a one-hour cost, a described workaround and a previous attempt that failed for a specific reason. None of this would have come up if the first question had mentioned the app.Mistakes that ruin an interview- Talk too much. If you spoke more than the interviewee, the conversation was about you.- Present the solution early. The person becomes a critic of the proposal, and polite critics praise it.- Ask two things at once. They answer the second, and the first gets lost.- Lead the answer. "Isn't it true that this gets in your way a lot?" already gives away what you want to hear.- Write down only the conclusion. Record the exact words, in quotation marks. You can reconstruct the conclusion later; you cannot reconstruct the words.- Interview only people who like you. Friends protect you, and that protection becomes noise.How many conversations are enoughFive is usually enough for the first round, as long as they are five people who truly have the problem and are not your friends. The signal to stop is not the number, but repetition: when the third, fourth and fifth people describe the same outline of a problem in different words, you have found something. When each describes a different problem, you do not yet have an audience; you have five isolated cases.Finding these people is the most feared and simplest part. Ask for a referral at the end of each conversation, go where they already gather and use the network of whoever is already involved in the project. Twenty minutes is a small ask.After each conversation, set aside five minutes to write down what was said in two columns: reported facts and your interpretations. It may seem pedantic, but it is the only thing that keeps your interpretation from appearing in the presentation three weeks later as though it were the interviewee's statement.What happens when the class goes out to askIn the Entrepreneurship and Practice course, the seventh week is devoted to a task that generates resistance: interviewing students from other terms before finalizing the project format. The class has already chosen a topic, divided the tasks and gotten excited. Stopping feels like a waste of time.The result changes the project more often than students expect. One class planning a lunchtime lecture series discovered that lunch was exactly the time when most people were working or interning. The topic was right, the format was wrong, and the information cost a week of conversations instead of an empty auditorium.There is a second pattern. Students rarely attend an event because of its advertised topic. They attend because a professor asked them to, because a friend is going, because it counts as an extracurricular activity or because they want to meet the guest. Finding out which of these drivers is available changes the entire outreach strategy.And there is an effect on the class itself. Those who conduct interviews begin defending the project with different language. Out goes "I think people will like it," and in comes, "Of the twelve people we spoke to, nine said they cannot come in the morning." The second statement survives a debate. The first does not.The problem was not the lack of a websiteUntil 2018, I distributed summaries of my classes as PDFs through Google Drive. It worked. From the outside, there did not seem to be any problem at all, which is why this case is useful.The problem was invisible from behind my computer. When I corrected an error in a summary, the correction did not reach those who had already downloaded it. Old versions kept circulating, forwarded between classes, printed and studied by students who would never know they were reading an incorrect sentence I had fixed months earlier.Pay attention to the shape of this problem, because it belongs to an entire family. The pain was not mine when it happened; it was the student's, and they did not know they were feeling it. Problems like this do not show up in complaints or satisfaction surveys. They appear when someone compares what should have happened with what actually happened.If I had started with the solution, I would have renamed files with version numbers, organized the folder better or sent an email with every change. All of that solves the symptom, which is file clutter. None of it solves the problem, which is that corrections do not reach the people who took the material with them. Framed that way, the requirement became narrow: the content needs to live in one place, where the latest version is the only version. That is why Lecursos.com was born, not because I wanted to have a website.The irony of the outcome: the hosting service I hired did not work well in Brazil, and for quite some time the website simply would not open for some students. The requirement I had gotten right on paper was exactly what failed in practice, and it took me a long time to see it because I only received feedback from those who could access it. Defining the problem well does not guarantee that execution will honor it. It only ensures that, when something goes wrong, you will know what you were trying to do.An exercise to wrap upSchedule five twenty-minute conversations with people who have your project's problem and are not your friends. Use this article's five-question script without presenting your solution until the end.After each one, fill in four columns: who (name or role), what they do today (the workaround), frequency and cost, and one exact quote in quotation marks. Once you have completed all five, write down only what appeared in at least three of them.If none of the five people said anything that contradicted your plan, you probably have not really interviewed anyone yet. And if someone did, congratulations: that sentence is worth more than the condominium app's seven months.
反応
0
0 コメント
最初にコメントする