The one startup that went up against Ben Thompson’s all-encompassing aggregation theory

And lived to tell the tale.

There is no escaping Ben Thompson’s Aggregation Theory in today’s world.

Ben Thompson of Stratechery, perhaps the world’s most sought after technology analyst, first introduced the term ‘Aggregation Theory’, in his July 2015 post. The theory describes how a significant number of the most influential technology companies of the past two decades have become world dominators by effectively commoditizing distribution, and instead aggregating suppliers by directly focusing on the end-user experience.

Before the internet, distributors were the kingmakers in any industry due to the fact that distribution of any good or service from the makers/suppliers to global consumers that were willing to pay for said good or service was a particularly complex task, and hence the most profitable part of the value chain.

But the internet has effectively made digital distribution of goods and services free, thus commoditizing the distributor’s role in the value chain. ‘Makers’ can now sell their goods and services direct to consumers.

“This means that the most important factor determining success is the user experience: the best aggregators win by providing the best experience, which earns them the most consumers/users, which attracts the most suppliers, which enhances the user experience in a virtuous cycle.”

This means that distributors are no longer king. Aggregators are.

The following graph from statistica shows the 10 most valuable public consumer internet companies by market cap as of 2016.

The singular commonality among all the companies above (with over a trillion dollars in cumulative value) is the fact that every single one of them is an online aggregator.

Every single one.

These companies don’t create the goods, services and content that they surface to consumers for the most part; their domination comes from providing the best end-user experience to consumers and making them stick, so much so that suppliers have no choice but to partner with these aggregators for distribution.

---

Zomato, started in 2008, is perhaps India’s finest and most well-known example of the Aggregation theory in action. Zomato started off by indexing information (menus) from virtually every restaurant in India on its online platform, organising said information and then making it easily searchable to become India’s ‘Google for food’.

Having effectively established a monopoly on the restaurant search and discovery market in the country with more than 12 million monthly unique visitors, Deepinder Goyal (CEO of Zomato) then made his next master move that would entrench users even more firmly onto the product.

Zomato laddered-up from being just a tool for users to a network, encouraging users to start checking-in, click photos and review restaurants on the platform.

This content then became the primary source of value for entirely new sets of users, driving both user acquisition and retention to drive a virtuous cycle, widening Zomato’s pre-existing moat. Even better, this was exclusive, valuable content that was being aggregated by the network for free, as the suppliers were its own users.

Now the service knew exactly what you liked, where you ate, who you went out with, all funnily enough because you told it yourself. This data helped Zomato personalise the service even more, improving the experience.

Each customer that joined the site and left a review made the next customer’s experience better by pushing good restaurants to the top. Additionally, since customers could now rate their experiences on a public platform, restaurants had an increased incentive to provide better service to customers, making the entire network’s experience even better.

This virtuous cycle meant that that Zomato enjoyed a rich-getting-richer effect, as it captured more and more users; attracting even more and more restaurants; effectively monopolising both sides of the network.

As if being the ‘Google for Food’ in India was any less of a moat, it had now became the Facebook (of food) as well.

Having become the leading aggregator in the restaurant category, the stage was now set for Zomato to capture a market almost as big as the e-commerce market in India: the market for food-delivery.

In 2015, Zomato launched Zomato Order: it’s on-demand food delivery service.

Zomato’s market-dominating restaurant search and discovery business, fueled traffic to their new transaction business at no additional (customer acquisition) cost — an unfair advantage over competition.

Zomato Order’s main selling point to restaurants was to bring in lots of new orders and replace their antiquated phone-ordering system with an optimised web/mobile platform, that integrated directly with their kitchen workflow.

Order depended on the restaurant’s own delivery fleet for a majority of food orders (around 80%) and third-party delivery partners for the remaining ones.

As they don’t touch the food itself (neither cooking, nor delivering it), and are primarily a lead-generation business, Order tends to charge a commission of 10–15%. But more importantly, as any pure-play software business, this model is highly scalable.

2014 and 2015 also saw a huge influx of VC money (more than $300 million) into India’s food-tech business as scores of food-delivery startups sprung up throughout the country, hoping to capture part of the multi-billion dollar opportunity in sight.

But while every other food-tech player struggled (read burnt money) to incentivise users to come onto their platform with discounts, Zomato Order chugged on unperturbed, powered by traffic from the mothership.

In the two years that followed, just as Deepinder had predicted, a food-apocalypse hit India as every other discount-led food-tech player in India fell by the wayside.

It seemed inevitable that Zomato Order would soon take the throne that was rightfully theirs: that of India’s biggest food-tech player.

Right?

Right?

Wrong.

In-spite of using the powerful network effects of the aggregation theory to lock-up two ends of the marketplace, Zomato Order is today second to the biggest player in India in terms of overall volume of orders: Swiggy.

How did this even happen?

The Food-tech ecosystem:

The value chain for any given consumer market is divided into three parts: suppliers, distributors, and consumers.

In the food-delivery space, these are the restaurants, delivery guys and the end-users.

It’s important to understand that the real challenge for food delivery platforms does not lie in the supply. Any aggregator that promises restaurants additional orders is a no-brainer for a business that has high fixed costs and low variable costs: every additional order goes straight to the restaurant’s bottom line and improves the margins.

Rather, it is the demand side of the marketplace that is the real challenge and where all aggregators build their moat.

The only way to attract a critical mass of users onto your platform is by offering a differentiated and superior user experience.

While a superior user-experience might be directly proportional to the superiority of the user interface for most internet businesses, in a transaction business the look-and-feel and interface of a product (where Zomato Order beats all competitors hands down) are purely hygiene factors.

Rather, it is things like faster delivery and lower prices, beyond the superficiality of what an app looks like, that impacts end-user experience far more directly.

Swiggy knew that offering lower (discounted) prices wasn’t a sustainable strategy in the longer term as losses would balloon with volume. More importantly, discounts do not lock-in users to the platform in any way meaningful enough to build a moat around.

But Zomato had not closed the food-delivery loop in its entirety: a gap through which it thought no one in their right minds would ever enter through.

“The best startups are good ideas that look like bad ideas. Good ideas that look like good ideas are already being worked on by big companies.” — Peter Thiel

The only way to offer any meaningful differentiation in the food-delivery space meant getting into the one part of the value chain untouched by Zomato: the messy and supposedly unprofitable business of logistics and last-mile delivery.

Owning delivery would allow Swiggy to assure users of faster delivery times, and hence a much better experience than what Zomato Order could offer.

And sure enough, selection of restaurants and pricing being equal, delivery times have the highest impact on user experience in food-delivery.

Giving users a superior experience lead to massive word-of-mouth growth for Swiggy, pulling in more and more users onto its platform. The more users it got, the more leverage it had to bring on more restaurants onto the platform and thus spur even more transactions, sparking it’s very own Amazonian flywheel.

More transaction volumes lead to much better utilisation and optimisation of the delivery fleet, faster delivery speeds, lower delivery costs and better cash flows for the platform.

Higher volumes also allowed the delivery guys to do more deliveries, increasing their earnings (who are paid mostly in incentives based on number of deliveries). This allowed the platform to hire more delivery guys which meant better service for users.

Lower delivery costs and faster deliveries lead to an even better user experience, bringing in ever more users, leading to an ever-accelerating cycle where each part of the value chain, acts both as an accelerant and a beneficiary.

Indeed, the much-maligned ‘delivery’ business acted as a loss-leader initially to spur the flywheel into motion. Now one could feed any part of the flywheel, immediately accelerating the whole loop. The concept, made famous by Bezos who made it the core principle behind Amazon’s operations strategy, was actually borrowed from strategy guru Jim Collins, and isn’t just unique to the Seattle behemoth.

Network effects work against you until you reach critical mass; then they work for you.

As Swiggy scales, the data the platform has gathered on past deliveries will allow it to keep optimising routes and pick-up/drop-off patterns, giving them an unassailable technological advantage, on top of the moat being ever-widened by the network effects of running a three-sided marketplace.

https://twitter.com/jmj/status/837044495262736384

This will allow Swiggy to predict delivery ETAs much more efficiently, offer ordering without any minimum order fee (which it already does) and charge lower fees for delivery than any individual restaurant can afford to offer.

Once the food-delivery business tips, Swiggy could even open up the delivery end as a platform.

In fact, this food-delivery war has much in parallel to a war that took place more than a decade back, as the upstart Amazon battled the then leaders eBay in the e-commerce arena.

“Wilke’s gradual success in making the logistics network more efficient would offer Amazon innumerable advantages in the years ahead. Tightly controlling distribution allowed the company to optimise deliveries and make specific promises to customers on when to expect their purchases to arrive, giving it a competitive edge over rivals, especially eBay, which had avoided this part of the business altogether. Owning the end-to-end technology, right from website to the supply chain, allowed Amazon to effectively bring down delivery times to 4 hours, when the ecommerce industry average was 12 hours. Fulfillment was the lever that Bezos had invested in, and he started using it to guide strategy.” — The Everything Store - Jeff Bezos and the age of Amazon

If anything, delivery times are even more important in food-delivery than e-commerce.

Domino’s had to understand that it is not in the pizza-making business, but in the pizza-delivery business to make profits. — J. Patrick Doyle, CEO, Domino’s Pizza

In fact, providing a superior user experience and owning the users is even more vital in an undifferentiated experience like food delivery.

“This may seem counterintuitive, but in fact in markets where:

Purchases are habitual and high frequency

Prices are similar

Products are not highly differentiated

Customers tend to build allegiance to a brand and persist with that brand unless they are given a good reason to change; it’s simply not worth the time and effort to constantly compare services at the moment of purchase.”

Once customers sign up, they rarely leave for another platform, creating a strong winner-take-all dynamic.

As more and more users hop on to Swiggy, sparking additional transactions on the platform, restaurant liquidity/availability will also swing in favour of Swiggy.

And market share in any transactional platform business will always be increasingly cornered by the side having more liquidity.

One might argue that a restaurant might work with both a Swiggy and a Zomato, but that doesn’t really matter. What would matter is liquidity which will increasingly be monopolized by the dominant service. Ben argues that in such markets, while suppliers are not exclusive in theory, they are exclusive in reality.

He explains this clearly in the context of the ride-sharing market.

It doesn’t matter that drivers may work for both Uber and Lyft. If the majority of the ride requests are coming from Uber, they are going to be taking a significantly greater percentage of driver time, and every minute a driver spends on a rider job is a minute that driver is unavailable to the other service. Moreover, this monopolization of driver time accelerates as one platform becomes ever more popular with riders. Unless there is a massive supply of drivers, it is very difficult for the 2nd-place car service to ever get its liquidity to the same level as the market leader (much less the 3rd or 4th entrants in a market).

Restaurants then, with limited availability of cooks, equipment and time, will naturally congregate towards the dominant platform, that drives the most orders. Every minute a restaurant spends on servicing a Swiggy order is a minute unavailable to a Zomato order and hence adds an additional minute of delay to a Zomato customer waiting for the food, again pushing users onto Swiggy.

The implication is a principle that anyone operating in an industry influenced by network effects should understand: in the long run, there will tend to be fewer players, and they will continue to grow ever larger.

Another common argument against Swiggy is that while Zomato is a pure-play software business and hence, much easier to scale, Swiggy being both a software and logistics play isn’t as easily scalable to multiple markets.

While this is true, there are a couple of powerful factors that do work in Swiggy’s favour:

1. Perfecting the playbook: While Swiggy’s network effects don’t naturally translate to new markets as would of a pure internet business; the technology, algorithms and playbook that has been iterated on and perfected in the initial cities can be taken to and applied in entirely new cities.

2. Driver network effect: While coordinating hundreds of delivery guys on the road in real -time is a massive operational headache, it is also a unique asset. With their branded tees and delivery boxes, Swiggy delivery guys act as highly-visible advertising panels on wheels, whose reach increases as the company grows, creating a second and powerful type of network effect.

So while these on-demand marketplaces may not be as easy to scale as the pure-software one, it will be very hard for a new entrant to compete against these optimized networks of restaurants and delivery guys, once they reach maturity in a city. They can also charge a higher commission of 25–30% on average vs the 10–15% of a pure software player.

Opening up a whole new market

Don’t forget that the biggest margin in food is inside the kitchen. This is what funds everything else in the restaurant. As soon as you step out of the kitchen, be it for dining or delivering, the restaurant margins go down.

In the near future, I see on-demand marketplaces like Swiggy offering a powerful enough distribution infrastructure (through both very large customer bases and efficient nation-wide delivery networks) to spare emerging chefs the hassle and high upfront investment of launching a physical restaurant.

Not needing a physical storefront could enable these new chefs to cut fixed costs (no more waiting staff, cleaners, rents, furniture depreciation, etc.) and drastically lower the price of food via delivery vs. in-restaurant, and hence grow the overall market.

They can instead test new concepts on the cheap in standalone kitchens, with Swiggy acting as the discovery/distribution channel — a sort of Appstore for food.

---

Someone famous once said:

The three most important things that matter for a restaurant are: location, location and location.

Just as Airbnb commoditized the trust advantage that hotels had, and Netflix commoditized time, what these new-age food-delivery platforms have done is essentially commoditized the location and brand-name advantage that big chains had.

This means that the real-world benefits of owning a great location have now been translated online into an abundance of relevant page or app views on a location now owned by these food-delivery platforms.

There is no doubt that Zomato has built one of India’s most-loved consumer brands, and one of it’s most well-thought out products to date; it is undoubtedly the winner in the dining out and nightlife categories. Indeed, that race has already been run.

However, I believe that the better business model will always eat the better product for breakfast, and even as India’s largest offline sector moves online in the coming years, my money’s where my mouth is: on Swiggy.

---

The Aggregation theory is one of the most important frameworks in tech today, governing the rise of some of the most significant internet companies in the history of the world.

Swiggy’s extraordinary rise, however sort of adds a corollary to this theory: in industries where the final good or service still has to be physically delivered to the end user, owning and controlling distribution is absolutely a core competency, a vital part of the flywheel, and not just a commodity.

Companies in these industries would do well to understand that delivering the best user experience is dependent on owning the actual distribution part of the value chain, and not just the interface between the supplier and the user, as is the case for most firms in the post-internet era.

Disclosure: Swiggy and Zomato are both customers of Freshdesk, the global leader in customer support, where I currently work.