My seven (hard-hitting) questions to Indian ecommerce businesses
[Editorial notes: This is a guest article contributed by Mrutyunjay Mishra, cofounder of JuxtConsult. If you are an entrepreneur building ecommerce business (or planning to build one), you should be reading this article.]
Recently, while preparing for a presentation for an event, I came across some interesting facts about the Indian Internet space from various studies we have conducted at Juxt. This was a private event of IUVP (Indo-US Venture Partners)– annual review of IUVP LPs (Limited Partners), where I was invited by Mr. Vinod Dham of to make a crisp (3 slides) presentation, followed by a panel discussion to be moderated by him at his Bangalore office. I wish to share these with you.
Exhibit 1: Presentation I made in that event.
Slide one (in Exhibit 1) is fairly simple for your understanding. Let’s not get into that perennial argument on different estimates from different sources (Juxt, IAMAI-IMRB, Google, etc.). If you want a one-sentence answer, then it is like “three different guys are taking three different routes to reach the destination, and depending on the route used, distance, and time taken to reach, the cost incurred varies.”
Similarly, the estimate of Internet users in India will vary because all of us are using different methodologies, different amounts of money/resources, and our own time to arrive at the same number. You as a fairly smart guy should have the choice to believe and follow one number. Keep in mind the story of the famous statistics professor who before making his son cross the river took 10 measurements of the depth of the river and calculated the average of all ten, and despite the average depth of the river being less than his son’s height, the son eventually drowned.
To keep it simple, only Indian Census is the most accurate, rest all of us do estimates. Therefore, just focus on the proportions and growth percentages. Internet users base is growing steady at a 24% CAGR for last 5 years. However, most of the Internet users are “Shoppers” (Search Travel & Non-Travel products online), and this proportion has remained around 75% to 80% of total Internet users in India. The proportion of “Online Buyers” (Buy Travel & Non-Travel products online) is increasing y-o-y, at a CAGR of almost 42%. This is what makes everyone euphoric and excited, that the number of buyers is growing at a rate that is almost double that of the regular Internet users’ growth.
Yet, as the CAGR definition itself suggests, it is fuzzy; it is an imaginary number calculated using a formula with numbers of a particular period. Please keep in mind that the real growth of buyers happened only between the years 2010 and 2011, growing almost 70% over that of last year (from 10 million to 17 million unique buyers). In essence, slide one tells us that “Internet is growing, will continue to grow; buyers are growing and continue to grow.” What is more important is that “many people search online but are not buying online.”
Therefore, the first question: “Why are they only searching and not buying online?”
If we try to understand the buyers and what they are buying from the Internet, then we will find that there are some people who exclusively buy travel products only, some buy non-travel products only, and some of them buy both. But they are all spending in the range of USD 5 to USD 90 across various categories, which makes the average transaction size around USD 45 (Rs. 2,300) for non-travel products.
This made the total non-travel ecommerce market in 2011 worth around USD 400 Mn. (Rs. 2,000 Cr.). But please remember that this average ticket size is a critical indicator to decide your cost of acquiring buyers on the Net. If you are acquiring someone for Rs. 1,000 and the average transaction size is Rs. 2,000 and your margin is 10%, then you need to make at least 5 sales of Rs. 2,000 each to the same person to earn Rs. 10,000, in order to recover your acquisition cost. The mathematics is not that simple, as in order to make those 5 sales you need many things in place and all of those things cost you money. Therefore, make a pressure of more than 10 sales of Rs. 2,000 each.
On my last slide, the key point I want everyone to watch out on ecommerce is: the number of unique non-travel buyers per month in India as on JAS 2011 was around 2 million (20 lakh) and growing at 26% CQGR (year is replaced by quarter in CAGR – over last 9 quarters).
The current buying behaviour with reference to average number of products shows that these buyers are not buying more than 2 products on an average. Most of them pay by debit cards, followed by credit cards and cash on delivery. You will find that these buyers are, on an average, using ‘two’ modes of payment for buying non-travel products online.
Finally, “the proportion of debit card, net banking and credit card users among the Internet users has remained almost similar for last four years.” Cash on delivery has its own challenges of logistics and, therefore, return rates. Return rates kill the ability to make money and increase your gross sales pressure. Going by our earlier example, if you make Rs. 10 on one sale and if you lose Rs. 2 because of the returns on cash on delivery, then in order keep your margin at the same Rs. 10 level you need to make 25% more volume. That changes the complexity of the game.
Check this simple grid to understand how returns really impact the margin and increases the volume pressure:
Exhibit 2: Impact of return on CoD on overall volume pressure
While I was preparing the presentation, the point on credit and debit card stagnation on the Internet made me dig a little deeper. I started searching for RBI (Reserve Bank of India) numbers around plastic card usage in India. To my astonishment, I found the following (first 4 columns and then derived the next 2):
Exhibit 3: Some relevant payment-related data taken from a RBI report
Source: RBI, Feb ’10 till Feb’11 Data
The 2nd column tells us that in India there are only 18 million (1.8 crores) credit cards and 240 million (24crores) debit/ATM cards. No wonder, we find that most of the transactions on IRCTC websites are undertaken through debit cards/net banking followed by credit card, and now on eCommerce websites, cash on delivery (calculate the share of transactions by gateway types over the total). Credit card numbers are definitely pathetic; almost 83% of all credit cards are already on Internet (25% of Internet users—15 million own Credit cards). In fact, more importantly, I also saw a Times of India article sometime back quoting some RBI numbers on credit cards going down year on year in India (check the figure ‘Discredited’ on top right corner of the article).
Juxt’s regular studies estimate that only about 30 million of these 240 million debit cards are currently on the Internet. I said cards, not people or users, because these are just cards – the number of people should be lesser as some own multiple credit and debit cards.
The 3rd column indicates that total amount transacted on credit/debit card swiping at POS* (includes POS / ECom / IVR / MOTO transactions) is around 10% of the total amount withdrawn from ATMs. This leads to some more questions:
2nd: “Can someone swipe credit/debit card everywhere?” (same RBI report says there are only 565,542 POS* terminals in whole of India)
3rd: “Do they really want to swipe their card everywhere?” (else, why is so much of cash getting withdrawn?)
4th: “Why would an average Indian like to swipe their card?” (in other words, “What is the incentive for general consumers in swiping their debit card?”)
Everything becomes more interesting when one looks at the number of transactions. If you divide the number of transactions with the number of cards, it emerges that only about 14 transactions happen per credit card per year. That’s it. You will find only around 0.96 transactions per debit card per year and about 17 cash withdrawals per debit/ATM card per year.
I wish RBI had provided me with unique cards used for these transactions and distribution of transaction sizes, and many more relevant indicators in the same report. Alas, everything is not public data. But this erroneous calculation of average transactions and transaction value per card essentially tells us that “not everyone is using their credit and debit cards”. The average number of transactions is going down because we are dividing it with the complete base of cards and not the real base of unique cards that were used for these transactions, not reported in the same RBI report. However, I can confidently guess that not more than 65 to 70 million debit cards are currently getting swiped, and maybe not more than 150 million cards are used to withdraw money. Rest all cards issued by banks are possibly not used by the owners. You are free to make your own guess or send me some numbers if you find them elsewhere.
Intuitively, the debit card numbers are good enough to drive a developed payment mechanism. However, even in my personal experience, whenever I try to offer my debit card instead of cash in many shops across the country, most of the merchants with POS generally ask me to pay up for the 2.5% or whatever the banks charge them for every transaction through debit card. By the way, that makes me ask the 5th and 6th questions:
5th: “Do merchants ask you to pay for the commission charged to them by banks?”
6th: “Why do banks charge more than the transaction cost, which should ideally be a few rupees, when the payment is made through Debit Cards?”
As you may be aware, in some of the developed countries where the adoption is higher, the banks charge a few cents per transaction as transaction-enablement fee. Probably I am an ignorant person and I don’t understand the banking economics and logic for the prevalent method of “transaction fee as a percentage of transaction ticket size” in India. I assume that if banks will charge a relatively smaller amount in case of debit card transactions, that would in turn make more merchants accept debit cards. The existing debit card numbers are good enough to drive electronic payment usage. I also believe that merchants, particularly eCommerce companies, in India should go to RBI and request for directives on a smaller transaction fee on debit cards. If not anything, it will help them pass on a little more discount to the buyer or increase own profit margin. I am sure both are good enough reasons to start lobbying. J
The point I am trying to make through this post is that “the electronic payment mechanisms in our country are still very underdeveloped and needs a lot of work.”
Now that you have read the post, do answer my final question:
7th: “Is Cash on Delivery a choice or a compulsion of the current mass ecommerce companies?”
Look forward to your answers to have a meaningful discussion.