The Data Entrepreneurs Story — Part 1: Finding a Case

The first step is to get into the conversations. Photo by Romain V on Unsplash.

“Who is an entrepreneur? Someone who jumps off a cliff and builds a plane on the way down.”

Reid Hoffman — Founder of LinkedIn

In the era of fast-changing markets and constant technological disruptions, those who do not innovate will become extinct. Innovation has been a central theme and at the heart of it are the entrepreneurs. Brave minds who dare not only to dream of a better world, but also willing to go face-to-face with the challenges. Among those who are foolish enough to embark on the journey, are three students united under the banner of Jheronimus Academy of Data Science (JADS) paving their way to become successful entrepreneurs.

Even though we have very limited experience as entrepreneurs, that does not stop us from learning. As part of our Master’s program of Data Science and Entrepreneurship, we are given the chance to experience what it is like to found a company. This series of blog posts convey a story of our experience and what we learn along the way: what are the steps that we take, how do we build our solution, and what we think can be improved in the future. First, we need a case to work on.

Part 1 — “What are your frustrations?”

A lot of start-ups failed not because of the technology itself, but rather because no one cares about their product. That is why we decided to start by asking: what are the things that other people care about?

In the first week of our journey, we try to gather as many concerns from our own circle. We listen to our friends as they describe their daily routines and frustrations. In total, we conducted five interviews and published one online survey. These are the main questions that we addressed:

1. What is your daily frustration?
2. What are the things that you believe is inefficient, too expensive, or non-existent?
3. Is there any nonsense that you notice lately?

The doors that we unlock by those three simple questions lead us to many interesting and eye-opening stories. Despite coming from different countries and different cultures, the three of us have the same background academically: computer science. Through the interviews and the survey, we got a glimpse of what are the typical challenges from a range of professions. We try to delve into their daily routines to find the nitty-gritty of their job. The background of the interviewees was quite diverse:

1. A head pharmacist in a clinic.
2. A supply chain manager.
3. A UI designer of a technology start-up company.
4. A sales and service engineer of an industrial instruments company.
5. A recruiter of a consultancy company.
6. A finance analyst.

Finally, we are able to extract some problems that we believe are worth getting a second look. During our first follow-up meeting, we come up with a compilation of the results:

1. Messy drug inventory management in hospitals.
2. Lack of traceability for doctor’s prescription.
3. No platform that specifically serves medical tourism demand.
4. Expensive time and money cost of handling an emergency shipment in a factory.
5. Unreadable documentation and hard to transfer knowledge on how to use a (legacy) complex applications.
6. Unstructured brainstorming sessions for mobile application development.
7. Slow response when communicating with local authorities/government.
8. Hard to find participants for application feedback and user testing.
9. Unclear agenda for meetings.
10. Lack of accountability for decision making process.
11. Distraction during work.
12. Feeling unsafe when walking alone in the night — yes, we also take note of some random thoughts like this.
13. Sitting down for a long time while working.
14. Manual verification of invoices.
15. Slow approval process of higher-ups (the manager level and above).
16. Cumbersome data entry for expense claims.
17. Working depression and loneliness.
18. Security issues in smart devices, in terms of accessibility (unauthorized access) and physical threat (intentional damage, got stolen, lost).

Not a bad list after just five interviews and a survey, right?

One problem at a time

The list contains many interesting problems, but if we try to solve many problems at once, we might as well solve none. We have to prioritize. What are the criteria that we can consider to choose which problem we shall focus on?

There are three criteria that we establish in order to find the most promising opportunities:

1. We do not want to ignore the “data” part of the “data entrepreneurship”. After all, we want to learn both data science and entrepreneurship. So, one of our considerations is if there is any dataset available as a resource to propose a data science solution?
2. JADS can provide us with extra support if we decide to work on some existing thematic areas: crime and safety, agrofood, smart cities, energy, smart industry, and health. Choosing a problem related to those areas will be preferred.
3. Since we will face challenges and the work is nowhere easy, it is better to work on a problem that we, personally, are also interested in.

Out of the 18 opportunities, we make a short-list of what we will follow-up:

1. Messy drug inventory management in hospitals.
2. Complex documentation and hard to transfer knowledge on how to use a corporate applications.
3. Lack of accountability for decision making process.
4. Cumbersome data entry for expense claims.

We believe those problems are prominent among many companies (and hospitals). In terms of thematic areas, it is clear that the first one is related to the health sector, but that is not the case for the rest. Given the limited time to the first presentation that we have to prepare, we decided to focus on the first problem for now. We keep the other three problems to discuss further next week.

In detail, the problem that we have chosen is primarily related to the inventory management of drugs in a hospital. Which and how much drugs should the hospital order for next month? How to make sure the number of expired drugs is minimum while still taking into account emergency cases? How to distribute the drugs more efficiently in a hospital and from the supplier?

Quantity over Quality

With a real problem in our hands, we proceeded with the ideation step. We decided to use diverging-converging thinking method. As the name suggests, we start with divergent thinking, where each of us should come up with as many potential solutions as possible in a span of 10 minutes. This method forces us to think come up with additional ideas besides obvious ones. It is also creating a safe environment where every member can come up with any ideas without getting criticized. The goal is quantity, not quality.

A typical scene of a brainstorming session: murder of post-its.

After the brainstorming session, we organize all ideas that we have written into 10 big themes. The ideas include the obvious ones to the not-so-obvious-rather-crazy ones. We will not list all our ideas here because the solutions themselves are still lacking a sufficient degree of detail and this is already a long post anyway. In the end, we agree to promote three themes as our current proposal:

1. Predicting which drugs are going to be ordered soon. We can use previous transaction history, preferences of doctors or patients, and drug expiration time.
2. A recommendation system about drug ordering, based on previous similar diagnosis, typical drugs bought together, or current drugs availability.
3. A managed drug inventory service for multiple hospitals, so each hospital does not need to maintain the supply-chain themselves.

We realize that some of the ideas might not be feasible due to resource limitation or legal concerns, but we never know. After all, we cannot be a good entrepreneur by being risk averse. Maybe the limitations are just our assumption, considering how minor our understanding regarding how things work internally in the hospitals is. That is why our plan for the upcoming week is to learn more about the subject and verify our assumption to the experts.

What do we learn this week?

A venture should start with customer’s problems in mind. A simple yet effective way to find out what kind of problems people have to deal with everyday is, well, talking to them. We can even start from our own circle: our friends and family. It is also important to empathize during the conversation. Try to put ourselves in their shoes, how would we feel? What would we have done differently? We made sure to leave our assumptions behind for the initial interview. There are a lot of things that we are not aware of about in each profession. Let them tell us what is truly going on and ask as many questions until we really understand.

Since we also published a survey, we can compare the result we get from the interviews and the survey. The survey was not as effective to gather insights. The answers tend to be short and we are unable to follow-up. Survey should be better if we already know exactly what information we want to get. In the future, when the problem is better defined, we can revisit the survey approach to further validate or improve our solution.

Lastly, we also had the opportunity to use some of the previously learned ideation techniques into practice using the data from the interviews and survey, which led us to derive at the four potential feasible solutions that were discussed.

Next Step

We are going to present our ideas to our fellow students and our mentors to hear their opinion and get some feedback. Afterward, we are planning to do in-depth research and conduct further interviews with practitioners that understand how drug inventory works to separate feasible ideas from the far-fetched. Our priorities for the second week are:

1. Find experts in the health industry, especially related to drug inventory management, to learn more about the details of how it operates currently and gather their perspective pertaining to the problem we have identified.
2. Make a list of our current assumptions that we should verify with the experts.
3. Decide the goal and prepare a set of questions for the interview.

About Us

Hi! This post is made possible by a collaboration of Yosef Winatmoko, Nemania Borovits, and Hameez Ariz. We are master students of Jheronimus Academy of Data Science (JADS) currently taking Data Entrepreneurship in Action III course. In the following weeks, we plan to continue to publish our weekly journal about what we go through and what we learn. Hopefully, you can also learn something from our story.

Update: read more on what we did next in

Part 2 (Secondary research on hospital pharmacy)

Part 3 (The problem is out there)