Hi Mohit, Thanks for pointing out the error. We have updated it on the copy. Cheers

Salman SH

Hi Saravanan, In the first part where I quote the Delhi-based Meesho seller, her maximum earning capacity is Rs 15,000. Her expenditure is nearly zero since she doesn't own or move stock. She only pays for mobile Internet. So that Rs 15,000 is actually her maximum profit every month. However, an average Meesho seller makes between Rs 10,000 - Rs 15,000. Also, its difficult to provide an accurate ASP, since Meesho sellers list everything from cheap long-tail goods as well as costlier premium/luxury items. Usually, the seller gets 20%-25% commision per sale. Hope this clears the doubt.

Salman SH

Hi Prashant, Thanks for pointing out the error. We have updated sentence with corrections.

Salman SH

Hi Robert, Thanks for pointing this out. AFAIK, most of Foodpanda India customers are on mobile, hence the app already has been designed to portray the food-first approach that Ola has strategized for Foodpanda. Currently, the Foodpanda India website does not accept orders and is only a discovery site for restaurants.

Salman SH

Hi Saravanan, thanks for pointing this out. The health insurance data points are directly sourced from IRDA 2016-17 report. The '450 million lives covered' mentioned on the graphic, includes lives covered under both private and government-led public health insurance programmes. While the 13 million policies issued is only an estimate of policies sold by private and individual insurance companies. For a breakdown of private+public health insurance penetration please do visit this site and download the latest 2017 report: https://www.irdai.gov.in/ADMINCMS/cms/frmGeneral_NoYearList.aspx?DF=AR&mid=11.1 You will find the break down on Pg. 50 of IRDAI's 2016-17 annual report. Thanks

Salman SH

Hi Kishore, Thanks for pointing out the inconsistencies. We have corrected the line that wrongly represented the value of the real estate market. Apologies. Regarding CommonFloor. In FY15, Commonfloor made around Rs 44 crore in *operational revenue* as per its filings. This has been compared with the operating revenue that Quikr generated in Fy17, which is Rs 64 crore (consolidated basis). As per Quikr's statement, 35% of its revenue is, in fact, real estate revenue which translates to Rs 22 crore. So a company like CommonFloor which used to make Rs 44 crores is now making even lesser than what it used to make before getting acquired by Quikr. This was the point I was trying to raise. Again thanks a lot for your feedback; it only makes my work and research better.

Salman SH

Thanks for reading Ankit, glad you liked it. On the fleet utilization bit, I agree with you that scale can only be achieved when routes are better optimized. I had written about this extensively on the Swiggy article we covered last month. Do take a look.

Salman SH

Hey Tejas, what I meant to say is that both Gold and Super are aimed at building habit or stickiness to the app. Both are loyalty programs targetted at high value users. Hence the comparison.

Salman SH

True. It should work both ways. Existing users will also get to try new verticals that Swiggy will offer. Here I believe re-targeting won't be costly since Swiggy is already in process of figuring out their high-value users with their loyalty product Swiggy Super. They know who they could sell to. While at the same time, hooking on new users with milk/groceries/booze and slowly make it a habit-forming app could become costly. But, these users will be different from existing Swiggy users who are largely people who depend on the app for ordering food. They can come from say for e.g. BigBasket's loyal users who do their grocery shopping once or twice a week, and users from Dunzo who depend on the app for late-nite booze delivery. So these are a new genre of users, and not necessarily users that may have tried Swiggy in the past.

Salman SH

Hi Srinivasan, Thanks for those insights. I agree with your contention that a business model that worked in a different setting or geography may not be easily replicable in another setting. Several businesses built around the hyperlocal model have failed badly, but there are multiple lessons out there from these failures. For e.g. expanding into multiple geographies with a bullish outlook--this hasn't worked for grocery delivery startups in the past. Now instead startups like Milkbasket, DailyNinja, Dunzo, etc. are expanding within colonies/residential blocks, and towns, rather seeing expansion as a multi-geographical function. There are more such lessons, but even the new hyperlocal models will not be unit positive from Day 1. Discounting and cross-subsiding will bear huge costs for Swiggy, and they might fail in 1 or 2 categories, but at the end of the day, if Swiggy can amass even 10k new user sign-ups from their new verticals, they could be re-targeted with food or other categories for example.

Salman SH

Hi Sharat, Thanks for reading it :) It's actually pretty difficult to say what segment works or may not work for an MVNO model. End of the day its resale of services and early MVNO players like Virgin Mobiles didn't see success until they failed the first few times. Also, another trend is that MVNOs work well when deployed in smaller geographies, hence it took off faster in Europe when compared to the US. In Africa, India and other developing markets, the spectrum is scarce and costly, hence telcos are protective about it. But in India, it seems like there is abundant (lower-end) spectrum, but the last Spectrum Auctions 2016 were so costly that some telcos decided to boycott LTE. So MVNOs may not strike the right balance here in India until the central government itself stops seeing telecom as a cash-cow (i.e via revenue collection and costly auctions IMO)

Salman SH