4 ms·

Hey, I don't know whom you may be able to get to help, but I would bet that the government doesn't actually have great data on custom wait time- I've never see

Hey, I don't know whom you may be able to get to help, but I would bet that the government doesn't actually have great data on custom wait time- I've never seen them track someone from entry to pass-through. I strongly doubt they have that. Rather, I think you may have come upon a really cool opportunity- find a way to track it! You'll have data no one else on the planet has. It's like when Uber starts in a new city- the error on their time estimate starts high and decreases over time. If enough people use it, you'll get really got at predicting the wait time based on the factors you mentioned. Idea: let the traveler download a free companion to the app. Then the user links to the traveler. Once the traveler lands, they can turn on their phone and let you see location. The user gets an alert "Bob has landed!" and actually see where the traveler is. Over time, you'll get really great at refining the wait time prediction algorithm. You can ask nationality, etc. Potential problems: No phone, international traveler can't use phone in destination due to roaming charges, can't have phone on in certain custom areas.

blakewatters 14y ago
We do this in GateGuru already. I think you are underestimating the scale needed to produce statistically significant models across airports (or even within one across checkpoints). MyTSA even has trouble with the crowd sourcing and they have TSA brand. It's a classic chicken and egg problem: nobody wants to use the app until its meaningful and it's not meaningful until you have huge numbers using the app. Our approach in GateGuru is to bundle the security wait time functionality with other, highly engaging and free services (such as flight tracking, itinerary management, and our amenity database) in a completely FREE app and scale our way into it. We are deep into it and its an ongoing part of our strategy, but its not clear to me how you could generalize this and build a business around it without an amazing solution to this fundamental problem.

blakewatters 14y ago

We do this in GateGuru already. I think you are underestimating the scale needed to produce statistically significant models across airports (or even within one across checkpoints). MyTSA even has trouble with the crowd sourcing and they have TSA brand. It's a classic chicken and egg problem: nobody wants to use the app until its meaningful and it's not meaningful until you have huge numbers using the app. Our approach in GateGuru is to bundle the security wait time functionality with other, highly engaging and free services (such as flight tracking, itinerary management, and our amenity database) in a completely FREE app and scale our way into it. We are deep into it and its an ongoing part of our strategy, but its not clear to me how you could generalize this and build a business around it without an amazing solution to this fundamental problem.