Why coffee chains are (still) not profitable in India.

 

In Oct 2015, before the Coffee Day IPO, I wrote a piece in LinkedIn on why I did not think CCD was a good investment or even profitable, due to fundamental problems the industry faces in India.

https://www.linkedin.com/pulse/coffee-day-ipo-worries-indian-qsrs-rahul-deans/?trackingId=3xvqbpePQkyIHi4UtWXbxw%3D%3D

I faced some flak for that, from veterans in the industry who argued that India’s rising income meant coffee chains like CCD or Starbucks, who spent a decade understanding and building the market, would now become profitable. I passed up an opportunity to be CEO of one such coffee chain as my view of the industry was too pessimistic for the owners liking – It turned out I was quite right in my assessment. 8 years have passed. CCD continues to be unprofitable, as does Starbucks and everyone else. The point of this article is not to disparage any chain – having run a QSR chain myself, but to point out some problems inherent in the industry.

Café coffee day (CCD) is 27 years old, integrated and focused on the café business. They have restructured the business twice, first before their IPO and more recently after the death of the founder.
Starbucks, the world’s biggest coffee chain partnered with the Tata group – India’s leading business house and have been running their cafés for over 10 years.
Barista is over 20 years old and have gone through several changes in ownership (which have brought alternate strategies to become profitable).
Costa is a leading International brand, which is run in India by Devyani foods, a leading player in the quick service restaurant business.

These are all companies with a very good pedigree. All have been around for more than a decade and
have over 100 stores each. 2 of the 4 companies have overhead partly shared by the larger businesses they are part of – whose overall profit, hides the losses from café business.  

To become profitable, CCD twice slashed the number of stores, to focus on fewer profitable ones.
Starbucks has looked at rapid expansion as a path to profitability. Costa has cut then increased stores, while Barista has looked at different product and outlet strategies (franchising) over the years. None have worked. I will argue that they are fundamental problems with the Indian market that conventional analysis has overlooked.

The alternative to organized retail in India is the unorganized sector, which does not have the overhead and tax of the organized players and offer a high level of efficiency and run by an entrepreneurial owner.
In the `out of home’ tea and coffee business, this competition to the café is the street `chaiwala’ or the
local South Indian restaurant (which I consider the best fast food, as their offering is fast, cheap, and healthy). Their price of a cup of tea of coffee is under 25% of the organized player. In a very price sensitive market, these are the go-to places for daily tea/coffee `out of home’ consumption. Many people who in theory are target consumers of coffee chains, will have a daily cup from a local vendor and will rarely visit a chain, unless it is a social occasion, or work-related meeting (as opposed to a daily habit).

When the only competition to a café chain is another chain, the price different between the highest and lowest priced chain is less than 2:1 (Luckin in China and Dunkin Doughnuts in the US are both 2/3 the price of Starbucks) the number of cafes increase, because there is no category of lower priced seller. Thus China has 6000 Starbucks (with the same population as India), while Indonesia, with a slightly higher per capita income (in PPP) has a similar number of Starbucks as India, in the same time (289 vs 333) with a fifth of India’s population.  

Tea is consumed by 90% of India (so coffee is an occasional indulgence). Tea chains, which have proliferated in recent years for no other reason than it is the fashionable thing to open, must benchmark themselves with the roadside vendor who can sell a cup at Rs 10. Their price per cup must be less than what it is for coffee, yet it is a multiple of the street vendor. This is a double whammy for tea chains – a lower per outlet sale than the coffee chain, yet a limited user base. The way out is premium tea cafes for specialty teas (e.g. Infinitea in Bangalore), but, by definition, these will not be able to expand.

 Starbucks has the same price per cup as the US, in a country with significantly lower purchasing power.
 If one uses the McDonalds Big Mac index, Starbucks is priced 2.5 times more than it should be. My sense is a significant consumer base for Starbucks are entrepreneurs who use the café as low rental office space (air conditioning, wi-fi and a clear restroom at a fraction of the cost of office space). 40% of Starbucks sales in the US are ‘to go’ but its negligible in India. The average Starbucks outlet in India has sales of Rs 29 lacs per month, compared to 75 lacs in the US and 40 lacs in China (with similar pricing).
My sense is the lower priced and local items introduced by Starbucks India will only cannabilise sales of higher priced items, not create new customers because price will still be a barrier to new customers. 

Some of this cannibalization also happens when Starbucks equivalents (3^rd wave, Blue Tokai) open in the same catchment with lower prices. The combined sales of both outlets will be higher than the original, but each can become unprofitable with lower volumes than what is needed to break even.  The problem is exacerbated because the number of malls or high street locations which are viable retail spaces are limited. Thus a locality will often have 4 coffee outlets all on one street, rather than spread across the catchment, allowing each to build a loyal base of customers who are walking distance from the cafe.  

It is difficult to increase sales with negligible take away sales, since sitting space is often unavailable at peak hours with sales volumes lower than what is needed for profitability (yet with pricing high enough
to lose sales to lower priced competition).  

For Coffee Day, (average sales 8 lac per month), pricing is lower than Starbucks, but a visibly less premium look and feel mean that it will neither attract many Starbucks customers and will encourage their customers (with less disposable income) to compare prices with street vendors or lower priced restaurants.

When there is higher reliance on corporate customers (employees who can sometimes debit the cost to a business expense) outlets in business parks have much lower sales on weekends. The sales slump during covid was higher for coffee chains and post recovery slower, since sales to office employees are a higher proportion of sales, compared to QSRs offering meals.

In my article 8 years ago, I had said that the size of the market is small with 60% of outlets in just 7 cities. It’s much the same today with 50% of sales from the top 7 cities. Even for the largest restaurant chain in India – Domino’s, with over 1600 outlets, has 40% of outlets in the top 7 cities. Devyani – the only profitable QSR chain (with Yum brands and Costs coffee) has 47% of its sales from the Metro cities. Unless sales per store (inflation adjusted) increase by 15% none of these chains will break even. The problem is that in Metro cities, customer spend on categories like out of home coffee is increasing by this amount, but it is going to new chains, who first set up shop in metro cities and take sales away from older stores that are just short of the sales required to be profitable. These new chains tend to focus on profitability only after reaching around 100 outlets – media coverage is almost entirely on how fats they are expanding and how customers love their product and not on unit profitability.

CCD had a peak of 1700 outlets in 2019 which have been slashed to 469 currently. Starbucks and Barista have over 300 outlets while Costa has 110. While expanding into new towns is imperative for growth, the problem is new markets are more price sensitive. If a coffee is priced more than a full meal at McDonalds or Dominos, will only be an occasional indulgence for a small number of customers in those towns. A larger proportion of these stores shut have been in the smaller towns. Zomato shut operations in 225 towns (from a total of 1000) because they collectively contributed to just 0.3% of turnover. Even the 500th biggest town in India (popl. 100,000) probably has just 2000 households able to afford a cafe.    

The QSR chain with the highest number of outlets in the `in between meals’ category is Baskin Robbins, where the number of outlets has stayed at 750-800 for over 10 years because of the problem of expanding to smaller towns with a pricing for an ice-cream (conceptually the same as a coffee consumed between meals) being the same as a full meal at the average restaurant in that town.
Domino’s has expanded in smaller towns because it has the lowest priced pizza in that town (priced lower than the lowest priced coffee sold in any chain). The average order value of a meal on Zomato (Rs 400) is about the same as the price of a coffee and a side snack at a coffee chain. 

The between meal category has variable sales depending on the season. Sales of hot beverages dip in summer but do not have a corresponding growth in winter. (Similarly, ice-cream sales dip in cold winters but do not have a corresponding growth in summer as very hot weather reduces the propensity to eat out, particularly if it only for a between meal occasion. Sales per outlet of coffee chains in Russia are higher than Pizza/Burger chains, unlike in India.

Interestingly, Devyani foods (though not its coffee business) is profitable with roughly the same sales per store as Starbucks India (Rs 33 vs 30 lac) similar gross margins (>75%) and a lower sales per employee (18 lac vs 25 lac) though the higher employee count of Devyani is partly the result of home delivery staff.  Sales per employee in Starbucks US, at the same pricing per unit, is approx. Rs 75 lac p.a. or 3 times India.
The difference in profitability between Devyani and Starbucks (which is relevant to other premium coffee chains) would be in rentals and store capex (high depreciation), both necessary for a premium image. With CCD it is lower employee productivity and higher rental and capex relative to sales.

India’s other consistently profitably chain (non-beverage) Domino’s Pizza despite lower unit pricing relative to international markets also has per outlet revenues of Rs 30 lacs (with a much lower capex than Coffee chains) a gross margin of 77% and a per employee revenue of Rs 15 lacs, despite most sales being home delivery executed by its own workforce.

With standalone coffee chains which are not part of a bigger organization (Third Wave, Blue Tokai, Gloria Jeans) corporate overhead relative to sales will often be insurmountable barrier to profitability – particularly when management has been poached and has not risen from the ranks, unlike in the West.
Royalties also tend to be higher as compared to brands where the Indian partner is a large domestic player concerned about a return on his investment.

 My other blog posts on startups. 

https://rpdeans.blogspot.com/2023/11/joining-startup-checklist.html

https://rpdeans.blogspot.com/2023/08/the-coming-unicorn-meltdown.html

https://rpdeans.blogspot.com/2023/08/the-problem-with-delivery-app-business.html