Our first meetup in San Francisco is coming on Jan 23, 2020. Learn more.

How can Lyft break into private ambulance market?
+6 votes
in Product Design by (717 points) | 543 views

2 Answers

+4 votes

I would first make sure I understand the scope of this question.

Lyft wants to break into the private ambulance market. From my understanding, private ambulances:

1. Usually aren’t called by the end user and are instead usually contracted either by the hospital or through 911. (I may be wrong, but I would ask for clarification).

2. Private ambulances usually require an EMT, the car, and necessary equipment on board

3. Possibly requires some sort of licensing with the state or city they operate in

4. Are prohibitively expensive and would therefore have to include some sort of insurance company tie-in

I am going to make the assumption that Lyft wants to build a product to allow both the patient and the emergency personnel to hail an ambulance.

So I identified the users in this case being hospitals/doctors/911/emergency services, the private ambulance, and the patient themselves or loved ones of the patient. I would like to focus on the experience on the patient hailing the emergency vehicle themselves if that’s okay. Because mainly I believe this would be the main differentiator for Lyft to enter this industry as I do not personally know of another way for a patient themselves to hail an emergency vehicle short of calling 911 or the hospital.

The goal of the feature is to bring a patient to the nearest emergency room.

Next I want to delve into the user journey of calling a private ambulance. A user currently on Lyft will open their app and enter in a destination, select the type of ride experience they would like based off of price and wait for the car to arrive. In an emergency situation, a lot of this will need to be streamlined. I can identify some pain points:

1. An user who is very sick or very injured may not be primed to go through all the steps in the app
2. An user may not know where the nearest emergency room is
3. An user may want an easy way to contact loved ones in case of emergency
4. Emergency personnel may not know exactly how to reach the patient
5. An user may not know if he/she truly needs to be rushed to the hospital at this point in time
6. An user may still be in danger
7. An user may not know if their insurance will fully cover their ambulance ride


1. I think the easiest way to solve this painpoint would be to develop a SOS signal. If Lyft can integrate with the phone itself’s SOS signal that would be best, but perhaps if you open the Lyft app, we can add functionality where if you swipe right on the Set Pickup button, it will reveal a button for SOS or cancel. The goal here is to make sure that users won’t accidentally hit it, but still easy enough for an injured person to (or a loved one of the injured person).

2. Once an ambulance is called, Lyft will automatically find the nearest hospital with an emergency room.

3. We can add functionality into the app that allows a user to enter phone numbers of a loved one in case of emergency. Once SOS has been triggered, Lyft will automatically contact the emergency contact.

4. Phone GPS is usually the way for current cars to find their riders, but in case of emergency, we may not always know the best way to reach the patient. I can think of several potential solutions for this.

One, Lyft should be able to ascertain what kind of location the patient is currently in (whether it’s a building or the wilderness, etc).

Second, Lyft should be able to call the patient or the loved ones of a patient after a SOS is called just to make sure to gather more information on the location and health of the patient.

Third, if nobody picks up the phone, Lyft can ping the phone GPS again once the emergency personnel is near to make sure the location hasn’t changed. They can also play a high pitch noise from the phone when emergency personnel is close to physically be able to get closer to the patient.

5/6. Following closely with the previous point, once a SOS has been signaled, Lyft should have an emergency personnel with medical experience call the patient or the loved ones of the patient to walk them through the process. This will help Lyft understand if the patient should actually rush to the hospital or if police should be involved, etc.

7. Lyft can offer an option to enter the most updated insurance before/after the SOS is called and determine if insurance will cover the cost. This depends on the user, but knowing how complicate dinsurance is, Lyft will have to work with all the insurance companies to determine a easy solution to this problem.

Judging all the potential solutions, I go back to the goal of this question which is to get a patient from where they are at to the emergency room. I would say 1-6 are all very integral to the experience and should formulate a MVP. 7, while interesting, is not crucial to the health and safety of the patient.

To wrap it all together, I believe Lyft can break into the emergency ambulance market by offering users a more direct way to contact emergency personnel and get them to the hospital. They can do this by integrating a SOS signal into the app itself and maybe even through the Phone OS. Lyft can then provide all the services from contacting the patient, to ascertaining the location of patient, and to connecting the patient with an emergency vehicle.

See less
Hi Scott,
Great answer! Well structured, easy to understand, and full of creative solutions with strong analysis from a UX perspective. I take it from your answer that you think features 1 to 6 should make it to the MVP. While I agree, the interviewer might want to see your approach to evaluating your ideas so maybe you want to walk them through your evaluation of the features with a bit more detail. Overall, I think it’s a great answer.
Great ideas. I particularly liked the idea of leveraging Lyft to automate a bunch of things that are hard to impossible when the patient is alone and in a bad condition.
The beauty here is that the users could configure possible ambulance needs way before anything bad happens to them – desired urgent cares, exact direction to the home, automatic unlocking of the door in case SOS was activated, etc.
+1 vote

@scott – This is really a great answer. I was thinking of another angle. Since Lyft is all about crowd-sourced personal cars, I was wondering if we could create a flow for drivers with large cars who can drive a semi-sick-injured person to the nearest hospital. SO its kind of like a private ambulance but not a professional private ambulance- but a person’s car being used as a non-urgent/non critical medical transportation vehicle.

So, my user base is:
People who are not in critical danger but need help getting transportation to doctors/hospitals – Elderly people going for regular check ups, busy single mom who does not have her own car to take kids to hospital, student with a twisted ankle
People who would like to offer their private vehicles as private ambulances without the need for oxygen tanks, EMT, doctors, aides etc.

User needs:
Getting an inexpensive ride to the hospital/doctor for non critical reasons.
Getting easy access to comfortable transportation where you can lie down potentially.
Someone can help you in to the car and help you get outside the car.
Someone can wait with you while you are at the appointment and then bring you back when you are done.
Someone who can be available on a regular basis (like a subscription service) for monthly/quarterly checkups

So it would be like a regular Lyft ride but there would be some provisions such as
1. Special licensing/permission to certify your Lyft vehicle as a care giving/transport vehicle.
2. Some insurance tie-up so that you can be protected from liability of carrying a sick person/caring for a sick person
3. Having the provision for someone to get in/out of the car from a wheel chair, stretcher, lie down etc.

In this case, I would design two flows – one for the Lyft driver who is signing up as a non-urgent medical transporter (a version of a private ambulance) that’s a lot cheaper and easily accessible. And the other for the non-urgent patient who wants to schedule pick-up/drop offs for one time or multiple (subscription style) service for medical transport

Driver sign up flow:
1. Go through a special flow for sign up as a non-urgent medical transporter

a. Enter personal details – name, address, driver information, back ground check etc
Ask driver if they want to sign up as a medical transporter for non-urgent medical needs (have on boarding video showing use cases, benefits and testimonials)
b. Vehicle details – Needs to be a van which has easy entry and exit capabilities for patients with disabilities (I am assuming that sedans might not work here). Any special tweaks to the car that Lyft might suggest?
c. Wage details: I dont know the regular driver sign up flow but I am assuming that the rest of the stuff will not change – there might be different (slightly higher rates for care giving rides)
d. Insurance tie-ups – Lyft can offer special insurance for liability protection by tie -ups with insurance providers.

Patient Flow:
e. Standard sign up flow with personal information, credit card/paypal integration + question – would you like to use Lyft for non urgent medical transportation for non-urgent appointments? There could be a video onboarding tutorial showing use cases, testimonials, benefits etc).
f. Front load information such as any ailments, name of preferred care giver, insurance or have tie-up with insurance provider to pull information with patient’s permission (with HIPAA complicance needs)

Patient flow for actual need based use:
g. A patient should be able to schedule Lyft rides for planned appointments in advance, provide feedback and change transport provider based on experience/feedback
h. A patient should be able to call Lyft medical transport for non-urgent needs (twisted ankle, child sick etc) – like they would call a regular Lyft but should be able to upgrade to Lyft Medical based on state/severity of condition so that there is a higher sense of urgency
i. Feedback/Tip process should remain the same to help improve service
j. A patient should be able to even play inspirational videos/music directly through their Lyft app or be given hopeful messages while they are waiting for Lyft or enroute to care provider to create a positive environment.

As part of the MVP (and as a separate initiative from Scott’s professional private ambulance project described above) – I would build out the driver and the patient flows with requirements a, b, c, d as well as e,f, g, h and i. j is a nice to have feature. I would need to list down the features in a table and assign priority based on customer impact and level of complexity (which I have not done here).

I would ultimately measure success by driver acquisition rate, driver retention, positive feedback from drivers and patients, usage of this service by patients (out of total Lyft customer base, how many opt for this service, out of people that sign up for this service, how many use it first time and multiple times etc).

See less
Hi Nabrapm,

Thanks for the answer. Really interesting idea. It’s certainly a segment that’s easier to get into given they don’t need as many licenses as a equipped ambulance. Your target user persona is also very specific and meaningful.

I have some feedback on the structure of the answer:

– I would have listed a few more use case / pain points for your target user (they might need help with dialling, they might not be able to wait outside to look for the car and need assistance in being picked up, each might have a different special need, if they are going for an appointment, they don’t want to be late to their doctor appointment, etc).

– After choosing the important use cases, I would only list out the solutions that solve the needs / pain points I have discussed earlier. The user flows have some of that information but by listing out the solutions rather than the flow, you’re able to map each solution directly to the particular pain point you’ve highlighted earlier, making it easier for the listener to follow the purpose and importance of each solution (e.g. solution i for patient flow doesn’t solve for a specific need highlighted. you could have skipped mentioning it).

Hope it helps. I think your measurements of success metrics are also great!
Your name to display (optional):
Privacy: Your email address will only be used for sending these notifications.
To avoid this verification in future, please log in or register.
Your name to display (optional):
Privacy: Your email address will only be used for sending these notifications.
To avoid this verification in future, please log in or register.

Related questions

+1 vote
4 answers
asked Jan 22, 2018 in Product Design by bijan (717 points) | 1k views
0 votes
1 answer
+1 vote
3 answers
+1 vote
1 answer
+3 votes
4 answers
asked Nov 26, 2017 in Estimation by bijan (717 points) | 1.8k views