No more fastest finger first on cure.fit

How cure.fit is trying to solve the demand problem using ‘waitlisting’

cure.fit, a healthcare and fitness startup founded by former Flipkart executives Mukesh Bansal and Ankit Nagori is already scaling at a lightning speed in major cities across India. I have been personally using cure.fit offerings since the past 6 months, particularly cult.fit, which gives users a platform to book different workout sessions at a centre of their choice.

As a user of cult.fit, I have frequently run into situations where I was unable to book a cult.fit class particularly for their popular time slots (for example, Dance Fitness at 6:00 p.m. on a Saturday). However, I was fairly confident to get such slots using the fastest finger first technique.

What do I mean by fastest finger first?

A cult.fit class can only be cancelled up to T minus 60 minutes, otherwise, the user is given a no-show, which can lead to deduction of a day from the subscription. The trick is to keep looking at the class availability between 90 to 60 minutes prior to the class time. For example, if a class is at 6:00 p.m. and the slots are full, the best time to find an available slot is between 4:30 p.m. and 5:00 p.m. The primary reason is that majority of users cancel their classes at the last minute as they wouldn’t want to risk having a no-show. As a result of such cancellations, users who are fast enough to refresh the app and book, will end up getting the class. Hence the name fastest finger first. Personally, there have been times when my friends and I were able to get the class of our choice at the last minute using this approach.

When I tried to play around with their APIs…

So, instead of using the mobile app, I have sometimes used the cure.fit website to find a slot as the app navigation is slightly time consuming to play this game of fastest finger first. The web portal on the other hand, makes it easier. I also tried automating this by intercepting the APIs responsible for the cult.fit class bookings via the usual ‘inspect element’ and Charles Proxy. At the end, I could not finish the whole automation and cure.fit had come out with these new set of rules and features.

What is this whole ‘Waitlisting’ feature on cure.fit?

Almost a month ago, I received cure.fit’s email communication where they mentioned that a large number of slots (more than 1 lakh slots per month!) go unused because of members booking but not attending, commonly called ‘No-Show’. To provide a little context, here’s a brief on what is ‘Waitlisting’:

• Once class slots are full, people can join the waitlist
• Once somebody cancels, the next person in waitlist queue gets confirmed

If you are interested to know more about this feature, you can read here.

Comparing ‘Waitlisting’ feature at cure.fit with IRCTC train bookings

As most people in India would know, the train booking system by IRCTC has existed for very long and one of the key features of it is the waitlisting. Lot of cancellations happen and people get confirmed from the waitlist. It is a much more matured waitlisting system as it has been present for years. Naturally, upon hearing of ‘Waitlisting’, IRCTC came into mind and hence the comparison.

One interesting thing which IRCTC offers is the ‘tatkal’, where people can book tickets at priority by paying a premium along with few caveats. Whereas, cure.fit doesn’t have an analogous feature to compare it with (yet!) but who knows. Perhaps, there would be few slots available like ‘tatkal’ where you can burn your workout points (if cure.fit wants to take the gamification factor up a notch) to book a class.

To solve this, one can explore the existing waitlisting systems (such as IRCTC) and see if some similar concepts would work well in case of cure.fit.

Potential gaps in the ‘Waitlisting’ feature

‘Waitlisting’ feature essentially works like FIFO (first in first out) i.e. once all the slots of a class are full, the users can choose to join it under waitlist and view their position in the FIFO queue. Once a cancellation happens, the first person in the queue gets the confirmation about the class. This sounds pretty straight forward, apart from few scenarios listed below which can hamper the core goal (ensuring people get class slots of their choice).

1. Problem with people getting confirmed under waitlist at the last minute

There are lot of cancellations which happen during the last 90 to 60 minutes before a class. In some of these cases, the users in the waitlist may not have enough time to make it to the class or cancel their confirmation. To solve this case, cure.fit app pardons the waitlisted users who got confirmed in last 90 minutes from receiving a no-show in case they don’t attend.

However, the loophole here still exists wherein such confirmed waitlisted users might not show up for the class as they have nothing to loose. There is no incentive for them to cancel it as well.

2. How many people can you keep under a waitlist?

Another bound which they have added is to keep a cap of maximum allowed users in a waitlist. This would be in order to avoid overflowing of waitlist and setting false expectations of the user. Although I have seen cases where the waitlist openings are also full but slots open up for grabs in the T minus 90 minutes (in the fastest finger first fashion). However, the other way around might also be true, where people who were under waitlist never got confirmed (though the chances of that happening seems lower).

3. Problem with changing free ‘No-shows’ to membership days

Apart from ‘Waitlisting’ feature, cure.fit also changed the policy for ‘No-Shows’ where a day would be deducted directly if someone missed a class. They have also compensated users with extra cult.fit membership days in place of free no-shows present in their account. As compared to earlier, this change will incentivise the users to attend the class to keep that extra day of subscription whereas previously the free no-shows simply gave the user an option to not show up. However, a user can still go ahead and miss classes (up to their older free no-shows) without impacting their original subscription period.

How to measure success of these changes?

• Number of no-shows should come down.
• Number of drop offs after clicking ‘book a cult class’ should reduce i.e. the number of users who went out of the app section without booking a class.
• Number of people who were under waitlist but never got confirmed

In my opinion, these are the primary metrics to measure the success of ‘Waitlisting’ feature incorporated by cure.fit. However, there could be other metrics as well.

All of these issues are actually good problems to have for cure.fit as they are facing a lot of demand and their supply is not able to meet it. However, there’s a flip side to it as well. There are some classes which hit 100% capacity whereas some might only be 10–20% occupied. They will keep adding more centres but this problem will persist in some shape or form. It will be interesting to see how cure.fit will continue to solve this.

PS: As a user, I just want to get the class of my choice without any hassle. As a product manager, it will be interesting to be on the other side :)