Zomato — Agile Feature Implementation and Its Study.

Zomato.com

Zomato is an Indian restaurant search and discovery online service (web and app). It is spread worldwide, And is known for its Online food delivery It has also launched Cloud kitchens in many regions across the world, It gathers and provides information of restaurants on a regular basis, it allows food lovers to share their reviews and photos of every restaurant. so that people can make an informed choice.

This story talks about the use of agile methodology to build up some new features for the zomato product.

Understanding What is Agile Methodology

Agile methodology is a software development process framework that adopts the iterative and incremental approach — open collaboration, and process adaptability throughout the life-cycle of the project

Incremental and Iterative characteristics

Incremental software development entails delivering finished components of the whole in parts. It allows the team to stagger the release of features which are of utmost value to customers.

The iterative approach produces earlier release and better stakeholders feedback. The feedback is used to review the project priorities and make changes to the requirements, functionalities, estimations, plans, resources, etc. as required.

Iterative and Incremental Agile Approach

Benefits of Agile Method

This iterative agile approach is more flexible and its short time-span iterations seek improvement for the project in small release, with minimal planning, rather than plan at length. This helps to minimize the overall risk, and allows the project to adapt to changes more quickly.

How is Agile Methodology more beneficial than other Approaches ?

• At any point in time, you are able to measure the progress by reviewing the working software, thus allowing a continuity in visibility from the start to anywhere in the middle, up until the end.
• Unlike in traditional methodologies, the gestation period is flexible when it comes to Agile owing to the incremental iterative product development process.
• Delivering a working software in two to three weeks as opposed to a year mitigates the technical risks that usually lurks in a traditional scenario
• Owing to its intensive nature, an agile method transcends the norm and encourages teamwork as part of the process, thus allowing communication at every level

Zomato India Stats.

Ideation: Rating would be done by users, and viewed by them too. For reviews, a provision to view them on restaurant details screen, along with a screen to capture the review, are needed. This could be giving an Option to Add Maps to see the list of restaurants around and heat-map the nearest restuarants.

Following are the required items to be developed and added, which will be present on the restaurant details screen:

1. Rating Landing Page, with list of restaurants on platform for certain city and area. Selecting a restaurant takes them to a popup that asks for rating of that particular restaurant.
2. Rating Popup to rate the selected restaurant, with a 5-star scale, and allowing users to rate in integral values of 1–5
3. Email to Curated Users, asking them to Rate Restaurants
4. Display Rating, and Number of Users who have rated
5. Add Rating Option on Restaurant Details Page
6. Review Landing Page, with listing of restaurants, for users to review a restaurant. Selecting a restaurant takes them to Review Page
7. Review Restaurants Popup, with options to add images, text description, and text heading
8. Email to Curated Users or Registered food bloggers , asking them to Review Restaurants.

Design: Following would be the design characteristics of all icons and screens, mentioned above

1. Popups are designed with minimal elements, and little description for the process to be followed
2. Emails are sent out normally, with mention of purpose of email, how and why to rate/review restaurant, and the link to perform action
3. Rating is displayed right below restaurant name, with no other fields close to it, to indicate its high value
4. Initially, option to Add Rating would be a long bar of sorts, to capture intimate users to add rating, which is a new option, and hence many might not notice it by themselves unless prodded on
5. Option to Add Rating will be placed beside Rating Icon, in a much smaller size to the rating icon, after finding that enough restaurants have gotten a sizeable amount of reviews to make the system self-sufficient
6. Reviews Count Icon, and the option to view them on clicking this icon, would be present close to the Rating Icon, to indicate similarity (Principle of Proximity)

Testing: Metrics to be evaluated would be very crucial for any organisation, it will be detailed with every aspect being mentioned

1. Conversion rate of emails and captured feedbacks (to test efficiency for reviewing, after learning from ratings email Tools used can be Mailchimp, Survet Monkeys )
2. Correlation between time spent by users on restaurant details page, after ratings have been introduced, and the displayed rating. Lower rating restaurant pages are expected to have lower time spent by users, as users would glance at the low rating and not consider that particular restaurant as an option anymore
3. No of Ratings per week added, from launch of Add Rating button
4. Initial stagnation, and continued growth, in no of ratings added per week, on shifting from Add Rating long-bar, to the smaller Add Rating button
5. Correlation between time spent on reviews, and number of reviews received by a restaurant. Higher count of ratings should imply that users are browsing through more reviews, and hence spending more time per restaurant

Above can be reviewed by using Google Analytics and Mixpanel.

Implementation and Deployment: Priority order for engineering build and deployment, in my opinion after impact-effort analysis, would be:

1. Rating Landing Page, listing all restaurants
2. Popup to Rate a Restaurant
3. Email to Curated Users, asking them to Rate restaurants
4. Display Rating, and Number of Users who have rated
5. Add Rating Long-bar
6. Review Landing Page, listing all restaurants
7. Popup to Review a Restaurant, with options to text description and a text header
8. Email to Curated Users, asking them to Review Restaurants
9. Popup to Review a Restaurant, with option to add images also
10. Reviews Count Text, to view Reviews on being clicking
11. Add Reviews Button, beside the Reviews Count text button

Filters for Search

Filters for search, as are everywhere, would add a sense of customisation and specificity to search. With this, users can limit the range of their search, without having to type in anything in the search, but instead chose and apply the most considered filters at the click of a button

Ideation: Filters would be provided based on existing tags, dishes, cuisines and the likes. The only developmental effort would be a modification to the search results page, with now additional options to set and view these filters. Cusisines play an important Role as it defines and broadgens the search scope.

Following are the required items to be developed and added, which will be present on the restaurant listing screen:

1. Ratings Filter: More than 1/2/3/4/5 stars
2. Cuisine Filter: South Indian, North Indian, Chinese, etc
3. Category Filter: Breakfast, Lunch, Dinner, Snacks, etc.
4. Location Filter: Indiranagar, Koramangala, etc (areas according to city)
5. Distance Filter: Less than 1km, Less than 3km, etc.
6. Cost for Two Filter: Less than 250, 500, 1000, 1500, etc.

Design: All filters would be displayed on the left of the screen, vertically, one after the other with options shown within each filter’s sub-section (like other general filters)

1. Hide and Show arrow buttons across each filter, ˆ and ˇ
2. Checkboxes across each option within each filter. Multiple selection allowed.
3. Filter should be specific and broad based on understanding of local Food cuisines to different Food cuisines.
4. White and Red colour coding would be sufficient for this

Testing: Metrics to be evaluated would be

1. Number of filters applied against total number of searches
2. Average number of filters applied per search (one search unit is search of a certain combination of words), for searches that had filters being applied. This should show a positive trend over time
3. Number of new users using filters, until critical mass is reached
4. Frequency usage amongst filters, and amongst options within each filter

Implementation and Deployment: For this activity, effort across all filters would be the same. I wouldn’t intimate the users of this change, as it is a normal factor in most sites, and users would be expected to spot it at the search listings page by themselves. The more clear and easy Implementation will enable easy and better searches.

Priority order for engineering build and deployment, in my opinion after impact analysis, would be:

1. Cuisine Filter (most users interviewed by me claimed they choose restaurants by cuisine first)
2. Ratings Filter (second most important option after cuisine)
3. Location Filter (jointly tied at second most important option)
4. Distance Filter
5. Cost for Two Filter
6. Category Filter

Online Ordering

Online ordering allows users to sit in the comfort of their home, and order the food of their choosing from the restaurant, while also not having to worry about how it is delivered. Zomato would either have the order delivered by their own delivery fleet, or with the help of contracted delivery partners. Online ordering gives an feasibility to search and buy any dedicated food cuisne no mater at what kMs radius it might me.

Ideation: Ordering online market was started on a big scale by Foodpanda in India. For Zomato’s entry into this market, with it being a direct complementary offering to their base product, it was an important feature to provide to the users. Users would want to have this option at both a home-page level, and also on the restaurant details page, as these are the two major times a user would choose online ordering.

Note: Foodpanda is now acquired by Ola Foods and acts as an direct competitor to Zomato.

Options to be developed:

1. Order Online option in Restaurant Details page, in close proximity to the Call Restaurant button
2. Order Online as a restaurant details option, in search results listings page, in close proximity to the Call Restaurant button
3. Order Online as a top-level option on the home-page, which redirects to the search results listing page, only containing order online restaurants
4. Interface to order items mentioned on menu, as is in general e-commerce and food ordering sites (changing quantities, adding customisations, prices displayed across each item and customisation)
5. Final Order details page, pending confirmation, containing details like items, quantities, order total and expected delivery time
6. Payment page, and integration with a payment gateway, which includes options to pay by credit/debit card, net banking, online wallets and COD
7. Order Placed page, containing order details like items, quantities of each, scheduled delivery time/time remaining, order total, and contact details
8. Developing an in-house payment gateway, after hitting a certain number of orders that economically justifies its development and management

Design: Following would be the design characteristics of all icons and screens, mentioned above. Most are text-based, with images as icons across dishes at a later point of time (not first version of menu listing)

1. Order Online option should be displayed near the Call option, in the restaurant details page, as the Call button would have been the normal way of ordering food by calling the restaurant and asking if they deliver with their own delivery boys
2. Showing an Map based option which allows users to understand how much time is taken for delivery boy to arrive at the address to delivery the food. (ETA should be considered as one of top priorities to be considered)
3. Menu Items listing page, with options to choose quantity across each item
4. Side-screen display of order details, at a quick-view glance, across the items listing page (important within the Menu Listings page)
5. Final Order Details page (as in other e-commerce pages today)
6. Payment gateway page (as in other e-commerce pages today)
7. Payment and Order confirmation page (as in other e-commerce pages today)

Testing: Metrics to be evaluated would be more on ETA taken for food deliveries.

1. Number of users using Order Online option
2. % of Order Online clicks as against the total of Order Online and Call Now clicks (this should show an increasing trend from launch of this feature, and reach a steady rate)
3. Number of Call Now clicks previously, now measured as the sum of Call Now and Order Online clicks (should have the general trend as seen before, and towards a more positive side)
4. Revenue generated over time from Order Online option (this should perform at a similar rate as the above point)
5. Average revenue per user
6. Average revenue per order
7. Average revenue per order, against rating of the restaurant of order

Implementation and Deployment: I would release this feature to those restaurants first, which receive the most number of Call Restaurant clicks, while also considering higher-ratings as a decision point, this coming from a usage and business point of view

Priority order for engineering build and deployment, in my opinion after an impact-effort analysis, would be:

1. Order Online option on Restaurant Details page
2. Menu Items Listing page, after clicking on Order Online option
3. Final Order Details page, pre confirmation
4. Payment Options page
5. Order Placed Details page
6. Order Online option on Search Results Listing page, across each restaurant offering it
7. Order Online option on home-page
8. In-house developed payment gateway

Summary –

· Agile development methodology attempts to provide many opportunities to assess the direction of Zomato app throughout the development lifecycle.

· In an agile paradigm, every aspect of development requirements, design, etc. is continually revisited throughout the lifecycle to preserve a product’s critical market relevance and ensure team’s work doesn’t wind upon a shelf, never released, is an attractive option for stakeholders and developers alike. Developers and Designers work simultaneously ensuring the roles is done one after the another.

The results of this “inspect-and-adapt” approach to development greatly reduce both development costs and time.