E-commerce – Platform Business Firms

From glass boxes to black boxes

I wrote about how platforms are glass box organisations, given their network effects. But there are a lot of black boxes out there. In the digital platform space, there are a lot of firms that operate as black boxes. The glass box metaphor was heavily drawn on the internal culture of the firm, that is, what the employees of the firms do. This post is about what the employees don’t do.

Before I elaborate, let me highlight the work of one of my doctoral students, Sandeep Lakshmipathy. He elucidates four primary value creation opportunities of multi-sided platforms: Discovery, matching, transaction, and evaluation. In short, discovery platforms help reduce the search costs for the sides of the platform (think Craigslist), matching platforms use filters and algorithms to ensure that the preferences of both sides are catered to while delivering a match (think Tinder), transaction platforms reduce the frictions and transaction costs in interacting with the other sides of the platform (think MasterCard), and evaluation platform enable ratings/ reviews/ recommendations/ feedback of the service (think Yelp). Sure, some platforms provide multiple value, like Uber provides discovery, transaction, and evaluation; whereas Airbnb provides all four.

Our focus today is just matching platforms, and how they create more and more opaque black boxes.

Matching platforms

In one of our conversations about the difference between discovery and matching platforms, Sandeep quipped about the difference between choosing a tomato sauce on an e-commerce platform and choosing a partner on Tinder. While it is sufficient for me to like the tomato sauce, it is not important for the tomato sauce to like me! Unlike this, in a matching platform like Tinder, it is imperative for both sides to have liked each other. Here is where the algorithms kick in. Magic! Matching algorithms.

Matching algorithms are technically taught in graph theory. Graph theorists discuss two types of matching – common vertex matching and bipartite matching. Common vertex matching is used to match, for instance, people with similar interests, like students in a class interested in a specific project. On the other hand, bipartite matching is used to match two subsets with each other, like buyers and sellers in an e-commerce platform. There are various algorithms used by graph theorists, including Hungarian maximum matching algorithm, Edmond’s matching algorithm, and the Hopcroft-Karp algorithm. The specificity of the algorithms notwithstanding, each of these algorithms work on the basis of three things – the specific preferences as limiting criteria, the minimum matches to be returned, and the maximum matches possible.

Imagine when you search for books to read on a peer-to-peer book-reading platform, and based on your preferences, you get exactly one recommendation. Just not sufficient enough choice, right? On the other hand, irrespective of the filters you add, if the recommendations do not change (the same titles keep appearing) and you get a recommendation of 5632 books, you feel overwhelmed. Both sides of the matching algorithms, there is a problem – of underwhelming and overwhelming choice. It is exactly to solve these problems that matching algorithms collect enough data from the users.

Some algorithms start with providing random matches, and based on the expressed user preferences, gradually mature their matching. Some others start at one extreme – like the shopping assistant in a mom-n-pop retail store. Once she’s estimated your broad preferences and budgets, she is most likely to start with showing you options very close to your budget. Pretty much like the default sorting algorithm (lowest price first) in the case of travel/ hotel aggregation platforms. Again, the platforms “learn” based on your expressed preferences.

Expressed preferences to profiling

Now, how do these platforms learn, and what do they learn about your preferences is the black box. Based on a few expressed preferences, these matching algorithms may end up profiling the user, and start providing more and more matches based on that specific profile. Sometimes, these preferences may be specific to a context – like me searching for a business class air ticket (when the client has agreed to pay – I travel economy otherwise!). The algorithm has no way of separating out such preferences without access to a large number and variety of such searches. And such preference-based profiling is easy.

Have you wondered why you get “more relevant” advertisements on a Google search results page than on YouTube? Both owned by the same firm, and possibly can share the matching algorithms. However, the way one could profile search users based on their text inputs and matching them with appropriate websites and advertisements is far easier than in the context of video content. Video content may be devoid of explicit tags that are indexed, may contain sarcasm, the audio and video content may be inadequate, or just non-existent. So, there are days I have been left wondering why I was exhibited a particular advertisement in the beginning of a YouTube video. Sometimes, I conjure up my own hypothesis about what actions in my history, the cookies on my browser/ device (my kids do use my iPad), and what specific search terms triggered those. Black box!

Customization-personalization or privacy: A trade-off

This is no simple trade-off: the one between customization and personalization and privacy. By exposing my preference for a particular sport (like cricket) to YouTube, I get to see a lot of interesting cricket video suggestions, right up front on the home page. By subscribing to specific channels, liking certain content, and commenting on some others, I help YouTube learn more about me. These actions provide enough inputs for the platform to customize the match and personalize suggestions. However, there are limits to when such expressed preferences breach privacy. Like when my phone’s AI assistant suggests a wake-up time for me based on my first appointment of the day, or when my wearable device chides for not walking enough during the day (how many steps can you walk in days like today, when the entire country is under lockdown?).

This is not new; and it could be creepy. Remember the 2012 story about the US retailer Target sending out mailers about baby care to a teenage girl, and the parent discovering it? Neither the local Target store had a clue, nor did the parent. The Target central database was able to predict something as personal as teenage pregnancy! Target realized that they were spooking people, and would randomize the offers, like putting ads for lawn mowers next to diaper coupons. But still, they knew!

Dealing with black boxes

I wish I had a set of recommendations for you! One day, possibly. But today, this is just a personal trade-off. As for me, I wouldn’t mind sharing my location data with my phone as long as it provides me good navigation services. And my food delivery app to know where I am, so that I get choices from hyper-local restaurants. I would go one step further and allow my photos app to have my location access so that I could organize my photos by date and location. But to allow location access to online newspapers, no.

Happy matching.

Stay home, stay safe, stay healthy.

Predatory pricing in multi-sided platforms

Over the last few days, living and commuting in Nuremberg, I realised that I was not missing Uber. While just about a month ago, commuting for a week in suburban Paris, I was completely dependent on Uber for even the shortest of distances. The penetration of public transport and my familiarity with the city of Nuremberg aside, I began wondering how would Uber price its services in a city like Nuremberg (when it enters here, which I doubt very much would happen in the next few years), where public transport is omni-present, efficient, and affordable. It surely should adopt predatory pricing.

In this post, I will elaborate on the concept of predatory pricing in the context of multi-sided platforms.

Theory alert! If you are uncomfortable with theory, skip directly to the illustration and come back to read the theory.

What is predatory pricing?

Economists and policy makers concerned about market efficiencies and fair competition have been obsessed with the concept of predatory pricing for a long time. The most common definition of predatory pricing is through the application of the conventional Areeda-Turner test. Published way back in 1975, in spite of its limitations, most countries and courts have used it consistently, due to, in some ways, lack of any credible alternative.

The Areeda-Turner test is based on two basic premises. The recoupment premise states that the firm indulging in predatory pricing should be able to predict and be confident of its ability to recoup the losses through higher profits as competition exits the market. The assumption is that the firm could reasonably anticipate the (opportunity) costs of predatory pricing, as well as have an estimate of the future value of monopoly profits; and the net present value of such predatory pricing to push competition out of the market should be positive and attractive. In plain English, the firm should be able to project the effect of lower prices in terms of lower competition and higher profits in the future.

How low can this predatory price be? That is the subject of the second premise – the AVC premise. The firm’s prices (at business as usual volumes) should be below its average variable costs (AVC), or marginal costs in the short run. If the prices were indeed above the AVC, the firm would argue that they are indeed more efficient than competition, due to any of their resources, processes, or organisational arrangements. It is when the price falls below the AVC that the question of unfair competition arises – the firm might be subsidising its losses.

Take for instance, a start-up that is piloting an innovative technology. It may price its products/ services at a price below the AVC to gain valuable feedback from its lead users, but in the absence of a recoupment premise such pricing might not qualify as predatory pricing. On the other hand, imagine a new entrant with superior technology who can bring costs down to a level where the prices fall below the marginal costs of the competitors but stay well above the firm’s AVC, it is just disrupting the market.

Only when both the conditions are met, i.e., when the predator’s prices are below the AVC and the firm could project the extent of recoupment due to monopoly profits as competition exits the market, that we call it predatory pricing.

Predatory pricing in MSPs

There has been a lot of discussion about how ecommerce firms in India have been indulging in predatory pricing and how various platforms have been going under. I had written about subsidies and freebies from a consumer perspective a few months ago (Free… continue hoyenga). Let us discuss how and why it is difficult to assess if a lower-than-competition price is indeed predatory in the context of multi-sided platforms (MSPs).

Multi-sided platforms have a unique problem to solve in their early days, that of network mobilisation. A situation that is like a chicken-egg problem, or a Penguin problem, where “nobody joins unless everyone joins” is prevalent in establishing a two-sided or multi-sided platform (for more details about the Penguin problem and network mobilisation strategies, read my earlier post here). In order to build a sufficient user base on one side, a common strategy is to subsidise, even provide the services free.
Another common feature of MSPs is the existence of subsidy-sides and money-sides of users. The platform might subsidise one side of users and make money from the other side, while incurring costs of providing services to both sides, depending on the relative price elasticities and willingness to affiliate with the other side of the platform. And the prices for the subsidy side would surely below costs for that side. It is imperative that the overall costs and prices are considered while analysing these pricing strategies.
These cross-side network effects will surely force the platforms to price their services most efficiently across both the sides. Even for the money side, the platform might not be able to charge extraordinary prices as such prices would themselves act against the sustenance of these cross-side network effects. It is likely that these extra-normal profits would evaporate through subsidies on the other side to keep the network effects active. Imagine a situation where a B2B marketplace charged the sellers higher than normal prices, only large (and desperate) sellers would affiliate with the marketplace, leading to buyers (the subsidy side) leaving the platform. In order to keep the buyers interested, the marketplace might either have to broaden the base of sellers by optimising the prices, or provide extraordinary subsidies to the buyers to keep them interested. So in order to maintain the equilibrium, the platform would have to price the sides efficiently.
Finally, in a competitive situation, not all competitors might follow the same price structure. So, a reduction of prices by one competitor for one side of the market may not force all other competitors to reduce prices; they may just encourage multi-homing (allowing users to use competitive products simultaneously) or manipulate the price on the other side of users.

So, a direct application of the Areeda-Turner test might not be appropriate while studying predatory pricing in the context of MSPs.

An illustration

Let us imagine a market for home tutors supporting school students. The market is inherently geographically constrained; it is very unlikely that either the teacher or the student would travel across cities for this purpose. For the time being, let us assume that there is no technology (like video conferencing) being used.

This market is apt for the entry of a multi-sided platform, like LocalTutor. This firm provides a platform for the discovery and matching of freelance tutors with students. LocalTutor monetises the student side by charging a monthly fee (that includes a platform commission), and passes on the fees to the tutor. We need to make two assumptions before we proceed with competitive entry and predatory pricing: the market is fully penetrated (all the students who are looking for tutors and tutors looking for students are all in the market) and there are no new students and tutors entering the market; and there are no special preferences between student-tutor matches, i.e., the student-tutor pair does not form a bond like a sportsperson-coach, where they begin working like a team. In other words, the tutor is seamlessly (with no loss of efficiency) replaceable.

Now imagine a new competitor enters the market and engages in predatory pricing to kick-in network effects. The new entrant, let’s call it GlobalTutor (a fictitious name), drops the student-side prices to half. In order to attract the right number and quality of tutors, GlobalTutor has to sustain the same fees that LocalTutor provides its tutors, if not more. So, it starts dipping into its capital reserves and begins paying the tutors the market rates while reducing the student fees. Anticipating a larger surge in student numbers, more tutors sign up to GlobalTutor, and seeing the number and quality of tutors on GlobalTutor (at least if it is not inferior to LocalTutor), students first start multi-homing (use both services for their different needs, like LocalTutor for mathematics and GlobalTutor for music classes), and some of them begin switching.

In a fully penetrated market, the only way for LocalTutor to compete is to respond with its price structure. It has two options – reduce the student-side prices to restrain switching and multi-homing behaviour; and tweak the tutor-side prices and incentives. The first option is straightforward; it is cost-enhancing and profit-reducing. The second option (which is not available for pipeline businesses) is interesting in the context of platform businesses.

There are various ways of responding to this threat. The intent is to arrest switching and multi-homing behaviour of tutors and students from LocalTutor to GlobalTutor.

Increasing multi-homing costs of tutors by providing them with incentives based on exclusivity/ volume: Like what Uber/ Ola provides its drivers – the incentives kick-in at what the company believes is the most a driver can do when they do not multi-home. In other words, if you multi-homed, drove your car with both Ola and Uber, you would never reach those volumes required to earn your incentives in either of your platforms.
Contractual exclusion: This might not be tenable in most courts of law, if these freelance tutors were not your ‘employees’. Given the tone of most courts on Uber’s relations with its driver-partners (drivers’ lack of control in most of the transaction decisions including choice of destination, pricing, and passenger choice), any such contracting would imply that the tutors would be employees, and that would significantly increase the platform’s costs (paying for employee benefits are always more expensive than outsourcing to independent service providers).
Increase contract tenure: LocalTutor may increase multi-homing and switching costs by increasing the tenure of the contracting from monthly to annual. Annual contracting will reduce the flexibility that students and tutors have, and might result in reduction in volume.
The next options for LocalTutor are to work at the two restraining assumptions we made at the beginning – penetration and perpetual matching. LocalTutor might want to add in more and more students and tutors and expand the market, providing unique and differentiated services like art & craft classes, preparation for science Olympiads, or other competitive tests. LocalTutor might also communicate the value of teaming of student-tutor pairs in its success stories, in a bid to dis-incentivise switching and multi-homing.

To predate or not to predate is not the question

Given the differences between pipeline and platform businesses new entrants seeking to mobilize network effects have very little option but to resort to predatory pricing. The choice is not if, but how. And as an incumbent, should you be prepared for a new entrant who would resort to predatory pricing? Surely, yes! And how? By being ready to expand the market and increasing switching and multi-homing costs. Unlike in the tutoring business that is inherently geographically constrained, a lot of businesses could span across markets. Even tutoring could leverage technology to reach a global audience.

Just one comforting thought, predatory pricing as a strategy to eliminate competition is inefficient in the long run. The new entrant might adopt predatory pricing to eliminate competition in the short run, but the act of predatory pricing breaks down most barriers to entry, and sends signals to others that there is a market that is easy to enter. It might attract a more highly capitalised competitor to enter the market with the same strategies … making the market a ‘contestable market’. And no one wants to make a fortune in a contestable market, right? More on competing in contestable markets, subsequently.

Cheers

App-in-app?

I recently got an email from my airline app that I could book my car ride within the same app. It was a way of providing end-to-end services. Much like the home pickup and drop service provided for business class customers by the Emirates. What are the implications of these for the customer, the airline, and the cab-hailing firm? Let’s explore.

It is an app-redirect

First, read the terms of how it works in the case of Jet Airways and Uber here. The substantive part of the T&C is hidden in the paragraphs quoted below:

“PLEASE NOTE, YOU ARE MAKING THE PAYMENT TO UBER DIRECTLY. JET AIRWAYS IS NOT RESPONSIBLE / INVOLVED IN THIS FULFILMENT PROCESS. JET AIRWAYS WILL NOT BE LIABLE AND/OR RESPONSIBLE FOR REFUNDS, DELAYS, REJECTIONS, PAYMENT AND FULFILLMENT OR OTHERWISE OF THE SERVICES OR IN RESPECT OF ANY DISPUTES IN RELATION THERETO, IN ANY MANNER WHATSOEVER.” (emphasis original)

Then, what is the value of this app-in-app integration?

Customer perspective

For the customer, it has the potential to work as a seamless end-to-end service. I imagine a future, where you would find a partner using Tinder or TrulyMadly, plan your evening to a game/ movie using BookMyShow, find a restaurant & book your table using Zomato, and take Uber whenever you are ready to move on, or better still, have an Ola Rentals car waiting for you through the evening. All in one app. Wouldn’t you love it, if all of it were integrated in one App? Just imagine the convenience if your restaurant-finder knew that you are in a particular concert at a specific place and you are likely to head out for dinner at a particular time. This specific knowledge could immensely help your restaurant-finder app to customize the experience for you – for instance, it could not only provide you those restaurant options that are open late in the evening after the concert was over, in a location that is close to the venue; it could possibly alert the restaurant that you were arriving in 15 minutes, based on your Uber location. And through the evening, post your pictures on Instagram and SnapChat, check-in to all those locations in Facebook, and Tweet the experience live.

Yes, you would leave a perfect trail for the entire evening in a single place, and if you were to be involved in an investigation, it would be so easy for the officer to trace you! No need for Sherlock Holmes and Watson here – the integrator app would take care of all the snooping for you!

Convenience or scary? What are the safeguards related to such data sharing across different entities? How will the data be regulated?

The Integrator perspective

Why would a Jet Airways provide an Uber link inside its App? Surely cab-hailing and air travel are complementary services. Plus, Jet Airways believes that its customers would find it convenient to book an Uber ride from within the Jet Airways app, as they trust the app to provide Uber with all the relevant details – like the estimated landing/ boarding time of the flight, drop/ pickup addresses, etc. Jet Airways also needs to believe that its customers would rather choose an Uber cab, rather than its competitor OLA Cabs, or any other airport taxi service. The brands should have compatible positioning. Given that Jet Airways is a full service carrier, and differentiates based on its service quality, Uber might be a good fit. But the same might not hold good for a low-cost/ regional carrier like TruJet connecting cities like Tirupati, where Uber does not operate.

Does integrating complementary services affect customer satisfaction, brand loyalty, customer switching costs, and/or multi-homing costs? In contexts where these services and brands are compatible, and there is a convenience involved in sharing of data between these services, there is likely to be some value added. Like airlines and hotels (hotels would like to know your travel schedule); currency exchanges and international travel (the currency exchange would love to know which countries you are visiting); or international mobile services. If there was no data to be shared between the complementary services, the user would rather have them unbundled. Think travel and stock brokerage.

That said, platforms find innovative complementarities. For instance, airlines (primarily the full-service carriers) have launched co-branded credit cards. In a recent visit to Chennai, there were more American Express staff at the Jet Airways lounge than the airline or lounge staff! And they were obviously signing up customers. What are the complementarities between credit cards and air travel, apart from paying from that card? A lot of business travellers have their business travel desks do the payments; consultants have their clients booking the tickets; and even for individuals and entrepreneurs, the credit card market is so fragmented that everyone holds multiple cards. And the payment gateways accept all possible payment options, including “paying cash at the airport counter”. They why co-brand credit cards – sharing of reward points/ airline miles. Either customers do not earn sufficient airline miles and using these co-branded credit cards help them earn more miles and retain/ upgrade their airline status (remember the 2009 movie, Up in the Air?); or they do not earn enough reward points in using their credit cards that they can redeem their airline miles as credit card reward points. Either ways, each one is covering up for the other.

In this covering up, or more diplomatically consolidation of rewards, the partners increase customer switching and multi-homing costs. Surely, redeemable airline miles might be more valuable to a frequent traveller than credit card reward points that have limited redemption/ cash back opportunities. But for loyalty to increase, it is imperative that both brands stand on their own – providing compatible services.

Mother of all apps

All this looks futuristic to you? A lot of you have been using an ubiquitous desktop app known as the browser for a long time, which has been doing exactly this! In a subtle form, though. However, there are firms that own multiple such apps, and they use a single sign-on – like all of Google services. Plus, even third-party sites like Quora allow for using your Google credentials to sign-in. The trade-offs are not always explicitly specified – it is always the case of caveat emptor – consumer beware.

So, the next time you experience some cross-marketing across platforms/ apps, think what data might be shared across both the apps; and if you would really value the integration.

Cheers!

Surge pricing for food delivery: when not to use surge pricing?

This post comes to you from Friedrich Alexander Universitat Erlangen-Nuremberg, where I am visiting for the past one week. I have been teaching a course on Platform Strategies here for the past four years. While in Nuremberg, the question has always been about food, how does a vegetarian, teetotaler survive in Franconia, Bavaria, Germany? To be fair, I have had great vegetarian food here in Nuremberg over the past so many years, and this year has been exceptional – we (my teaching assistant and I) have found great Indian restaurants, that I have had an Indian vegetarian meal for dinner every day of my stay here (except one night of Italian food). Thank you, Nuremberg.

Coming back to food, I was intrigued when I read in the Uber company blog (read it here) that Uber Eats (Uber’s food delivery service) would begin charging customers surge pricing. Much like the way they charge for their ride-hailing services. I began looking for when and how surge pricing can work. I believe it is a function of customer willingness to pay in part, but most importantly, the platform’s ability to scale up and down service levels at will on the other part.

Economics of the surge

A market is made up of demand, supply, pricing and the norms around exchange. For a market to function, the norms of exchange should be fair and acceptable to the transacting parties. Some markets are defined by the actions of intermediaries who set the norms of exchange, like a stock exchange, a municipal council, or a platform like Uber. In most cases, these intermediaries are third parties in the true sense of the word, “third”, meaning independent of the transacting parties. And in a ‘efficient market’, the intermediary sets the boundaries of behaviours of the transacting parties, and let them transact with little or no involvement. However, in platforms like Uber, the intermediary takes a much larger role, say in pricing. It not only decides the prices of the rides (for both riders and drivers), it also uses pricing as a tool to modify demand and supply conditions. Surge pricing is used as a mechanism to increase supply of cars (by motivating more driver partners to join the system at that point of time), and decrease the demand for cars (by getting riders to either postpone their rides to off-peak times or move away from Uber to other modes of transport, like bus or train). There is enough that has been written about surge pricing, including in this very blog, previously.

Surge pricing in food delivery

Alison Griswold wrote in the Quartz online magazine about why surge pricing for food delivery by Uber Eats is a bad idea (read his article here). She definitely writes wonderful stories about the sharing economy. She argues that once Uber Eats introduces surge pricing, customers would shift away from Uber, and move on to other services, may be even Amazon (with its Prime services). Given that food delivery services do not have high multi-homing costs (customers can simultaneously affiliate with multiple service providers at the same time), and some services may cater to special preferences like a specific cuisine, customers might surely switch in terms of choosing their delivery partner, their restaurant choice, or both. But that can be overcome by just simple speed and other aspects of service quality.

However, her main argument is that the economics of surge pricing might work for increasing more delivery partners to join the system in times of peak demand, but might not get the restaurants to produce more food. She avers that increasing the supply of food available for delivery is not the same as increasing the supply of delivery partners. Fair point. But, don’t restaurants anyway plan for increase in food supply during lunch and dinner times? Don’t they build in some buffer of raw material, ingredients, and/ or semi-processed food before they toss them on the stove? Aren’t there some limits to which they can extend?

Where does surge pricing not work?

Surge pricing works in markets where the intermediaries can, at least at the margin, increase the supply of goods and services and/ or decrease the demand for goods and services. In the case of ride-hailing services, surge pricing can shift people away from ride-hailing to use buses/ trains or just walk. Surge pricing works best when there is idle capacity not available to the users – when the driver partners are present but are themselves taking a break (not logged in) and are not available to take rides. Surge pricing motivates these ‘idle’ capacity to join the market, and restores the balance. In summary, surge pricing works when the demand side has ‘substitutes’ and the supply side has ‘excess capacity’.

If either of these conditions are not met, surge pricing might not work. Take an instance when a cricket/ football game or a concert ends in the middle of the night, and there are no public transportation options. Any amount of surge pricing is unlikely to reduce the demand for cars. Or try surge pricing of rail tickets in Indian trains. Any amount of surge pricing is not going to motivate the rail authorities to increase capacity to balance the market (I am not even convinced it should be called surge pricing – it is just differential pricing of different tickets, depending on whether I am the first person booking the seat or the last). In both of these conditions, differential pricing might be grudgingly accepted by the transacting parties, without any impact on the demand-supply mismatches. Take for example, Kayani Bakery in Pune, India, where by noon they are sold out! Surely, no amount of surge pricing is motivating these businessmen to increase supply. In fact, the scarcity increases the demand for these biscuits.

What are the welfare effects of surge pricing?

Scarcity principle tells us that when supply is far less than demand, prices will rise to ensure that supply matches demand. In an ideal world, both supply will increase and demand will fall. However, in contexts where supply is limited or inelastic, it will be demand that has to come down. In the case of essential goods and services (inelastic demand), prices continue to rise to point where only the wealthy could afford it. This is precisely the reason why governments indulge in market intervention mechanisms. For those interested in how commodity prices can bring down governments, read this!

The lesson for platform business firms: engage in surge pricing only when you can work towards increasing supply, or your demand side has (at least imperfect) substitutes.

FirstCry.com: Leveraging the power of offline

In my blog post last week, I wrote about how a hybrid online and offline strategy is useful for collecting small data. As a couple of my readers pointed out, what marketers and strategists call small data, ethnographers and sociologists call as thick data. Honestly, I had not heard of thick data. @fernandogaldino introduced thick data to me. I dug through the online references on thick data, and realized we are talking the same thing. Exactly the same thing. Thank you Fernando! So, I am going to continue using the term small data (that is what I have read academic articles about) with the caveat that small data is also thick data. Last week, I promised to delve deep into FirstCry.com and its online-offline strategy. Here it goes.

FirstCry.com

The firm was founded by Supam Maheshwari and Amitava Saha in 2010 as a pure online venture. By 2011, FirstCry.com opened its first offline stores. In an interview to TechCircle, co-founder Supam Maheshwari elucidated how a vertically focused ecommerce firm could survive and make money in a market dominated by horizontally spread competitors (you can read the interview here). He talked about replicating Quidsi’s business model in the Indian market, by owning a set of vertical markets like diapers.com and soap.com. In replication, firstcry.com has a sister website goodlife.com. The key difference, he said, between the Indian and the US market for baby care products was that, more than 95% of the products were imported. In fact, that was the seed for the enterprise – his own difficulty in finding good quality products for his child in India, whereas he could buy a lot of them during his international travels. That effectively makes this business inherently inventory-heavy. One needs to leverage economies of scale and scope in sourcing, hold inventory and invest in logistics to be able to service customers across the length and breadth of the country (read about firstcry.com inventory model here).

Omni-channel strategy

Here is where the omni-channel strategy helps. Instead of keeping inventory in dark warehouses, ready to be shipped, it was possible for firstcry.com to open retail outlets in tier II and tier III cities (where real estate was also likely to be cheaper), where ecommerce penetration was not as much as the tier I cities and the metros like Mumbai, Delhi, or Bangalore. The inventory holding was thus distributed across the various franchised retail outlets. The outlets also provided customers with the look and feel of the products before they bought them – you need to appreciate that baby apparel and shoes dominate the market, only to be followed by toys and diapers. Clothes and shoes … when was the last time you bought your own shoe purely online? Inventory provided increased footfalls to the store, created brand awareness, and inventory off-take. The decision to have the same prices between online and offline stores, coupled with large touch screen interfaces to shop online from within the offline store could have provided exponential growth in traffic and sales.

Promotion: The FirstCry Box

Firstcry.com began promoting using traditional mass media – television and online ads. They invested in Bollywood’s longest serving (possibly) celebrity, Mr. Amitabh Bachchan as their brand ambassador and launched a few television advertisements (see some of their ads on YouTube here). However, they soon realized that mass media advertising was highly expensive and yielded low returns for a niche range like baby care products. That is when the idea of the FirstCry Box was born. The FirstCry Box is a bundle of some essential products that the mother would need during the first few days of the baby and mother reaching home from the hospital. Firstcry.com has agreements with over 6000 hospitals, through which these FirstCry Boxes are gifted to the new mothers, congratulating them on the birth. These boxes also contained gift and discount coupons from major brands of baby products, that the parents could redeem at either online at firstcry.com or any of their retail outlets. This ‘welcome kit’ to parenting provided firstcry.com a significant opportunity to build brand equity and recall amongst the over 70000 mothers receiving these kits every month. Some marketers call this permission marketing (read about it here), or direct-to-parents strategy. For me, it is a wonderful platform, a two-sided platform mediated by firstcry.com. Parents, especially first-time mothers, are initiated into parenting with the help of these grooming products (basic diapers and lotions) and the gift coupons for free. The new mothers as a subsidy side is being financed by the brands that provide the products and coupons to be included in the box and act as the money side in the platform. For the brands, this is highly targeted sampling of their products, and most mothers would stay loyal to quality brands/ retail stores in baby products. In the entire transaction between the mothers and the brands, firstcry.com benefits significantly in three ways: (a) store loyalty resulting in increased sales, (b) small data about how these mothers use these products, the basket, frequency of purchase, and willingness-to-pay for quality; and (c) good quality prediction of demand in specific geographies, leading to efficiencies in inventory and supply chain management practices.

By the way, such welcome kits are not entirely new – a lot of employers have been on-boarding their employees with such welcome kits. I first heard/ saw such welcome kits when I was part of a team that delivered a customized training programme for the ITES service provider ADP India, a few years back. It was fascinating to see how the entire family was on-boarded into the firm! Not just ADP, a variety of other new age firms, I see have adopted this practice (read this article on how some Indian organizations welcome their employees). I wish I was welcomed like this by my employers!

Are hybrid models here to stay?

I would say, yes. We saw how Amazon was opening stores in our blog last week. We also discussed how Amazon.in was using firms like StoreKing to reach the Indian retail hinterland. I read last week that their Indian competitor, Flipkart.com was also opening offline store to reach users in small cities (Flipkart to open offline stores as well). And in vertical markets like baby products, it has become all the more important to target your promotion very narrowly, and focus on the backend (inventory, supply chain, and logistics) efficiencies, while at the same time achieve scale.

Is vertical ecommerce a winner-takes-all market?

Three industry conditions define a platform market as a winner-takes-all market: presence of strong cross-side network effects, high multi-homing costs for the users, and the absence of special requirements. The baby products retail market is dominated by imported brands, is a highly fragmented industry, and the brand owners are dependent on their retail partners to promote their brands. The demand for these products are relative price inelastic, and consumers would be willing to pay premiums for sustained quality and reliability. An aggregator platform like firstcry.com would significantly aid in establishing and reinforcing the cross-side network effects between the brands and consumers. Second condition – the quality and reliability concerns of the parents would ensure significant store loyalty and brand loyalty. As long as there are no serious concerns, consumers would be loath to switch; and when the fill rates are high (there are no stock-outs of items that they want to buy) in their preferred stores, would not multi-home. In other words, consumer switching costs for brands are high, and as long as these brands are available with their favorite retailer, they would not shop from multiple outlets. And most infants have the same needs – diapers, creams, lotions, oils, and basic toys. Special preferences begin showing up only when they ‘grow up’. Some of them don’t ever grow up, but that is a different matter!

Firstcry.com and BabyOye merger and further consolidation

Given the industry conditions of geographically distributed year-round demand, operational efficiency and leveraging economies of scale and scope become key success factors. Consolidation is inevitable to achieve both backend (sourcing, inventory, and supply chain/ logistics) efficiencies as well as frontend scale (online and offline stores distributed across the country). That is why we would see waves of consolidation in such strong vertical markets. Like how firstcry.com and BabyOye merged their operations, I agree with Supam that this market will see more and more such mergers (read his interview here).

Lessons for enterprises focused on vertical markets

Based on what we have discussed over the past few weeks, I would urge enterprises focused on vertical markets (like firstcry.com) to (a) seriously consider your business model to include online and offline consumer touchpoints … for instance, online furniture store, Urban ladder is ‘pivoting’ to offline stores (read the news here) and are positioning their offline stores as customer experience centers; (b) invest in collecting and analyzing small (or thick) data through these omni-channel (or hybrid) business models; and (c) critically evaluate if the market conditions favour winner-takes-all dynamics.

Hope my readers from India and the diaspora had a great deepavali festival! Greetings from Bangalore.

Disclaimer: I am in no way related to FirstCry.com, Goodlife.com, its investors, or its founders.

Regulating Platforms

Over the past few months, there have been a lot of disputes between platform businesses, governments, and a lot of these have gone to courts as well. Last Friday (26 August 2016) issue of the Mint newspaper carried an opinion piece titled “the tricky business of regulating disruptors” (read it here). The editorial while labeling almost all platform businesses as disruptors, just stopped short of calling all of them disruptors. In this blog post, I dig deep into the issue of if and how platform businesses need to be regulated with respect to consumer protection, without impeding innovation and thence providing fair business opportunities to businesses (and returns to investors).

Defining the industry boundaries

One of the key determinants of “competitive” behavior is the definition of the relevant industry. What is competitive and what is anti-competitive can depend on how narrow or broad you cast your net while defining the industry. For instance, the Mint editorial explains in detail how in a 1953 verdict on DuPont’s monopoly on the cellophane as a result of “result, business skill, and competitive activity”, despite having over 75% market share in the cellophane market, because the courts defined the “relevant” market as flexible packaging material, and not cellophane, the product. However, in most cases against platform businesses like Uber, the competition commissions and other regulators have defined the market as app-based taxi services, and therefore looked at the market being usurped by monopolies (Didi-Uber combine in China) or a duopoly comprising of Uber and a local operator (like Grab in SE Asia, OLA in India, Lyft in the USA).

Is Uber a competitor or substitute to Taxi?

In a detailed response to Prof. Aswath Damodaran’s 2014 article on Uber’s valuation (read it here), Bill Gurley (a series A investor and board member of Uber) defined three things (read Bill Gurley’s blog post here).

He argues that Uber has since transformed the industry so much that one’s market size estimates based on current taxi market sizes is flawed. In other words, Uber was providing customers with far more value and a very different set of value propositions than a traditional taxi service – quick discovery, easy payment, predictability of service, quality (dual rating of riders and drivers), and trust/ safety. He talked about how Uber’s customers are using it to transport young adults/ children or older parents in the “comfort and safety” of an Uber, rather than a taxi.
He argued that given the economies of scale that arose due to the positive cross-side network effects, more and more drivers and riders adopted Uber, and Uber expanded to more and more geographies, and the prices fell. And the price elasticity contributed to more demand and therefore more network effects. The economics of Uber (and therefore other ride-hailing app-based services) are very different from the city Taxi services.
Uber is not a taxi alternative – it is a car-ownership (or a car-rental) alternative. When the liquidity (availability + density) of Uber vehicles is so high in every geography you want to travel to, you would rather not rent/ buy a car, but use Uber. The convenience and reduced cost of Uber as an alternative to ownership is something that he substantiates with data and analysis.

In other words, Uber was indeed a disruptor, and therefore was entitled to be treated as a separate industry. It is not a competitor to the for-hire taxi, it is an alternative; much the same way Kodak was bankrupted by digital photography (and not by competitors like Fuji).

Creative destruction and Schumpeterian waves of technology innovation

The Mint editorial called for Honorable Judges to not set taxi fares, simply because these disruptors would transform the industry through their technology innovation, and any restraining regulation would hinder these Schumpeterian waves. It is therefore an indirect call for letting these disruptors alone, let the waves of Schumpeterian technology innovation hit the markets, before we arrive at a stability of sorts. Regulation can wait.

Can regulation wait, and allow for a disruptor, in the excuse that the market is a “winner-takes-all” market monopolize the market? The popular arguments against monopolies is that of consumer protection, and that when monopolies rule, consumers suffer – prices rise, service levels fall, and there may be no alternatives. This is exactly the case for another wave of creative destruction.

My primary thesis is that when such disruptions happen on the basis of network effects, leading to economies and scale, and the disruption is based on parameters like improved customer service, lower prices, and transparent/ fair transactions (trust/ safety and the like), monopolies are not necessarily bad. When such monopolies emerge and the customer experiences deteriorate, as dictated by traditional industrial economics theory, the market will be ripe for another wave of Schumpeterian technology innovation. The waves of market entry in the Indian airlines market is testimony to these (1990s – privatization and shake-up leaving two state-owned and two private competitors; 2000s – entry of low-cost carriers leading to the demise/ consolidation of all stuck-in-the middle competitors; 2010s – entry and strengthening of regional airlines, is it?) waves of creative destruction.

Yes, there is space for other competitors, but not so much for Uber replicas. The market is indeed a winner-takes-all market (as I have argued in the past), and therefore there is just enough room for small, losing replicators. Look around the markets for Uber competitors, you do not find any market fragmented. While differentiation and creating niches is the prescription for firms competing with Uber, I request the regulators to begin treating such platform businesses as an independent market and let the inefficient product-markets fail, if required. No one cried when the offline ticket counters of Indian Railways are declining sales, thanks to the volumes garnered by IRCTC (some claim that this is the world’s largest ecommerce platform, is that true?). No one complains about bookmyshow.com garnering huge market shares in the app-based movie seat booking market, claiming that the livelihoods of the ticket clerks are under threat. Why cry about Uber, or for that matter, OLA, Grab, or Lyft?

There is already sufficient discrimination against these disruptors. In a recent visit to San Francisco, I made an extra effort (okay, walked down a flight of escalators) to click a picture at the SFO airport that read, “app-based taxis to pick-up from departures level”. Honorable Judges, please leave them alone, enjoy your ride/ movies/ every other service, contribute to the economies of scale, and let the market be disrupted.

Cheers.

Flipkart Ads – Is there a shift in online advertising economics?

Yesterday, I read an interview with Sanjay Ramakrishnan, Senior Director & Head – Business development & Marketing, Flipkart Ads in Advertising Age India (read it here). It set me thinking, why is Flipkart into advertisements? Is it competing with Amazon or with Google, Facebook, and Apple as well?

Though I am tempted to label this development as the advertising market becoming a contestable market, I will refrain from doing so. Let me first explain what is a contestable market (in simple terms, of course, let me try; and in the context of platform business models), and then proceed to analyze if the success of Flipkart Ads is a source of worry for other platforms whose principal business model is based on advertising revenue.

The theory of contestable markets originated from the works of Baumol as early as 1982 in this seminal paper (available through JSTOR here). He (and his co-authors in subsequent papers) defined a contestable market as one with absolutely free entry and costless exit. Which implies that such a market would be vulnerable for a hit-and-run entry, i.e., by any competitor with no need for any specific assets, process capabilities, or differentiation.

A key characteristic of these markets is that the new entrant takes the prices prevailing in the market (of the incumbents) as given, and enters with the same price. In a perfect competition, any new entrant will increase the supply in the market, and should lead to a reduction in prices. Even when the market shares of incumbents and new entrants change, the industry price levels should ideally fall with increase in supply. In contrast, in a contestable market, the new entrant could enter the market with the same price as the incumbents. The justification for this assertion could be based on two arguments, that the new entrant enters the market at such a small scale compared to the incumbents that there is no visible change in the total market supply to warrant a price correction. The second argument is founded on the thesis that the incumbents cannot retaliate with sufficient speed to counter the threat posed by the new entrant, due to their systems and processes that bind them to a particular cost structure and a positioning in the market. In such a case, the new entrant could enter the market with a prior contract, preferably a long-term contract, at least as long-term as it takes for the incumbent to respond. In perfect competition or monopolistic competition, incumbent firms will adopt limit pricing strategies (if profitable for them) to keep new entrants at bay, i.e., as the incumbents sense the threat of new entry, they would reduce the prices to a level where it would be profitable for the incumbents and not for the new entrant. Take for example, when cola firms entered the bottled water market in India, the incumbent, Bisleri International embarked on a strategy of keeping market prices so low that it took a long time for Coca Cola Company, and Pepsico to break even.

The second aspect of contestable markets is the absence of any sunk costs whatsoever for both the incumbent and the new entrant. If any upfront fixed costs were to be incurred by a competitor either prior to entry (including in studying the feasibility of making money in that market) or at entry (like setting up manufacturing and distribution capabilities), the costs of entry will prevent this market from becoming contestable. Let me provide an example. In today’s world, setting up an online store entails no sunk costs for any retailer. The domain registration and hosting, website design, payment gateways, and fulfillment are all functions that are unbundled and offered as independent services (as SaaS) by different vendors, which makes all of them variable costs, rather than fixed costs. Such costs are neither fixed nor specific – one could use the payment gateway for any other online transaction, should this venture fail. Such markets with no sunk costs result in no barriers to entry and exit and therefore, are contestable. Contrast this with our previous example of Coca Cola Company and Pepsico entering the Indian bottled water market – this is a market that requires significant bottling and distribution capabilities. Though the cola firms enter this market with significant synergies from their core business, there were certain unique capabilities that the bottled water market required – sourcing of good quality water and plastic bottles, bottling lines that were specific to water, unique branding, and wider distribution networks.

The third characteristic of contestable markets is that the products are absolutely non-differentiable. That means, the new entrants can enter the market and imitate the products/ services offered by the incumbents at the same costs or even lower, and therefore maintain the same price levels. It is also possible that the new entrants enter with lower prices, and offer the same ‘standard’ products or provide additional features at the same or lower prices. Such standardization is highly visible in the context of platform services, like a C2C marketplace. In the absence of any product differentiation between competitors (any new feature is imitable quickly and is almost costless to do so), Quickr.com and OLX.in entered the market and took market share from incumbents like Sulekha.com or asklaila.com.

In summary, a market can be (or become) contestable when either of these conditions are met – no changes in prices (no limit pricing by incumbents), no fixed sunk costs, and no differentiation in products and services offered.

Is digital advertising becoming a contestable market?

For digital advertising market to become (and be) a contestable market, it has to allow for costless entry and exit, no sunk costs, and no differentiation. In the case of Flipkart Ads, I would argue that it would have cost Flipkart next to nothing to build the platform. The ecommerce store was in any case dealing with sellers, and all that they had to do was to extend the relationship to brands. And remember, in the Indian market, a lot of the brands had their own ecommerce retail operations and some of them were already on Flipkart as sellers. For instance, when I searched for the Prestige brand of pressure cookers on Flipkart, I found about 40-odd sellers including TTK Prestige, the brand owner.

And when Flipkart entered the digital advertising space, did Google and Facebook respond with limit pricing? I am not sure they did. A Feb 2015 LiveMint article that announced Flipkart and SnapDeal’s entry into online advertising space gave Google ad revenues as US$55bn compared to Amazon’s US$1bn (read it here). Given these sizes, it is unlikely Google and Facebook would have felt the need to respond to their entry by lowering prices.

Developing the advertising platform would possibly not involve any sunk costs for Flipkart. There is sufficient traction in terms of relationships with sellers and brands, the technology platform costs next to nothing to build, transaction costs are variable (including cloud storage and payment gateways), and even brand building is costless (as they are extending the same brand – Flipkart Ads).

It is the third condition of contestable markets that protects the online advertising market from becoming a contestable market, i.e., lack of differentiation. In the case of online advertising market, differentiation is created and sustained by superior targeting of advertisements to the right users. Measuring and monitoring engagement of the audience is the key in data collection; deep understanding of the consumer behavior and decision-making process is critical in analyzing this large volume of data; and close relationships with a wide variety of advertisers is imperative to ensure narrowcasting of advertisements to specific audience profiles. Here is where the product differentiation kicks in – the kind of browsing habit data that Google has access to is very different from the ‘buying intent’ that Flipkart can derive out of its customers’ behaviors. And especially in the context of mobile apps, the Flipkart app has access to other information like the person’s location, WiFi/ data connection information, and even his contacts; all of which could be useful to provide targeted narrowcasting (or even unicasting) of advertisements. Such shrinking of segments and the ability to serve what the marketers call ‘the segment of one’ customer can differentiate the new entrant, Flipkart’s services from the incumbent ‘Google’ and ‘Facebook’.

So, what are the implications for entrepreneurs?

First, evaluate if your market is indeed contestable, or is likely to become contestable. If there is a likelihood of your primary market being or becoming contestable, consider one of the following options:

Change your business/ business model (pivoting is a fancy word these days in the startup ecosystem)
Erect barriers to entry and exit – use regulation if you must (see how Airport Taxis in Bangalore are competing with OLA and Uber)
Differentiate – even if it is not the most significant of your product offerings, focus on those value creation opportunities that involve sunk costs
Wait for a new entrant and bleed him to death with limit pricing (you better have easier access to capital than the new entrant)!
Wind up, sell out, and take your family (if you have one) on a holiday to Seychelles! And don’t forget to thank me!

StoreKing: Taking ecommerce to rural India

Each of my visits to Europe has taught me something new over the past few years. My recent visit in April-May, I had to travel through three countries – Switzerland, Germany, and Italy. What struck me this time was how much the local language was used in a lot of business and commerce, with English being the common language. While looking for similarities between India and the European continent, I was amazed at how much they value their local languages. For instance, my colleagues in Germany did my hotel bookings for Nuremberg and Rome through Booking.com and HRS.com, and the entire communication cycle was in German language. Not surprising. But it triggered the thought about “why don’t we have websites and mobile apps in India’s languages?” What would be the impact of an ecommerce site in a local language like Kannada on a rural consumer in say, the Dakshin Kannada district?

I began my exploration and in a recent road trip to Tiruchchirappalli (Trichy for the phonetically challenged) in Tamil Nadu, I experienced the power of StoreKing. StoreKing is not a traditional retailer or an ecommerce firm. It leverages the power of ecommerce and solves the three major problems faced in rural penetration of ecommerce – language barriers, non-specific addresses, and trust. A detailed description of the StoreKing business model is available in a write-up on YourStory.in (read here), but for the sake of explanation, let me briefly introduce the same.

StoreKing approaches rural retailers (brick and mortar) and convinces them to install the StoreKing kiosk/ buy a StoreKing tablet in their shops. These kiosks or tablets are powered in the local language, and has a large variety of SKUs, ranging from electronics, appliances to digital goods like mobile/ DTH recharges. Customers walk in to the store, and with the help of their trusted retailer, browse and shop on the StoreKing kiosk. Once they have placed an order, they pay the retailer in full, StoreKing communicates with the customer through their mobile phones in their local language. The problem of poor (ill-specified) addresses is taken care of by dispatching the goods to the local retail shop (from their central warehouse in Bangalore) within 48 hours. The retailers receive a 6-10% commission on the sale proceeds. Though I am not sure how StoreKing sources the goods, it uses the standard FMCG distribution network to ship the products to the retailers.

StoreKing’s last-mile connectivity to its rural consumers addresses the three main problems faced by traditional ecommerce firms – lack of scale in rural markets to justify investments in delivery infrastructure, lack of sufficient data about rural consumer habits and preferences, and their (misplaced) perceptions about rural buying power. An older YourStory.in report talks about how StoreKing’s customers bought dishwashers (not one, but two for the same household) and iPhone 6s (read here). The lack of scale has been overcome by adopting a hub-and-spoke distribution system that piggy backs on the FMCG distribution network.

I am not sure this happens, but would it be possible for the customer to change the default language of communication? I appreciate that rural India would not have enough linguistic diversity to justify this, but if StoreKing were to penetrate into border towns like in Belagavi (nèe Belgaum) district, where multiple languages are used, it would definitely need customization.

StoreKing has partnered with Indian Oil petrol bunks (gas stations) as retailers (see here); as well as Amazon.in, presumably for expanding their breadth of products. The recent media reports talk about Amazon.in’s Udaan initiative to reach rural customers with limited internet connectivity, and the synergies Amazon.in will have through this partnership with StoreKing, but not the perspective of StoreKing. Amazon.in would leverage their deep local presence and established distribution network; and I would guess StoreKing would significantly benefit from Amazon.in’s breadth of products list.

StoreKing claims to be neither a discounter nor a premium seller of goods. The primary value proposition is the trust its customers have on the local retailer; and that has enabled them to even collect cash in advance, rather than cash-on-delivery that has become the dominant mode of ecommerce transactions in India. This trust placed by the retailers on StoreKing provides it with a significant first mover advantage. At over Rs.10,000 investment and significant local knowledge of the customers, the switching costs and multi-homing costs for the retailers are very high. Even when a competitor enters the market directly, it would be difficult to convince a retailer to shift out of the StoreKing kiosk/ tablet to another competing solution. It is here, that I believe StoreKing should follow the classic Wal-Mart strategy of “regional rural saturation”, and convince every retail shop/ kirana store in a particular geography to host a StoreKing kiosk.

Four questions pop up in my mind, for which I have no answers right now.

Should StoreKing open its own exclusive stores, as they grow big? What are the costs of signing up with competing retail stores in the same village? Can these costs be overcome by stand-alone StoreKing kiosks?
At the other extreme, should StoreKing allow for a tight integration of the brick-and-mortar retailers’ inventory and their inventory? For instance, if a customer ordered a Micromax mobile phone through the StoreKing kiosk, which was already available with the retail store in his physical inventory, should he fulfill it from his store (and forego the StoreKing sales commission) or block those items that he sells in his store?
If these brick and mortar stores who are trusted by the local customers offer discounts and credit for their offline sales, how does that affect StoreKing operations and business model? Should StoreKing allow a retailer to extend the same credit terms he offers to his customers for ordering good through StoreKing?
As StoreKing expands into more and more geographies (as of June 2016, they operate in the five South Indian States, plus Goa, Maharashtra, Gujarat, and Odisha), is this model scalable? What challenges would a market like Eastern Uttar Pradesh pose?

I am watching this firm and its growth trajectory from the outside. Any answers?

PS: I am nor in any way related to StoreKing or its investors/ founders.

Durability of network effects – importance of multi-homing costs

In their recent HBR article, David Evans and Richard Schmalensee argue that winner takes all thinking does not apply to the platform economy. In the article, they cite instances of how popular multi-sided platforms like Facebook, Google, and Twitter haven’t won every market. In fact, in spite of being near monopolies social networking, internet search, and micro-blogging, they compete very hard for the advertisement revenue. They also posit that network effects are not durable enough in the case of digital goods, as compared to physical networks like railroads and telephones. In this blog post, I am going to discuss these two assertions.

In the meantime, I ordered their book, Matchmakers, and my favorite ecommerce bookseller just delivered it to my desk, as I begin writing this blog. Will read the book in the coming week, and possibly update the note; but for now this blog post is based on their HBR piece. Now, if you have not read their HBR post, please read it.

Winner-takes-all markets

In their very popular HBR article Eisenmann, Parker, and Van Alstyne elucidate three conditions for a market to exhibit winner-takes-all (WTA) conditions. One, the network effects should be strong and positive; two, multi-homing costs should be high; and three, there should not exist any special needs by the users.

Network effects

In the case of the three multi-sided platforms that Evans and Schmalensee quote, the network effects are very strong. You signed up to Facebook because all your friends, family, and acquaintances were on Facebook (same side network effects); you use Google search because Google has learnt enough about you and only pushes “relevant” advertisements to you (cross-side network effects); and you micro-blog using Twitter because everyone who you want to reach are already looking for you at Twitter, as well as everyone who you want to follow are micro-blogging using Twitter (a combination of same and cross-side network effects).

Multi-homing costs

Multi-homing costs imply the costs of affiliating/ maintaining presence on multiple platforms at the same time. My most popular example is the case of internet-based email services. Even though it is literally free for anyone with an internet to have an unlimited number of email accounts, most of us cannot really maintain more than three email accounts. The monetary costs of creating and operating multiple email accounts may be zero, but the effort required to remember passwords, periodic logins to each of the accounts, and ensuring that you are communicating using the right email account is too much for most people. These are multi-homing costs.

Multi-homing costs exist in all the three markets we are discussing – social networking, internet search, and micro-blogging. In the case of social networking, it is difficult to maintain multi-home as the updates that we are likely to share in multiple networks are likely to be the same. And, the strong network effects (all my friends are on Facebook) make sure that there is virtually no-one else who is active in any other competing social networking site who is reading my updates. Multi-homing costs in internet search manifest in the form of the search engine’s ability to customise its advertisements and offers to my preferences and behaviour, which is based on my behaviour over time – with my past preferences, I have actually trained the search engine to customise. Search on the same key words across different internet search engines are unlikely to provide different results, but it is the overall experience including advertisements and personalisation that matters in the case of Google. This is somewhat similar to being loyal to a particular airline and gaining miles in that frequent flyer program; as splitting one’s travel across multiple airlines’ loyalty programs would ensure that one does not remain a frequent flyer anywhere! Similarly, having invested sufficiently in training Google on my personal preferences, I would rather stick with Google search. Similar is the argument for Twitter – the network of micro-bloggers and followers exist on Twitter; and I have carefully curated the list of which micro-bloggers I want to follow. Multi-homing costs include creating multiple lists of people I want to follow, and getting others to follow me.

Special preferences

The third condition for a market to exhibit winner-takes-all characteristics is the absence of any special preferences. Let us take the case of social networking – when professional networking and sharing of professional thoughts is a special need, different from social networking, LinkedIn thrives. Most people with a need to separate out their personal networks from the professional networks will maintain a Facebook account, as well as LinkedIn account. And, when a LinkedIn user turns into an active job seeker (from being a passive expert), she would open an account with a focused careers site like Monster.com. Similarly, someone’s work/ passion may require sharing large sized file attachments over email, and therefore push her to open multiple accounts for different kinds of uses.

In sum, winner-takes-all markets are characterised by the presence of strong network effects, high multi-homing costs, and the absence of any special needs. What Evans and Schmalensee ignore in their HBR post is the presence of high multi-homing costs. Yes, these firms do contest in the market for advertising revenues, but in one side of their respective markets, their strategies have been to continuously raise multi-homing costs. Take Facebook’s acquisition of WhatsApp for example. When more and more people took to social networking using a mobile phone than the ubiquitous desktop, and were increasingly constraining the breadth of audience for their posts, it was important for Facebook to be present on the users’ mobile phones, not just enabling broadcast social networking (with its Facebook mobile App), but also including narrowcasting or unicasting social networking using WhatsApp. Same is with Google – over the years, Google has come to dominate the internet search in more ways than one – YouTube and Maps to name a few.

Durability of network effects

The second thesis of Evans and Schmalensee is that network effects in multi-sided platforms are not durable. They cite how easy for a new entrant to challenge these leaders with little or no physical investments. Digital goods like software have high fixed costs and almost zero marginal costs for every additional unit produced. Economics has taught us that in markets with near-zero marginal costs, prices will fall continuously to eventually make the product free. There are a variety of other goods where such cost structures prevail. Take for instance, news media. The cost of replicating (or is it plagiarising) a news article across multiple outlets is close to zero, and therefore news producers are under tremendous pressure from consumers to respond to the threat of potential new entrants to provide news at prices cheaper than free. Yes, cheaper than free, which means that you may actually be paid to consume news! Like what Google did to the handset makers to use its mobile OS (for more details, read here). In the initial days of building the platform, firms are under severe pressure to kick-in network effects, and adopt pricing strategies that are cheaper than free. For instance, the Indian cab aggregator OLA Cabs, incentivises drivers handsomely (as the markets mature, the incentive rates are falling) to undertake a certain number of rides per day. This is apart from the amounts they earn from the passengers. In the entire bargain, drivers get paid by both the riders and the aggregator, and OLA keeps the rider fare low to encourage more usage, leading to faster growth of network effects.

Evans and Schmalensee argue that faster the network effects grow, faster they will disappear. I contend that this may not be true in markets with higher multi-homing costs. Take the OLA Cabs business model for instance. At the rider’s side, there are no significant multi-homing costs; at best it is limited the real estate available for multiple apps on the rider’s smartphone. It is the drivers’ multi-homing costs that are of interest here. OLA Cabs and its primary competitor Uber, have been working hard on increasing the driver’s multi-homing costs by limiting the incentive payouts only when the driver completes a certain number of rides per day. And as the market grew, this number of rides required to earn incentives has risen sharply. That means, a multi-homing driver has to ensure that he completes at least the minimum number of rides on one of the aggregator platforms before accepting rides on another. And soon, drivers who cannot meet the minimum required for earning incentives on both platforms would choose one of the two, and those drivers who cannot even meet the requirements of one aggregator would leave the market. Even though the cost structure of cab aggregation is similar to digital goods (high fixed sunk costs incurred upfront) and close to zero marginal cost of adding a new driver/ cab to the fleet, these firms have sustained the winner-takes-all characteristics by increasing the multi-homing costs of the drivers.

To sum up, network effects are durable when the platforms invest in increasing multi-homing costs of at least one side of the platform. Better so, the money side (not the subsidy side) that has the highest switching costs. These multi-homing costs arise out of asset-specific investments that the participants make in affiliation with the platform. In the case of OLA Cabs, multi-homing costs do not arise out of having to carry multiple devices, but in ensuring minimum number of rides per day on a particular platform to earn incentives. And these incentives are significant proportion of the drivers’ earnings, as the aggregators keep the rider prices low.

The importance of multi-homing costs

Evans and Schmalensee write:

With low entry costs, trivial sunk capital, easy switching by consumers, and disruptive innovation showing no signs of tapering off, every internet-based business faces risk, even if it has temporarily achieved winner-takes-all status. The ones most at risk in our view are the ones that depend on advertising, because even if they dominate some method of delivering ads, they are competing with everyone who has or can develop a different method.

In this post, I argue that creation and maintenance of high multi-homing costs is an effective insurance against low entry costs, trivial sunk capital and easy switching by consumers. Fighting disruptive innovation requires platform firms to understand the economics of envelopment, which we will discuss next week.

Cheers

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this: