【NTU DAC】Custermer Retention Study for Shared Parking Space Provider
In retrospect, I am thankful for my decision to join NTU DAC and engage in the project despite the fact that the project was highly related to my increasing weight and full schedule in 109–1 semester. For me, the most surprising thing is how poles apart I feel at the end of the project, given the countless problems we encountered during the project. As I put more and more efforts, the initial impatience while doing tedious works unwittingly turned into the determination of accomplishing the goal. These ups and downs made it a memorable journey.
Facing the Challenge
Before everything else, I would like to briefly introduce our team and the problem we had to solve. This project is a three-month cooperation with a shared parking space provider. Every two weeks, we reported our progress and latest discoveries to proprietor. The ultimate goal was to find methods to boost customer retention rate.
We decided to construct the business problem solving framework first. Because most of our team members came from college of management and commerce, we expected it to be a relatively easy task. The reality is just the opposite. On our first meeting, we spent four hours just discussing what “retention rate” meant and how to represent it as measurable indicators. This is beyond my imagination.
Our succeeding meetings were not very different. To ensure logical coherence and data availability, and to kept everyone on track, we argued over every minor dispute and were in a really slow progress. After endless discussions, we finally came up with a seemingly feasible methodology, including research design, target customers, and data process procedure.
Hit the Wall
In the interim report, our suggestion of identifying loyal customers’ behavior and making the new customers the same slightly disappointed the proprietor. He appreciated our efforts but none of our “findings” really surprised him. The fact that loyal customers tended to use coupons or reserve parking lot in advance is not a big news to him. “Instead of logical thinking, I wish you to think out of box and let your imagination run wild in the next phase.” He told us.
At the same time, our personality diversity began to emerge and affected the overall meeting quality. Some of my teammates was kind of a perfectionist and couldn’t move on to next topic unless being fully persuaded. Some are the opposite. Compared with making zero mistake, they paid more attention on effectiveness of meeting. As a result, we often ended up with an unsatisfying meeting. Some felt we haven’t finished yet while others felt boring and tedious.
Solution
Solving two things at the same time is not a simple task. Fortunately, immediately after interim report, we reached a consensus that we should also take another research methodology into consideration. We dived directly into the most obvious question: “Why don’t they use our service again?” We used design thinking to list all possible reasons we could think of and verified them with the provided data.
Because not being reconciled to give up all results came from the rigorous research, we combined the new discoveries with the old ones. Altogether, we redesigned a complete research which included why loyal customers are loyal and why disloyal customers were being so. And finally, we successfully meet the proprietor’s expectations while keeping our own characteristics at the same time.
As for diverse communication style between panelists, we made ourselves distinguish between “must have” and “nice to have”. Whenever someone split hairs, we would tell them it’s a “nice to have”. Certainly, this method didn’t always kick in. We still struggled for persuading each other.
Once before the weekly meeting with proprietor, we had a discussion after the club class for the purpose of meeting the deadline. We started at 10:00 p.m. and everyone wanted to finish it as soon as possible. However, because we had a dozen solutions to decide whether to drop or keep and we discussed every detail about them thoroughly. The discussion lasted for 4 whole hours until 2:00 a.m.
At that time, I knew that maybe we were not the most talented group, but we are definitely the most hard-working one. And I was proud of it.
What I’ve Learnt
1. Communication can reduce unnecessary but time-consuming discussion.
It’s applicable to both consultation with proprietor and discussion between teammates. If we had understood fully what our proprietor wanted and what his personality was like, we might have allocated more time on conceiving creative solution and less time on arguing whether the logical deduction is correct. As for discussion within the team, we could be more dauntless when pointing out what each other can do better, and less butthurt while receiving constructive suggestion.
2. Opportunities are not offered. They must be wrested and worked for.
Although being introverted is not a shortage, our society do pay more attention to those who speak louder and act more proactive. That’s why I recommended myself for reporting in the final report despite the fact that I usually keep a low profile. My aggressiveness turned out to be a good choice and I got to made the proprietor a deep impression.
3. Comparing with learning from textbook, I like learning from doing better.
The last one is about hard power and technical skills. Being an economics student, I learnt lots of econometrics models but never had the chance to utilize them in solving real-world problem. In the project, in order to test the feasibility of the solutions, we discussed about using econometrics methods, including ‘Difference in Difference’, ‘Random Experiment’, and ‘Instrumental Variable’. Thanks to it, I got to know more about these “familiar strangers”.
According to club cadres, apart from exploratory data analysis, we are also using machine learning models to solve corporate problems the next semester（109–2）. With 2-semester of intense training in machine learning principles and various machine learning models, I can’t wait to put what I’ve learnt into practice and contribute to the corporate!
