6 ms·

Background: Every day in South Africa low income people suffer through 2+ hour commutes into work using public transport. This has a serious impact on their liv

by ismail 11y ago

Background: Every day in South Africa low income people suffer through 2+ hour commutes into work using public transport. This has a serious impact on their lives, due to long wait times and inefficiencies. I interviewed a single mother, who only got home after 7pm and leaves at 6am in the morning. We have built a ride share matching system: Our goal is to make use of unused seats in cars matching drivers and passengers. We do not put any cars on the road (i.e Uber), We match to drivers on their daily commute. Our Biggest Problems: 1. We are quite successful at converting people who are posting to online classifieds etc. We have strategies to get these users. We grew week on week by as much as 50% but then we hit a negative growth week. In order to continue growing, we need to expand into other areas. We also have an un-balanced market. Demand for rides is much higher than rides offered. 2. We have not figured out how we will make money yet. Our initial hypothesis was a transaction fee. The passenger and driver will be traveling together on a regular basis. Once we match them, we run the risk of being dis-intermediated. You could make an argument that a payment directly into the bank account is more convenient than cash, though i am not sure that is a strong enough incentive. 3. Our churn/attrition is very high, There is no need for the system once they start traveling together. We have churn built in to our model.

kevin 11y ago
When I think about marketplace problems, I ideally want to solve one side of the equation at a time. That means understanding whether this is supply constrained or demand constrained. Looks like you have an abundance of demand (people who want rides) and not enough supply (drivers). So all your efforts should be focused on how to get people to offer rides. Usually the incentive for drivers when I've seen other companies work on this problem is they get to drive in the carpool lane. Does that not work in your instance? Why couldn't you have drivers be paid by passengers and you take a cut of that process? I think a lot of people are selfish about their cars because they're valuing efficiency and control of their commute over everything else. Adding a passenger reduces those primary values. Can you think of other values that might overcome those? Money is one, but considering the lack of interest to meet that demand, it might not be enough. If I were creating a ridesharing program that I wanted people to talk about organically, I'd want to design it in a way that commuting via my program was FAR better than commuting by myself. Off the top of my head, the best part of my commute is listening to a great podcast. If you could get exclusive content that only ridesharers can get access to if they ride together...that could be interesting. Doesn't need to be entertainment. Could be learning a language together or business skills, etc. Why did you think you had a negative growth week?
- ismail 11y ago
  Thanks for the response Kevin. - Yes we realize that the supply side is constrained, and lacking. Passengers seek out a solution and find us easily. - There are no car-pool lanes in South Africa, they had a trial a few years back but did not continue with it. We have spoken to people at a few municipalities. For the city, they have parking constraints so ride-sharing is something they want to encourage. However, passing new legislation may take quite a while. We can not be sitting around in the hope of this. - This was our initial plan. The current behavior is: The driver & passenger will agree on pricing, timing etc. Then they pay weekly or monthly. This happens mostly in cash, currently. The passenger and driver will now travel together for the foreseeable future until something changes (i.e someone moves, new job etc). At this point, there is no incentive for the driver to continue receiving payment via the platform. The triggers to search for a new ride share per a year is quite low. So strategies like ratings, and reviews do not encourage keeping transactions on the system. People on the passenger side may not even have a credit card or bank accounts. - Yes people are selfish, and value convenience. Spending an extra 10-15 minutes to pick-up and drop-off is an issue. Which is why most people do not do it. In an ideal world, if we had complete information on all drivers and passengers, we could match person x with y who is just 2 minutes away. One of the other blockers to offering a ride, is the 'commitment'. They now need to think twice, if they want to sleep in late or if they want to call in sick. - This is a great idea, do you think having a PR stunt like "Ride with someone famous" Or "Ride in a cool car" would work? The content idea seems feasible, we would need to figure out what content is important to our market. From our analytics our #1 segment is Jobs/Employment - Growth example: Week 1 our growth was +86% # trips posted. The following week we had -26% # trips posted based on the previous week We are saturating our current acquisition methods. Other strategies have not been working as well (Off-Line) The # people searching for a ride on classifieds are small, the market of people commuting via public transport is massive, the people using public transport are not searching online for a solution.
  - kevin 11y ago
    Yes, people are selfish. That's why you should think of what you can offer them that will make them want to sit in the car for 10-15 minutes longer. That's why unique content came to my head, I've sat in driveways much longer just to hear the end of a good radio program. Also, I don't like stunts. It's not sustainable and it's short term thinking. I think you should shoot for building a great community. That means thinking for the long term and building a lasting relationship with your customers and have them build those relationships with each other. You seem more focused on what you can extract from your users. It's better to start with what can you give that they can't help themselves but want to take.
ismail 11y ago
If anyone else has any thoughts mail me : (hnuser)@codiez.co.za
ismail 11y ago
If anyone else has any thoughts mail me : (hnuser)@codiez.co.za
biot 11y ago
How about this: riders pay a reasonable monthly membership fee based on distance; 0-15km daily = $X/month, 15-25km daily = $Y/month, etc. Drivers pay nothing. Payment to the driver works like with Flattr... every month, each rider took m rides totaling n km. Take your cut of the rider's monthly fee and the remainder is divided proportionally across each ride based on some allocation of base amount per ride plus amount per km, and the driver gets paid. This may avoid disintermediation as the price should be more than fair for the rider that they wouldn't want to have to deal with forking over cash (from what I've read about South Africa, I suspect there's a risk of getting robbed which this prevents). And what happens if a rider is sick or has a day off and they need to pair up with another driver? Additionally, look at what AirBNB has done: offer supplemental insurance to cover unfortunate circumstances. Investigate what it would cost to provide some additional insurance to riders and/or drivers such that if they try to go outside the system, they're not covered. I'm not sure what kind of insurance coverage is mandatory for drivers, but if there's something many drivers routinely avoid (eg: they have collision only, but no liability) there's an opportunity to step in and provide something extra. And/or what some auto clubs have: drivers are covered for so many flat tire repairs, jump starts for dead batteries, etc. Leaving the system cuts both drivers and riders off from this coverage. Also, look for ways to partner and offer additional benefits. There's probably some underdog gas/petrol chain which would love to be the exclusive provider and could kick back a small percentage of purchases. Ideally it's kicked back to your company via a reward card and is part of your payment to drivers. Rate drivers on vehicle cleanliness and reliability and now there's an incentive to have a clean, well-maintained vehicle... and you can partner with a chain of mechanics and auto detailers and give your members a 25% discount or whatever you can negotiate. As you're dealing with many low income people, not sure what level of tech you're facing, but you could also provide mobile app notifications or SMS a few minutes before the driver arrives. Finally, if the local government wants to make ride sharing a priority why not partner with them to subsidize the program on a per-rider basis? This may let you offer a cash incentive to riders for signup and/or get 100% of your fees through this subsidization, allowing you to pass through all of the monthly fees to the driver. With whatever other benefits you offer, there's really NO reason for them to leave the program.
dpatru 11y ago
Expecting regular commuters to continue paying is like a dating site expecting married couples to continue paying. Focus instead on helping irregular travelers find rides or minibus entrepreneurs find paying passengers.

by ismail 11y ago

kevin 11y ago

When I think about marketplace problems, I ideally want to solve one side of the equation at a time. That means understanding whether this is supply constrained or demand constrained. Looks like you have an abundance of demand (people who want rides) and not enough supply (drivers). So all your efforts should be focused on how to get people to offer rides. Usually the incentive for drivers when I've seen other companies work on this problem is they get to drive in the carpool lane. Does that not work in your instance? Why couldn't you have drivers be paid by passengers and you take a cut of that process? I think a lot of people are selfish about their cars because they're valuing efficiency and control of their commute over everything else. Adding a passenger reduces those primary values. Can you think of other values that might overcome those? Money is one, but considering the lack of interest to meet that demand, it might not be enough. If I were creating a ridesharing program that I wanted people to talk about organically, I'd want to design it in a way that commuting via my program was FAR better than commuting by myself. Off the top of my head, the best part of my commute is listening to a great podcast. If you could get exclusive content that only ridesharers can get access to if they ride together...that could be interesting. Doesn't need to be entertainment. Could be learning a language together or business skills, etc. Why did you think you had a negative growth week?

ismail 11y ago
Thanks for the response Kevin. - Yes we realize that the supply side is constrained, and lacking. Passengers seek out a solution and find us easily. - There are no car-pool lanes in South Africa, they had a trial a few years back but did not continue with it. We have spoken to people at a few municipalities. For the city, they have parking constraints so ride-sharing is something they want to encourage. However, passing new legislation may take quite a while. We can not be sitting around in the hope of this. - This was our initial plan. The current behavior is: The driver & passenger will agree on pricing, timing etc. Then they pay weekly or monthly. This happens mostly in cash, currently. The passenger and driver will now travel together for the foreseeable future until something changes (i.e someone moves, new job etc). At this point, there is no incentive for the driver to continue receiving payment via the platform. The triggers to search for a new ride share per a year is quite low. So strategies like ratings, and reviews do not encourage keeping transactions on the system. People on the passenger side may not even have a credit card or bank accounts. - Yes people are selfish, and value convenience. Spending an extra 10-15 minutes to pick-up and drop-off is an issue. Which is why most people do not do it. In an ideal world, if we had complete information on all drivers and passengers, we could match person x with y who is just 2 minutes away. One of the other blockers to offering a ride, is the 'commitment'. They now need to think twice, if they want to sleep in late or if they want to call in sick. - This is a great idea, do you think having a PR stunt like "Ride with someone famous" Or "Ride in a cool car" would work? The content idea seems feasible, we would need to figure out what content is important to our market. From our analytics our #1 segment is Jobs/Employment - Growth example: Week 1 our growth was +86% # trips posted. The following week we had -26% # trips posted based on the previous week We are saturating our current acquisition methods. Other strategies have not been working as well (Off-Line) The # people searching for a ride on classifieds are small, the market of people commuting via public transport is massive, the people using public transport are not searching online for a solution.
- kevin 11y ago
  Yes, people are selfish. That's why you should think of what you can offer them that will make them want to sit in the car for 10-15 minutes longer. That's why unique content came to my head, I've sat in driveways much longer just to hear the end of a good radio program. Also, I don't like stunts. It's not sustainable and it's short term thinking. I think you should shoot for building a great community. That means thinking for the long term and building a lasting relationship with your customers and have them build those relationships with each other. You seem more focused on what you can extract from your users. It's better to start with what can you give that they can't help themselves but want to take.

ismail 11y ago

If anyone else has any thoughts mail me : (hnuser)@codiez.co.za

ismail 11y ago

If anyone else has any thoughts mail me : (hnuser)@codiez.co.za

biot 11y ago

How about this: riders pay a reasonable monthly membership fee based on distance; 0-15km daily = $X/month, 15-25km daily = $Y/month, etc. Drivers pay nothing. Payment to the driver works like with Flattr... every month, each rider took m rides totaling n km. Take your cut of the rider's monthly fee and the remainder is divided proportionally across each ride based on some allocation of base amount per ride plus amount per km, and the driver gets paid. This may avoid disintermediation as the price should be more than fair for the rider that they wouldn't want to have to deal with forking over cash (from what I've read about South Africa, I suspect there's a risk of getting robbed which this prevents). And what happens if a rider is sick or has a day off and they need to pair up with another driver? Additionally, look at what AirBNB has done: offer supplemental insurance to cover unfortunate circumstances. Investigate what it would cost to provide some additional insurance to riders and/or drivers such that if they try to go outside the system, they're not covered. I'm not sure what kind of insurance coverage is mandatory for drivers, but if there's something many drivers routinely avoid (eg: they have collision only, but no liability) there's an opportunity to step in and provide something extra. And/or what some auto clubs have: drivers are covered for so many flat tire repairs, jump starts for dead batteries, etc. Leaving the system cuts both drivers and riders off from this coverage. Also, look for ways to partner and offer additional benefits. There's probably some underdog gas/petrol chain which would love to be the exclusive provider and could kick back a small percentage of purchases. Ideally it's kicked back to your company via a reward card and is part of your payment to drivers. Rate drivers on vehicle cleanliness and reliability and now there's an incentive to have a clean, well-maintained vehicle... and you can partner with a chain of mechanics and auto detailers and give your members a 25% discount or whatever you can negotiate. As you're dealing with many low income people, not sure what level of tech you're facing, but you could also provide mobile app notifications or SMS a few minutes before the driver arrives. Finally, if the local government wants to make ride sharing a priority why not partner with them to subsidize the program on a per-rider basis? This may let you offer a cash incentive to riders for signup and/or get 100% of your fees through this subsidization, allowing you to pass through all of the monthly fees to the driver. With whatever other benefits you offer, there's really NO reason for them to leave the program.

dpatru 11y ago

Expecting regular commuters to continue paying is like a dating site expecting married couples to continue paying. Focus instead on helping irregular travelers find rides or minibus entrepreneurs find paying passengers.