Customer Research
There exist plenty of chats where people are looking for rides (those requesting normally don't have cars).
People who do have cars are normally not using the chats, but they are carpooling. These people often are working out amongst themselves any monetary transactions of trades for food.
Some places that people are going (outdoor activities) have roads that are not in the best condition or are gravel roads. In these cases a car that can handle the terrain is important and the person who's vehicle can handle it will drive.
The main pains that people are having is arranging rides, how to share the burden of driving, and the uncertainty of people showing up on time or even at all. There is a feeling of obligation, as a passenger, to compensate the rider or else their "bumming" a ride. There is worry about if someone is available on the days the passenger needs and ride. For the most part, people are looking for a way to organize a group that completes repeat trips. Carpooling with one other person is often easier since there is less coordination required.
A key difference in pain was identified. People with no cars looking for people with cars vs. people with cars looking to share the burden of driving.
Some important positive were also identified. It is a fun social experience to carpool. Lots of positive memories of laughing and having a good time when carpooling. Carpooling reduces emissions.
Some features people would like to see include who is available to drive on a given day, who is driving, where they're driving, and when they're driving. Also where to pick people up, who to pick up first and how to optimize this route. Keeping some kind of record to track how the driving has been balanced and if people are contributing equally (either in also driving or in monetary contribution). Finally, a feature for selecting a designated driver for a night out was also mentioned as a desired feature.
People said that they would pay for this app, but not very much money. Either 3.99 once or 1.49 a month if the features implemented were really solving all their pains.
It seems that many people are casually organizing commuting, but there is no structure and it's very circumstance based. From the people I talked to, most of them plan carpooling when they happen to see each other in person or just go by themselves and hope to see friends at their destination.
The main pain-point for carpooling is the organization between passengers and drivers. The passengers feel guilty asking a driver for a free ride so the drivers (even those who are more than happy to give a friend a ride) often don't know that other people are trying to complete the same journey as them. It's also difficult to plan the exact duration of a trip such as planning to be on the ski hill for 5 hours before returning.
The passengers' main concerns were about safety, since they don't want to get in a car with a complete stranger and feeling a burden of guilt over not contributing for a ride.
People without cars reported feeling uncomfortable asking for rides, and declined invitations to events outside of transit range because of this. They also mentioned being discouraged at having to bus or walk to a meetup location and then drive to an event. A passenger noted that if they had a car, they would want to be able to offer rides to people because of their experience having needed rides.
The lack of accessibility to nature for hiking or other activities with transit, especially when bringing a bike or kayak, was mentioned as a high pain point.
It was also stated by one respondent that if the only benefit was organizing a ride amongst friends, they would not pay for that. They were also aware of some community organizations that organized rides for events but had never participated because of the pandemic and the effort involved in contacting and joining an event.
Some drivers that I have talked to who own vehicles were willing to carpool, especially when they are travelling longer distance (e.g. drive up to Nanaimo), as long as they get some rewards, like gas money, food or coffee from the passengers. Their main pain points were it is often awkward to ask passengers to "contribute". And often organizing over chatroom takes too much time, since there is a lot of back-and-forth texting. And it is hard to find necessary information over previous chats (e.g. who's being picked up where).
Interstingly, most drivers' concern was not just safety but also cleanness of the car after carpooling.
Several concerns were raised with a subscription model that everybody has to pay. First, it doesn't allow for organizations to pay the entirety of the cost to run a group (such as a UVic club or a Tofino surf shop). Also, the ownership is put in question when everybody pays equally. For example, who gets to decide which people are accepted to the group? At this point, these questions aren't answered.
Somebody purposed an interesting idea of slightly increasing the subscription price and allowing paying users to "sponsor" a new user so they don't have to pay.
Concerns with having one person always being responsible for paying the entirety of the server cost. The server boosting idea is popular.
Didn't find much interest in extra features like push notifications, however there was interest in a custom theme for a price.
A donation system after a ride is complete was suggested. Just like free tax return software asking for donation at the end of tax return process, the users might be more willing to donate a small amount of money if the process was satisfying. It seems like people are willing to donate a small amount of money when the ride was smooth & they got the contributions from the passengers. Most people agreed that for small groups(<20 people), they would not pay more than $5 per month.
People were willing to pay for a subscription if the cost was cheap enough (around $1.50). Thought premium features were helpful but doesn't think that traffic was a valuable addition.
Our first question asked users how often they carpool currently. This question was meant to get a baseline of how familiar people are with the idea of carpooling.
The second question provided more context by asking if the user owns a car. The feature requests will probably be different for drivers and passengers so it's important to know what side they are on.
The third question was the main focus of the survey. After users tried the app, it asked them to choose which feature would give the best experience improvement. 5 choices were given to the users who assigned them a ranking from favourite to least favourite.
A total "score" was calculated for each feature by giving 5 points for every first place vote, 4 points for every second place vote, and so on until 1 point is assigned for least favourite feature. A graph was generated for each feature's total score.
Multi-car rides received the highest score so that will be our next major feature.
Our next question asked users to suggest their own feature they would like to see in the app. Since these are one-off ideas we are not going to implement these immediately, but it gives a direction for possible inquiry in the future. Some ideas include the following:
- Have the map show my route and the cost of fuel, while I edit the start and end locations
- Rating drivers and passengers.
- Allow drivers to approve pickup points and say where they are not willing to drive.
- Public calendar for people to post their trip dates that others can see
- shotgun feature (call the front seat).
- Would like to see a Refuse ride to someone I don’t know feature
- I think being able to send out regular reminders would be useful, something that could be preset by the group creator so that people getting picked up could be reminded on a regular basis to be ready to go for that time.
- Friends list
- A homepage button. (this was implemented)
Next we wanted to determine what parts of Tandem are working well and what needs to be improved. We asked the users what they liked about Tandem and what they found challenging to use. What do you like about Tandem?
- Clear layout & fun to use!
- It is really simple and straight forward. Being able to chat is also a huge plus because most restrict that so you don't just arrange a ride out of the app 3.Free small group planning 4.Simple to use because app users don't need to download the app. 5.The ability to connect with people for activities. 6.Interesting idea, pretty website 7.Fairly intuitive creation flow. Gas calculation is very handy!
- the interactive map!
- simple, unique
- Dark mode is a nice touch
- E.g. for a sports team, would encourage carpooling and make it easy for people who are less inclined to carpool.
- Fun and fresh perspective to look at car pooling
- Clean interface, gas breakdown estimate
- I love the dark mode. I think the layout and visual aspects of the app are all appealing, and the mapping system for plotting route and adding new pickups works well.
- Nice interface; seamless and logical all of the way. I like that the fuel cost is calculated per person and shown on the app. Dark mode is great.
- Website loads pretty fast
- simplicity of the website
What challenges did you have while using Tandem?
- no bike option as a solo or group 'vehicle'
- Creating a ride got stuck when entering a time. Unable to test creating a group (fixed)
- It would be better to see how other drivers or passengers get rated so that users can safely share a rider with strangers.
- The complete ride button in the ride viewer is slightly confusing. (fixed)
- "Many text explanations were unclear;
- What does start time indicate?" (fixed)
- Navigating between features (starting new ride, chat, etc) was a little confusing and I ended up back in the create ride flow (fixed)
- There are some confusing parts about the menus, such as knowing whether or not you are actively typing in a text field and some of the fading of the buttons make it seem like an option is not available rather than being the selected option.
- Some scrolling interfaces weren't intuitive because of the minimalism. Auto-suggestions of locations doesn't always work when I tried inputting the name of a location (e.g., the Roost Bakery in Sidney) instead of the location's street address, but the street address worked.
- I think instead of choosing car type and associating that with the seating and fuel usage, it would be better to have some common car manufacturers(nissan, ,mazda, toyota etc) as well as models to select from, and the seat information comes from that. Maybe i'm just dumb, but I have no idea if my car is subcompact or mid-size. As well, I have absolutely no clue what the fuel usage is like for my car, I just know it's good. I think fuel usage could also be attached to the car type if they can just select make and model from the start. Lastly, after creating a group, I wish the group name button and the create group button were distinctly different, as of right now they are the same size and font which is off putting.
- Selecting a time for scheduling a ride was janky, and 12h/24h selection difference would be appreciated. Dark mode doesn't work on the active rides interface. (fixed time and dark theme issues).
- little confusing to use. Although the web is simple, it was missing some basic features such as a main menu or a home page that has all the basic options such as settings. Some more pictures would be also appreciated on the website to make the website more vibrant when the user is on the website. (added more pictures, added home button, better settings page).
Next we asked people how much they would be willing to pay for the app. We figure that the app has more value if more people use it (since there are more people to participate in rides), so we wanted to confirm that our business model should not be a subscription model where everybody needs to pay equally.
The results confirm our hypothesis, so we will continue pursuing a model where people pay what they can.
Our last section asked for some demographic data from the users such as Name, Age, and City they live in. Our privacy agreement prevents us from sharing this data directly, but we can show generalized results. This section was used to determine if our surveyed users were in the target demographic for this application.
All surveyed users were in Western BC, but we will not share exact cities or their names.