Comparison of Food Delivery Unit Economics in India with that in the US

This guest column is authored by Vivek Singla, SVP, CLSA

Food delivery companies have been the subject of constant debate since the model first caught fire in India a few years back. Fuelled with easy money, a lot of food technology companies, ranging from ordering cum delivery platforms to fullstack models had jumped on the serve India bandwagon. A lot of these on-demand eateries either folded up or merged in a wave of restructuring due to weaker execution (read Foodpanda, TinyOwl). But even execution aside, most virtual chains have failed to make a big splash.

Swiggy, one of the few remaining but successful online delivery companies here closed a US$15mn round some months ago, valuing the startup at a post money valuation of approx. US$160mn. Modelled on the US market leader Grubhub (or UK’s JustEat), it earns largely from the commissions on transactions. Its investors hope to import Grubhub’s phenomenal overseas success into India.

But there are some fundamental differences between the Indian and the American food delivery markets.

Firstly, the Average Order Value (AOV) is comparatively low in India. While it’s true that the delivery costs in India, largely consisting of incentives driven compensations paid to field executives, of US$1-1.5 per order are much lower than the US$7-8 seen in the US but part of that benefit is negated by the lower US$5-7 post discount AOV here. In comparison, AOVs in the US hover around US$30, almost 5x those of India. Ceteris Paribus, lower the food ticket, lower is the amount earned in commissions.

Secondly, the market driven take rate for pure delivery in India is maximum about US$80c per order (normally included in the commission itself), while Grubhub is reported to charge the restaurants and/or customers as high as US$4, again almost 5x.

Thirdly, the Indian customer is also not known to be as generous as her American equal when it comes to delivery charges and tips. In the US, the drivers can earn as high as 15% of the AOV in tips. This is not the case here. Hence, even if the Indian customer does pay a delivery charge (when the order value is below the minimum threshold), the amount is generally routed back to the field executives in the form of perks.

Put it the other way, while Grubhub loses 50% on an incremental basis, the Indian platforms may lose even upto 100% at margin on pure delivery. Even Zomato’s delivery partners lose about 25% per order, despite handling others types of deliveries like ecommerce and groceries in order to optimize utilizations during lean phases.

There are other lessons that can be learned from Gruhhub’s growth. Its Q2’16 earnings call revealed that it is losing US$4mn a quarter on its delivery operations, subsidized by the deeply profitable demand generation division. Also, Grubhub’s growth is mainly driven by increased penetration and increase in AOV, but the average frequency of orders per active diner is dropping due to expansion into new territories beyond the top cities. This leads to certain takeaways for its Indian counterparts. It is unlikely that the restaurants here will partake more than what they are doing already. As a hyperlocal model, food delivery is limited to the fixed catchment in the vicinity of the restaurant. Once the customer loyalties set in, restaurants are unlikely to pay more to win repeat orders for the same customer. In fact, the take rate may decrease as the restaurant rating inches up. Individual AOV may increase with inflation but the combined AOV would take a hit as the business expands outside the metros.

Then there are other externalities which are specific to the delivery business like traffic and weather conditions. Not to forget the significant cost of recruitment given the high rate of attrition in this industry. It would be interesting to see whether the penetration, density and frequency of takeout customers, especially in new markets outside the metros, are able to justify the cost of operating a full scope pan India delivery network.

The views expressed in the article are strictly of my own.

Have ideas to share? Submit a post on iamwire


  1. 1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>