5 ms·

HN - this was a project I built for a class at HBS (Online Economies). The class covered strategies for solving the network effects problem of a lot of online b

HN - this was a project I built for a class at HBS (Online Economies). The class covered strategies for solving the network effects problem of a lot of online businesses. I thought it would be great to actually build one of these platforms and then try to apply the strategies discussed in class. Looking for feedback on the app but happy to discuss strategies for cracking the chicken and egg problem....

jwb119 16y ago
thanks for sharing. out of curiosity, since you mention you are at HBS, are you a programmer as well? if not, what strategies did you use for building out the site?
- alooPotato 16y ago
  Yes, my undergrad degree was in computer science/software engineering - so fairly easy for me to pick up iOS programming
shasta 16y ago
The chicken and egg problem would seem to be the obvious problem with something like this. Can you broadly explain what you have in mind? I assume you're only pairing people with other Gobicab users, right? You have the following testimonial displayed prominently: "GobiCab is so easy to use. As soon as I get off my flight, I enter in my destination and have a match by the time I arrive at the cab stand." Is that experience a fluke or is that what I'd expect to find if I used Gobicab at an airport in a random city of a million people?
- DanI-S 16y ago
  Presumably the testimonial is part of solving the chicken-and-egg problem.
  - stcredzero 16y ago
    This could be managed as an exercise in expectation management. Perhaps GobiCab could do a some sort of beta at a particular airport for a particular conference? This could actually be presented as an exclusive perk for conference goers. This exercise could generate testimonials.
stcredzero 16y ago
alooP, a perfect way to crack the chicken/egg problem would be to offer a "beta" of this service at particular conferences as an exclusive conference perk. This provides you with a high population of people arriving in the same time period, all going to the same place. The fact that it's a beta can be used to manage the expectations. It could also be used to generate a lot of testimonials. I have been talking to a startup which is targeting such conferences. Send my username email at gmail dot com. --SCZ
- alooPotato 16y ago
  Yes this is one class of mobilization of a network effect business - exploit irregular network topologies. In our case that means targeting areas with high trip density (defined as locations where people are all leaving by cab to nearby destinations). So not only do we want to try and target specific cities, but also specific landmarks within those cities where this will be ultimately useful. We've identified two of these - airports and college campuses. We are working with Boston Logan airport to see if we can do some co-marketing with them and advertise to all the passengers arriving on flights and in the cab stand. As for colleges, there are a lot here in Boston! Its just a matter of reaching out to the right student associations and newspapers who can recommend the service to their students. Any connections here would be greatly appreciated! I've never thought of the conference idea before, I wonder if cabs a re a primary mode of transportation there or not. I'll contact you offline for more info....
  - stcredzero 16y ago
    One thing I'll point out: co-incidence in time alone is going to have a greater effect on user-populations and the ability to effect a match than location alone, though both are important, naturally. (Gedankenexperiment -- which is more of a stretch, time match being a stretch or location match being a stretch?) By targeting events, you have high densities of coincidence for both.
- wittjeff 16y ago
  Speaking of use at conferences, I have wished for something like this while standing in line (with hundreds of other people) waiting for a cab at Las Vegas airport. Too frequently they have under-loaded cabs going to the same hotels only minutes apart. In Vegas it scales up further though. You have lots of people hopping from hotel to hotel to see shows etc. So if I want to get from the MGM to Bellagio to see O (particularly at predictable intervals before shows start), it's a good bet a bunch of other people will too.

jwb119 16y ago

thanks for sharing. out of curiosity, since you mention you are at HBS, are you a programmer as well? if not, what strategies did you use for building out the site?

alooPotato 16y ago
Yes, my undergrad degree was in computer science/software engineering - so fairly easy for me to pick up iOS programming

shasta 16y ago

The chicken and egg problem would seem to be the obvious problem with something like this. Can you broadly explain what you have in mind? I assume you're only pairing people with other Gobicab users, right? You have the following testimonial displayed prominently: "GobiCab is so easy to use. As soon as I get off my flight, I enter in my destination and have a match by the time I arrive at the cab stand." Is that experience a fluke or is that what I'd expect to find if I used Gobicab at an airport in a random city of a million people?

DanI-S 16y ago
Presumably the testimonial is part of solving the chicken-and-egg problem.
- stcredzero 16y ago
  This could be managed as an exercise in expectation management. Perhaps GobiCab could do a some sort of beta at a particular airport for a particular conference? This could actually be presented as an exclusive perk for conference goers. This exercise could generate testimonials.

stcredzero 16y ago

alooP, a perfect way to crack the chicken/egg problem would be to offer a "beta" of this service at particular conferences as an exclusive conference perk. This provides you with a high population of people arriving in the same time period, all going to the same place. The fact that it's a beta can be used to manage the expectations. It could also be used to generate a lot of testimonials. I have been talking to a startup which is targeting such conferences. Send my username email at gmail dot com. --SCZ

alooPotato 16y ago
Yes this is one class of mobilization of a network effect business - exploit irregular network topologies. In our case that means targeting areas with high trip density (defined as locations where people are all leaving by cab to nearby destinations). So not only do we want to try and target specific cities, but also specific landmarks within those cities where this will be ultimately useful. We've identified two of these - airports and college campuses. We are working with Boston Logan airport to see if we can do some co-marketing with them and advertise to all the passengers arriving on flights and in the cab stand. As for colleges, there are a lot here in Boston! Its just a matter of reaching out to the right student associations and newspapers who can recommend the service to their students. Any connections here would be greatly appreciated! I've never thought of the conference idea before, I wonder if cabs a re a primary mode of transportation there or not. I'll contact you offline for more info....
- stcredzero 16y ago
  One thing I'll point out: co-incidence in time alone is going to have a greater effect on user-populations and the ability to effect a match than location alone, though both are important, naturally. (Gedankenexperiment -- which is more of a stretch, time match being a stretch or location match being a stretch?) By targeting events, you have high densities of coincidence for both.
wittjeff 16y ago
Speaking of use at conferences, I have wished for something like this while standing in line (with hundreds of other people) waiting for a cab at Las Vegas airport. Too frequently they have under-loaded cabs going to the same hotels only minutes apart. In Vegas it scales up further though. You have lots of people hopping from hotel to hotel to see shows etc. So if I want to get from the MGM to Bellagio to see O (particularly at predictable intervals before shows start), it's a good bet a bunch of other people will too.