Mar'21
Case Study - Swiggy's Delivery Model
Shubhanjali Chakravarty
Former Research Associate, IBS Hyderabad (Under IFHE - A Deemed to be University
u/s 3 of the UGC Act, 1956), Hyderabad, Telangana, India. E-mail: shubhanjali@icmrindia.org
Sanjib Dutta
Research Lead, Case Research Center, IBS Hyderabad (Under IFHE - A Deemed to be University
u/s 3 of the UGC Act, 1956), Hyderabad, Telangana, India. E-mail: sanjibdutta@icmrindia.org
The case study "Swiggy's Delivery Model" discusses how online food delivery
company Swiggy used technology to power its delivery model. Swiggy was the
brainchild of two engineers who were staying away from home and were struggling
to get good food delivered at their doorstep. The duo worked on the idea of hasslefree
food delivery, giving rise to Swiggy in 2014. Swiggy initially started taking
online orders from customers in the Koramangala locality in Bengaluru which
housed some of the best restaurants in the city. It gradually expanded across
Bengaluru and other Indian cities like Delhi, Pune, Gurgaon, Hyderabad, and more.
It launched its mobile application in May 2015. The case describes how Swiggy
created a system that connected all the three partners in the delivery model - Delivery
Executives (DEs), customers, and restaurant partners - seamlessly. The case also
discusses the challenges Swiggy encountered when it began scaling up in 2016 and
the initiatives such as Just in Time (JIT) assignment, Next Order Assignment (NOA),
and batching that it took to overcome the challenges. The case concludes with
Swiggy's plans for the future such as diversifying into medicines and grocery and
the use of Artificial Intelligence (AI) technology for further scaling up and expansion.
Swiggy's goal is to revolutionize the food delivery market in India by providing a
cheap, fast and efficient alternative to taxing the already short-staffed restaurateurs.
There are several existing and even more upcoming food-ordering platforms in the
country. However, Swiggy sets itself apart from the competitors by having their own
delivery fleet, thereby minimizing the risks in delivering an order.1
- Sriharsha Majety (Majety), Co-Founder and CEO of Swiggy, in 2019
Introduction
On December 20, 2018, India-based online food delivery company Swiggy raised $1bn in the
largest-ever funding round by an Indian food tech startup. Post that round, Swiggy's valuation
went up to $3.3 bn.2
Since its inception in August 2014, Swiggy had focused on creating a robust delivery
model consisting of its own fleet of Delivery Executives (DEs). Swiggy's founders realized
early on that to deliver the best on-demand consumer experience, Swiggy had to have its own
delivery model in place rather than depend upon a third party delivery partner. Swiggy aimed
to create a robust delivery model initially through its website and subsequently through the
launch of its mobile application in 2015.
However, Swiggy faced challenges in delivering on time when the company began
scaling up in 2016. As a response, it recreated its delivery model by bringing in the Just
in Time (JIT) assignment method (to ensure timely delivery of the order), Next Order
Assignment (NOA) (to reduce the time taken by DEs at the restaurant waiting for the order),
and batching (for increasing the efficiency of a DE by getting him to deliver multiple orders
in less time).
As of 2019, Swiggy was focusing more on using digital technologies to enhance its
delivery model. It invested in Artificial Intelligence (AI) technology to enhance its delivery
capacity. Swiggy also planned to invest in groceries and medicines. It remains to be seen how
the company would defend its position in the online food delivery market, which is projected
to reach $161.74 bn by 20233, as new players entered the market.
A Brief History of Swiggy
In 2012, Majety, a BITS Pilani and IIM Calcutta alumnus who worked as a trader at an
investment bank in the UK, met Nandan Reddy (Reddy), another BITS Pilani alumnus. Reddy
was working as a consultant and had built a startup "Galla" around a tablet-based Point-of-
Sale (POS)4 for restaurants after heading operations of SourcePilani, India's first rural Business
Process Outsourcing (BPO).5
During 2012-2013, both Majety and Reddy came together to build a platform called
"Bundl", a logistics aggregator for e-commerce companies. However, things did not work out
quite the way Majety and Reddy wanted and Bundl was shut down in June 2014. But,
working on Bundl and tracking the logistics space made both realize the potential of
hyperlocal delivery in the Indian metros. After the shutdown of Bundl, Majety and Reddy
planned to get into hyperlocal delivery.
Both Majety and Reddy stayed away from home and had to order food from restaurants
on an everyday basis. There were no players offering a complete food-ordering and delivery
solution in the segment apart from Zomato and Foodpanda.com (Foodpanda), but they too
depended upon their delivery partners (restaurants) to deliver food to their customers. Majety
and Reddy were now convinced that they needed to work on the idea of hassle-free food delivery. Majety said, "When already we decided to shut Bundl, we knew we have to dive
into hyperlocal delivery and give it a shot. We also understood that ordering platforms for
restaurants were very broken. Ordering food has always been a big hassle, and hence, we
decided to take the plunge and fix it"6
Rahul Jaimini (Jaimini), an IIT-Kharagpur alumnus who had left his job as a developer at
Myntra Logistics (Myntra) and was eyeing entrepreneurial opportunities, also joined Majety and
Reddy. Together, they named their venture, "Swiggy" (meaning, unbelievably legit and real), a
food ordering and delivery company based out of Koramangala, Bengaluru's startup hub. Swiggy
started its operations in August 2014 by signing up with a few restaurants in Koramangala. It raised
its first round of funding in May 2015, and secured $2 mn from venture financing companies,
Accel7, and SAIF Partners.8, 9 The same month, it came up with its mobile application, which was
designed as an on-demand food ordering and delivery platform through which customers could
place food orders from their preferred restaurant(s), and get the orders delivered within 40 min.10
Swiggy was operating in 11 neighborhoods in Bengaluru by May 2015 and it spread out all over
the city over next two months. It had tied up with more than 300 restaurants and was delivering
1,000 plus orders on an everyday basis.11 From July 2015, Swiggy started expanding across India,
covering Pune, Mumbai, Gurgaon, Delhi, and Hyderabad.12
In 2015, Swiggy had 650 DEs from diverse backgrounds-college students, who worked
part-time and full-time employees. By the end of 2015, it had hired over 12,000 DEs. Swiggy
developed its technology keeping in mind the "changing requirements based on geography,
logistics, etc."13 The technology was developed entirely in-house.
As of June 2019, Swiggy connected consumers to over 100,000 restaurant partners across
over 200 cities in India. By using technology, it aimed to provide a hassle-free, fast, and
reliable delivery experience. Every order delivered by Swiggy's 0.195 million active delivery
fleet, the largest in India, ensured quick delivery, no minimum order value, live order tracking,
and 24/7 customer support.14
Swiggy's Delivery Model Powered by Technology
In order to ensure that food was delivered on time, Swiggy introduced live order tracking. Like
other on-demand delivery startups, Swiggy too integrated the Google Maps application
program interface15 (API) which let the customers know where their order was and how much time it would take for the order to be delivered. That allowed them to track the DE's current
location with accuracy.16
Swiggy aimed to offer an integrated ordering solution by deploying multiple proprietary
applications, including a vendor management application provided to restaurant partners and
a delivery application powered by routing algorithms, provided to driver partners. When a
consumer placed an order with Swiggy through the iOS or Android applications or website,
the order was immediately transmitted to the merchant and the DE, based on availability and
distance from the restaurant.
The merchant confirmed receipt of the order and prepared the food, while the DE reached
the restaurant JIT to collect the prepared food and deliver it to the customer. Swiggy's in-app
navigation was made to facilitate hassle-free delivery. According to Jaimini, "The technological
backbone of Swiggy, the start-up's delivery fleet is powered by algorithm-enabled smartphones
that are constantly enforcing more efficient ways to complete deliveries."17
Swiggy concentrated on establishing coordination between the consumers, restaurants, and
the DEs. This triad constituted a separate marketplace in a hyper-local setting for Swiggy
(refer to Figure 1 for interconnection in the market place). That meant that at every given
location in which it operated, there had to be a critical mass of consumers, restaurants, and
DEs at all times to deliver an optimal experience. The demand from consumers also needed
to match with the supply of restaurants available in the area to enable delivery within a
reasonable time. And that required the management to maintain real-time accurate
information.
The Swiggy application for the customers was designed to offer a simple interface and a
hassle-free customer experience. Once the application was installed and the customer signed
in, various restaurants were listed, to help the customer decide where to order from and what
(refer to Figure 2 for the design of Swiggy's mobile application). As Swiggy was a real-time
platform (the actual time during which a process or event occurs), the application opened with
thousands of options instantaneously like customers' food preferences, the restaurants present
in the customers' area, different cuisines, and the ranges across all the price points. The
application also provided a live snapshot of the DE in the customers' area.18
After the customer placed an order and the restaurant had confirmed it, the assignment
algorithm looked into factors like assigning a DE to an order accordingly and the best route
for him to follow. A broadcast signal was then sent to all the drivers in the vicinity who had
their own driver application. Those willing to accept the order could choose to accept it on
the application.
Swiggy's communication engine was a decentralized system. That pattern worked well
when the teams were small but while scaling up, Swiggy encountered difficulties in its
communication channels relating to uncertainties in dish-preparation times, the DE's
familiarity with the area and speed, and changing traffic conditions, to name a few. Hence,
Swiggy's founders felt the need for a centralized platform to cater to the needs of the large
teams in the company.
With the help of Big Data analytics and software developers, Swiggy resolved its pain
points and built a system as a rational business communication engine, which helped users
to use a "rendering engine to build templates and send communications to users via multiple
channels"19. The system helped track open rates and Click Through Actions (CTAs) and
conversions, but the system worked independently without depending upon other programs.20
A software developer at Swiggy said, "This seemed like a good idea, as the communication
engine will become an isolated system with no dependency on any other system. This system
was designed to be a push based system, where other services will send an event to the
platform, which will then send it ahead. The whole process will be asynchronous with users
having an option to either provide a webhook or a queue (topic) if they want to know when
communication was routed if there is a high requirement for consistency."21
Business Model and Expansion
Swiggy earned some part of its revenues from commissions. It charged a 15%-25% commission
on the order bill amount received by the restaurant. This commission was calculated on the full
bill amount which was inclusive of the Goods and Service Tax22 (GST) charged over and above
the menu price. However, the company often aimed to get restaurants to be available on Swiggy
exclusively. For this, it gave certain benefits to the restaurants like greater visibility (with the
restaurants appearing on the application more number of times) and its commissions sometimes
being cut by 2-3% (refer to Figure 3 for business model of Swiggy).
Swiggy did not have a minimum order requirement for delivery. That meant it often received
orders for less than 100. That increased the logistics cost per order. However, the company
started charging delivery fees after 2016 which depended upon the geographical location of the
restaurants and the cities they were located in. Delivery charges typically amounted to around
20 for orders less than 250. Swiggy also introduced surge pricing in 2016 in times of high
demand, rains, and special occasions and midnight delivery in select markets. It also earned from
advertising, banner promotions (promotion of restaurants in its application and website
pertaining to regions), and priority listing of restaurants. Swiggy tapped its revenue stream by
charging a premium from restaurants in return for giving priority listing to them. The higher up
the list the restaurant was displayed, the higher was the fee the restaurant had to pay.
Swiggy also earned through "Affiliate Income" by referring credit cards to its customers.
The company partnered with financial institutions and banks like American Express, Citibank,
ICICI Bank, HSBC, etc. to sell their credit cards to the customers.
In January 2017, Swiggy launched its own cloud kitchen service23 in Bengaluru named
"The Bowl Company", which offered a limited menu. In March 2017, Swiggy's services were
expanded to Delhi-NCR, including Faridabad, Noida Expressway, and Greater Noida. In April
2017, it introduced post-midnight delivery till 2 a.m. in Delhi-NCR, and Hyderabad. In July
2017, Swiggy marked five million monthly orders. In November 2017, it launched "Swiggy
Access", a central kitchen base-like facility which housed kitchens of different restaurants
including Swiggy's private labels, "House of Dabbas" and "Punjabi Rasoi" in Bengaluru.
Swiggy Access was its first virtual kitchen facility for existing restaurant partners, which
served as a central base for delivery with five kitchens of different restaurants. As of 2019,
Swiggy Access was available in four metros in the country with kitchens in Delhi, Mumbai, Kolkata and Hyderabad and Swiggy planned to on-board new restaurants in over 30 cities by
2020. As of 2019, about 8%-25% of the order volumes at Swiggy came through its Access
kitchens.
Swiggy also launched "Swiggy Pop" in 2017 to offer food delivery to singles. As on 2019,
Swiggy Pop was available in seven major Indian cities - Hyderabad, Bengaluru, Chennai,
Mumbai, Pune, Kolkata and Delhi.24 In February 2018, Swiggy launched 'Swiggy Scheduled',
which allowed users to plan and order their meals in advance. In July 2018, Swiggy rolled
out its subscription model "Swiggy Super" with a monthly and quarterly plan priced between
99 and 149. Swiggy Super offered unlimited free delivery on orders above 99 with no surge
pricing applicable if a customer had already opted for Swiggy Super.
Delivery Challenges
While creating an exhaustive and hassle-free delivery model, Swiggy faced challenges on a
daily basis from the DEs and restaurant partners. The first kind of challenge for Swiggy was
to fix the problem of restaurant listing and the delivery time promise. Whenever a customer
opened the Swiggy application, a query was sent to the delivery system regarding the
serviceability of restaurants to the customer and the expected time of the order from the
potential restaurant. But the expected time was difficult for many of the restaurants to estimate
due to their geographical locations.
The second challenge pertained to the feasibility of finding all the restaurants such that
a DE could reach there, pick up the order and deliver it to the customer on time. Swiggy faced
difficulties in assignment delay, first mile (time required to arrive at the restaurant),
preparation time (time required to prepare the food), and last mile (time taken to reach the
customer). The third challenge was the central delivery equation or the dilemma in finding
just the right length of time to ensure delivery, which would not be so long that it would
discourage the customer from placing an order and also providing an ample choice of
restaurants to the customer so that the delivery time remained as quick as possible.25
Cracking the Delivery Challenges
However, Swiggy cracked the delivery challenges and brought changes in its existing
environment by taking certain measures. To ascertain and resolve the problem of delivery
timing, Swiggy brought in the technique of JIT Assignment (a way to minimize the time spent
by the DE at the restaurant waiting for the food to be prepared"26). Through JIT, Swiggy
earmarked the DE available nearest the restaurant to pick up the order and service its
customers. The engineers at Swiggy came out with a new model. Under this model, when the
order was received, rather than dispatching the DE immediately to the restaurant, Swiggy
earmarked him for the delivery (according to Swiggy executives, the preparation of food took
longer than the time it took for a DE to reach the restaurant, which meant the DE would waste his time waiting at the restaurant to receive the food). Likewise, by allotting a DE on a JIT
basis, Swiggy took care of the timely delivery of the order and sent the DE to pick up multiple
orders in one go, thus ensuring timely deliveries. And Swiggy could use one DE for multiple
orders.
However, Swiggy executives faced challenges in earmarking and execution during peak
time loads (like during office hours or in the evenings). To resolve that issue, Swiggy adopted
the technique of NOA (a way to minimize the time spent by the DE waiting to be assigned
to the next order and increasing the potential pool of executives considered for an order27).
For instance, if a DE could deliver his first order (O1) within 10 min of picking up the order
from the restaurant and was 5 min away from the restaurant preparing the second order (O2)
(where estimated preparation time was 20 min), the DE could clearly finish delivering O1 and
then pick up O2 in time. An analyst at Swiggy opined, "However, while this is a great way
to increase the pool of eligible executives and at the same time reduce the wait times at the
Restaurant, it largely depends on the accuracy of the prediction for when the executive will
get free. Because, if the DE takes longer than what was estimated to get free, then the order
is unacceptably delayed."28
Swiggy also adopted, "Batching" (A way to increase the efficiency of DEs by delivering
more than one order at a time29). If a restaurant received two orders at the same time from two
customers located near each other, Swiggy dispatched a single DE to deliver both orders. That
helped increase the number of orders significantly and ensured delivery with a fixed set of
DEs. Devansh Gupta, executive at the delivery team at Swiggy, said, "Cracking the food
delivery challenge, not only requires us to come up with creative solutions to hard problems,
but also requires us to get an in-depth understanding of how the entire ecosystem behaves
in the real world, to ensure we make the right trade-offs".30
In 2018, Swiggy launched a program, "Swiggy Packaging Assist' to enable packaging
solutions on its platform as a move to diversify its play in the food technology sector. The
initiative was taken to offer restaurants access to a variety of packing solutions for their menu
needs with about 30 products to begin with. The marketplace connected restaurants with
vendors of these materials and offered eco-friendly and food grade certified materials
including leak-proof, sturdy, stackable, eco-friendly and heat insulant packaging materials.
Vendors on the marketplace offered discounts of up to 5% to restaurants opting for these
solutions.31
Business Performance and Competition
In the FY 2018, Swiggy's revenues rose to 4.42 bn up from 1.33 bn in FY 2017, as it
continuously expanded its geographic footprint across the country. Swiggy had expanded its operations from seven cities in March 2017 to 12 by March 2018, even as order volumes had
increased by three times during the period. Swiggy ended September 2017 with more than
21 million orders in absolute terms.32
However, amidst the growing revenues, Swiggy also increased its losses manifold to 3.97
bn in FY 2018 up from 2.05 bn in FY 2017, as it had to spend heavily on funding
discounting and low-cost deliveries.33 In 2018, Swiggy led the online food delivery market
in India with a 35-38% market share, followed by Zomato at 25-30%, according to a report
from Red Seer Consulting, an Internet-focused consulting firm in India.34
According to company officials, in 2017-18, Swiggy "focused on delivering exceptional
value to users, while investing significantly for the future and improving operational
efficiencies".35
Majety further opined, "We are being fairly measured about our own
dependence on these things (discounting) to drive business and its super important to keep
building a real long-term, sustainable business. We will react in ways to be competitive, but
it's not our focus area to figure out ways to spend on discounts. For us, it is well under control
and probably a few quarters of such behavior is to be expected because of new players who
are trying to win market share suddenly."36
Swiggy's biggest rival was Zomato. Uber Eats and Ola's Foodpanda had not been
expanding due to heavy losses and the cash burnt by them respectively. And both the
companies had rolled back investments in their food businesses.37
Other companies like
TinyOwl, Dazo, Spoonjoy, Eatlo, and EatOnGo had shut down operations after incurring
heavy losses. That made Zomato the biggest competitor of Swiggy as of 2019 with revenues
worth 4.46 bn, as of FY2018.
Swiggy had started delivering food (using its own fleet) since its inception in 2014,
whereas Zomato started its own delivery fleet a little late in February 2015 (after Swiggy's
launch). It continued to depend upon its restaurant partners (for food delivery), from its
inception in 2008 till February 2015. Changing rules, delayed deliveries, problems with the
restaurant partners and problems with training and retaining staff were the issues that affected
Zomato's growth and prompted it to go in for its own fleet like Swiggy.
Swiggy had been catching up fast with Zomato in terms of revenue and expansion (refer
to Figure 4 for revenue growth of Zomato and Swiggy). According to Majety, "We do our
basic service really well. We are boringly predictable. Our hope is there should be no touch
points between a customer and us beyond them placing the order on the app and us handing
over the food to them. It is a challenging business and we were able to get the basic service
right through a mix of technology and operational excellence. One of the big contributors
was having our own fleet. We had five riders when we started; we have crossed 20,000 riders
now. We have India's largest last-mile delivery fleet."38
As on 2019, Swiggy was focusing on increasing its order intake capacity and apart from
the restaurant chains and bigger players, it was also collaborating with small restaurant
owners, (who formed a big part of the unorganized restaurant business). For many small
restaurants, Swiggy was the primary route through which orders flowed into their restaurants.
As the competition between Swiggy and Zomato heated up, both increased the salaries of
DEs from 10,000- 25,000 per month to 25,000 to 50,000 per month from July 2018. Ujjwal
Chaudhry, engagement manager, e-tailing, at RedSeer, said, "With such fast growth, there are
challenges for the companies to fulfil which leads to higher pay out to DEs. It is a short-term
trend and cannot be a sustainable strategy because any hyperlocal industry is tight on unit
economics and they cannot keep burning money,"39 Both the companies had been eyeing
expansion and Swiggy had already burned cash on its marketing, advertisements, discounts,
and low-cost deliveries, to remain ahead of Zomato in FY 2019.
Testing New Waters
In June 2019, Swiggy announced its entry into the home chef space with the launch of a new
venture, 'Swiggy Daily', an application to give consumers access to a variety of simple home
style meals prepared by home chefs, tiffin service providers, and organized vendors.40 "Swiggy
Daily" was expected to list over 30 options for every meal. The platform would include meal options from a mix of organized vendors like Homely, Lunchly, Fig, iDabba, and
Caloriesmart; popular tiffin services like Dial a Meal and Dailymeals.in that specialized in
food fit for daily consumption, and expert home chefs like Sumita's Food Planet,
Mrs. Ahmed's Kitchen and Shachi Jain. This was expected to be a separate application of
Swiggy focusing on providing customers with lunch and dinner options.
Swiggy Daily would be a subscription-based service which aimed to deliver food with
order values ranging between 50 and 150 without a delivery fee. According to Majety,
"There is a growing demand for quality and affordable everyday meals. With a mix of
organized vendors and home chefs, Swiggy Daily will cater to this latent demand for
homestyle meals that are an affordable, long-term solution for our daily food needs."41
In February 2019, Swiggy also announced the launch of "Swiggy Stores" in Gurugram.
Swiggy Stores aimed to deliver different products from 3,500 stores in Gurugram, and had
partnered with 200 stores in categories like Kiranas and supermarkets, fresh meat, baby stores,
and health supplement stores among others. It had also partnered with Licious, The Mom's
CO, and Apollo Pharma, to name a few.42
In 2019, Swiggy became the first aggregator to offer the 30 min or free delivery feature
to customers. It was available at select restaurants in select cities but it solved a huge pain
point of customers having to wait for 40-60 min for their food to be delivered. Once it was
fully launched-Swiggy would bear 70%-80% of the cost in case it failed to deliver on time
while the rest would be borne by the restaurants.
Future Plans
As of 2019, Swiggy was looking to further strengthen its tech efforts and focus on an AIdriven
delivery platform for hyperlocal and on-demand delivery. According to Dale Vaz
(Vaz), Head of Engineering and Data Science at Swiggy, "The first big tech challenge was
moving from a food tech stack to a platform in the hyperlocal category. Today, even though
we support food, we should able to do the same for other things like buying medicines,
grocery, pet food, etc. With our AI-first approach, we are looking at how can we replace and
rethink Swiggy products, and replace some of our hard-coded system assumptions with
intelligent AI-based systems. We are doing this across the board to look at how Swiggy can
become an AI-first product. Last year, we hired multiple PhDs from international
universities, and tenured people from places like IBM and GE Research as senior
scientists."43 Swiggy had also focused on building a dedicated applied research team that
would be separate from the data science team. That team would focus on areas around voice,
vision, and Natural Language Processing (NLP). According to Vaz, "The thinking is that
there is a team that will look at immediate problems while the applied research will look
at how to solve for the next set of 100 million customers."44
Swiggy had been building data infrastructure since 2017. It also worked on the Swiggy
data platform, which housed all the data (from customers, drivers and restaurants with 40
billion messages per day) that flowed through the Swiggy network. The company also aimed
to build the Swiggy AI stack and intelligence models based upon that. Swiggy sought to build
long-term capability as well by partnering with external research institutes and industries.
According to Vaz, "We have partnered with universities to enable this kind of research under
a grant program where Swiggy can fund that research."45
Swiggy also aimed to democratize AI across the organization to build models around it.
Vaz opined, "We launched this program called 'AI for All' where we are training our teams,
analysts, and even business leaders. Eventually, our vision is to make AI a part of our culture
where we have built the capability and knowledge in the organization."46
As of 2019, Swiggy was looking to make inroads into the non-food business. Majety
opined, "We've been closely observing our consumers' lives and we feel that there are some
unique capabilities that we have that can take a shot at solving more problems, even outside
the realm of food. You'll see us start making small forays into new areas."47
Notes