How would you improve the recommendations for Zomato?

My approach to this problem would be first to clarify the problem statement,  followed by some clarification questions to define the scope clearly. Then agree upon the customer segments which will be impacted followed by the pain point for the customer segment. Then we can try to come up with solutions for those pain points and prioratise them. Finally we can think of metrics which will help in assessing the impact of the improvements made.

1.        Zomato is a food app which primarily does the food delivery, but it also provides suggestions for dine in and help making reservations at the restaurants. Thru its subscriotion options like Zomato gold, a user can get discounts in both food delivery and dine in reservations.

2.        So, what do we mean by improve recommendations and what is the goal to be achived by improving the recommendations?- to improve user engagement on the app so that the user is not overwhelmed by the options suggested and has a happy experience.

3.        So, I cannow reprhase that my problem statement as ‘As a user, I want to have better recommendations on  Zomato, so that I get quality suggestions based on my preferences and make a good choice.

4.        Now, I would also like to understand different customer segments of Zomato, which will help in deciding how to proceed for a solution. So, we can create customer segments in different ways like on the frequency of usage – power, casual and passive users. We can also segment like types of users based on age – like bachelors, family folks, health concious , foodies etc.

5.        For narrowing the scope, I think we can focus on casual users , I feel they have a potentioal to convert to power users if the app exp is extraordinary.

6.        Within casual users there can be all kinds of users like bachelors, families etc.

7.        Now that the customer segment is identified, next I would foucs on the user journey because it will help in coming up with pain points at every stage which we need to understand in order to improve recommendations.

8.        So, I log in to the app, and there are two possibilities – 1. I have something in mind to order like biryani or paneer dishes and the other option is that I don’t know what I want to eat and open for options. > either ways I make a selection of place followed by dish and then place the order.

9.        So, in this entire journey, recommendations will come at the second step after a user logs in. Lets try to find out the pain points here with priority based on how important they are to customer.

10.  Next step is to find out a solution for our pain points.-

So, I think we can focus on the first two pain points for the customer.

It is also impotant to understand what derives the recommendations – so if we think of a real life when we have to recommend a place to someone, we like to understand what the other person is looking for . For e.g. in case of a restaturents we ask u want veg or non veg or buget or a fine dine or a quick meal place. These things matter so we need to keep these factors in mind while dsigning the solution.

11.  Finally we need to have some metrics to measure the impact of these improvements. So for this one I think would be to see the increase in numberof orders per month for a casual user. We also need to see the impact on other user groups like power users for this change. Another metric we can see is the number of times users are engaging with the app even if the order is not placed, which might help us in understanding if the new features are helping or not.