As a PM at Swiggy, you observed that the number of orders went down by 20% last week as compared to all the weeks before. How would you go about doing the Root Cause Analysis here?

Question

As a PM at Swiggy, you observed that the number of orders went down by 20% last week as compared to all the weeks before. How would you go about doing the Root Cause Analysis here?

Should focus more on the metrics that we should choose to go about identifying and clarifying the problem further.

Asked at Swiggy over a year ago

2.7k views

Problem Solving category

Asked at

2.7k views

How to answer Problem Solving questions

Interview Guide

Top Problem Solving interview questions

Invite members
Invite by email

Add another

Tarun Kumar · Answer 1 · 2023-10-25T18:58:14+0000

Starting with few clarifying questions to begin with:

Has this impacted specific region or across all region? (Answer: All region)
Is this a trend or seasonality? (Ans: Not due to seasonality)
Is this happening for particular category or across all categories? (Ans: Food category is taking this decline)
There is no issue with the data (Ans: Yes, data is fine)

Ok, going ahead I'd like to break down the analysis, I'd first understand if there are any changes in the following:

User behavior
Funnel
Internal factor
External factors

Going ahead

Is this happening for some specific segment of users or all users? (Ans: All users, no segment)
Are we seeing decline in
- Total of orders per week or (Answer: This is happening)
- Avg. orders per user week

Nice, going ahead I'd like to look at few KPIs to understand more about the situation

User launches app
User goes to food tab or User goes to food discover/rec. page (V%)
User adds item to the cart (W%)
User goes to cart (X%)
User completes the payment (Y%)
User gets the order delivered (Z%)

Are we seeing a decline in any of the metrics in the funnel?

Case 1: If decline is at payment stage

Answer: Yes, we are observing a decline at some stage of the funnel i.e., When user adds item to the cart and when user completes the payment.

Great, to know this. Let me again dig deep into what could be happening here.

Users are adding items to the cart but the payment is failing. Let me understand a bit better by looking at the following data:

Instances of payment failure has increased? (Answer: Yes)
Instances of payment by a particular payment method has increased? (Answer: What do we need to check here)
- Payment by Card
- Payment by Cash
- Payment by UPI
- Payment by BNPL
- Ans: Among the above, we are observing a decline in payment by UPI in the last week and it can be attributed to that.

Analysis: Due to UPI payment failures happening in the app, the users are not able to order. UPI is one of the highest contributor to the payment on Swiggy's platform.

Recommendation: This can be an anamoly as UPI failures generally don't tend to happen. To avoid this in the future, we can remove UPI options from the payment page in cases when there are UPI outages.

Secondly, we can ask user to try another payment method in such cases and offer some discount (5%) to test if the traffic doesn't drop after looking at this discount.

Case 2: If decline is at app launch stage.

I'd like to understand more about the user behavior here,

Is there increase in the no. of customer issues? I'd check the following data
- No. of support tickets for Quality
- No. of support tickets for late delivery
- No. of support tickets for items spilled or missing
If there decrease in the number of dishes being added
Increase in the instances when item not available?
Is there an increase in the delivery fee recently?
Are we not running any discounts?
Are we seeing increase in the comments on app store?
Are we seeing the average rating of dishes and restaurants go down significantly?
Has the number of calls increased from delivery partner to the user?

Answering these questions will point towards some user behaviour changes that might have happened. Next I'd look at the internal factors

Did we roll out some design or UX changes?
Did we roll out some change experimentation?
Did we change our algorithm a bit?
Has the crashes increased?
Are we seeing an increase in the app crash numbers?
Has the page load time for Swiggy has increased?

Answering these questions would help me understand the internal factors that are responsible towards this decline. Next I'd look at the external factors:

Was there a bad PR for Swiggy recently?
Is there a competitor that's running heavy discounts to sweep our customers?
Is there a competitor that's not charging delivery fee for order?

Based on the answers I'll be able to form some hypothesis around the external factors driving this decline.

This would be my approach on the case.

Can someone review this please? Thanks!

Udit Nigam · Answer 2 · 2023-04-21T14:28:34+0000

I'm an aspiring PM and writing the answer for the first time here . So would appreciate feedback from everyone.

A:-

Problem Statement here is to do RCA for 20% drop in the number of orders on Swiggy.

I would go ahead with asking some clarifying questions:-

I’ll start with some possible internal reason for this drop after that I’ll take into account external factors if needed.

INTERNAL FACTORS:-

Q1:- Is this problem specific to Android or iOS or both?

Q2:- Did our Application servers face any glitch or were down during the week?

Q3:- Have we introduced any changes in the recent past i.e in UI or in the user journey of placing the order or any sort of version upgrade by us or any other team apart from us?

Q4:- Have we removed any payment type from our Apps or were they unserviceable during the particular week ?

EXTERNAL FACTORS:-

Q1:- Have we seen such behavior in past years during this time frame due to any religious festivals, as Swiggy has been operating in India since 2014 so it must have accumulated the data & we can quickly infer this .

Q2:- Is this drop more severe in some specific states or throughout the country?

Supplementary Q:- Is there any new legislation passed by the govt which we have missed to inculcate in our system ?

Q3:- Were there any high valued(high frequency) restaurants which stopped taking orders during peak hours ?

Q4:- Did customers face higher than expected ETA for their deliveries or were we not able to supply delivery partners to fulfill the order?

Q5:- Did this happening to us only or our competitor Zomato is also faced it?

Supplementary Q:- Has Zomato introduced any new feature ( include new payment gateway also) or discounts due to which this drop has occurred on Swiggy?

Q6:- Was there any negative/boycott trend running in that week over social media against Swiggy ? If yes , reason for that?

These could be the plausible reasons which i can think of right now & we can discuss the solution if any of the above issues are causing this drop else we can move further down the funnel .