How would you asses the impact of weather on consumer spend – You first ask clarifying questions – What is consumer spend- Can we assume retail spend across all categories and across ecommerce and brick and mortar? If Yes, What is winter? Which place is this? Is it October to November in US because Winter in India or Australia could be different or is this generic winter in any place? And then the most important question of all – What do you mean by Impact? Typically, you have forecasts of consumer spend – How much you thought consumers would spend as a whole or by category or by demographic or geo location etc. You also take seasonality in to account as you forecast. When the actual number is different from forecast, you try to attribute the forecast error to various factors and weather could be one of them. So you first look at actual weather vs forecast – was it an exceptionally cold winter, was it a hotter winter than expected? This would typically manifest itself by people buying more or less of winter specific products – like clothes, food, furniture. You can also look at the regular products such as toilet paper, bread etc that people buy irrespective of weather and see if they met forecasts or if people could not shop because of bad weather and they were impacted as well… I would do a forecast vs actual analysis and then do specific weather related attribution to the forecast error

Clarifying questions
By assessing impact means change in spending last winter vs previous winters and whether the change was because of weather?

1. Any geography or all?
2. Any specific category?

Steps
1. Spend means – Change in spending on goods(essential & non-essential), Change in spending on services(vacation travel/hotel), Change in seasonal spending
2. Look at YoY spend during the same period.
3. Look at change during holiday period when spending peaks
4. Look at environmental factors such as unemployment, recession, war etc…
5. Look at weather related issues such as storms, harsh winters etc…
6. Look at spend on essential vs non-essential items. Essential means groceries vs non-essential means luxury goods, clothing etc…
7. Look at spending across diff. geographies. Especially areas with severe winters vs low winters.
8. Look at other factors such as change in tariffs, taxation which can contribute to pricing thus spending spending.

Here’s a framework for answering this Google product strategy question:

1. CLARIFY:
   1. Should we be assessing the impact on Google’s consumer spend or any consumer spend? Google.
   2. Do you want to focus on individual consumers v. business? Your choice.
   3. Is there a specific region or global? You choose. (I will focus on US.)
   4. Is there a specific product or service I should focus on? Your  choice. (I will focus on Google Pay. Reasons below.)
   5. Was there anything unique about last winter I need take into consideration (i.e. I’m assuming non-COVID, etc.)? Correct. Nothing crazy uniques.
   6. Is it OK to assume the winter is cold? Yes. (Here, I am assuming the weather is the weather we see in the North East in areas like Boston or NYC.)
2. GOOGLE BACKGROUND: Goog;e’s mission is to organize the world’s information and make it accessible to all users. It offers a variety of products and services in both hardware (phone, Chromebook, Home, Nest, etc.) and software (search engine, Youtube, Wallet, etc.). For this question, I’d like to focus on Google Wallet / Google Pay because we have deeper spend analysis potential here.
3. GOOGLE PAY BACKGROUND: Google Pay is Google’s digital payment system similar to Samsung Pay or Apple Pay. On mobile devices (Android) and watches, it offers contactless payment via near field communication (NFC). On web, it offers pre-populated payment on browsers including Chrome, Firefox and Safari. A user authenticates payments via a pin.
4. ASSESSMENT CRITERIA: I will evaluate the impact of weather on consumer spend in the winter based on the following criteria. For each metric, I’d make a note of severe weather changes (for example, days of heavy rain or snow, extreme cold (ex. 30 degrees F or lower) or extreme heat (50 degrees F or hotter)) and compare those days to an average day in winter (i.e. average temperature, no crazy rain or snow, etc.). (Note, we could also compare the consumer spend of winter in general to other seasons, but I am interpretting the question as how does severe weather events during winter impact spend. )
   1. PAYMENT VOLUME:
      1. Total Volume: Volume of total payments on Google Pay in day
      2. Total Transactions: Total # of transactions on Google Pay in day
      3. Average Transaction Size: Average transaction size on Google Pay in day
      4. Average # of Transaction: Average # of transactions / customer on Google Pay in day
   2. LOCATION OF SPEND: 
      1. Merchant location: Changes to where customers spend on average based on merchant classification (ex. retail shops v. grocery stores, etc.)
      2. Channel: Volume of payments and # of transactions on mobile v. web. Could take total volumes / #s and also breakdown average for customer. For example, during bad weather, transactions are more likely on web than mobile (i.e. outside).
   3. TYPE OF SPEND:
      1. Line Item Details: If Google Pay has data on purchases, changes in trends to purchases. (Ex. on heavy snow days, higher purchases of coffee.)
   4. FRAUD:
      1. Fraud Complaints: # of fraud complaints in day
      2. Fraud Volume: Total volume or average volume / customer of fraudulent spend.
   5. AD REVENUE:
      1. Data on User: Google doesn’t earn fees from Google Pay transactions but does collect more data on user which it can use to run ads. How much data on user does it collect on extreme weather days v. regular winter days.
      2. Ads / User: Is Google able to run more ads on user based on data collected via Google Pay on inclement weather / super warm days in winter.
5. EVALUATE CRITERIA: 

   1. Theme	Criteria	Insight into Impact of Weather
      Payment Volume	Total Volume	High
      Payment Volume	Total Transactions	High
      Payment Volume	Average Transaction Size	High
      Payment Volume	Average # of Transaction	High
      Location of Spend	Merchant location	High
      Location of Spend	Channel	High
      Type of Spend	Line Item Details	Medium
      Fraud	Fraud Complaints	Low
      Fraud	Fraud Volume	Low
      Ad Revenue	Data on User	Low
      Ad Revenue	Ads / User	Low

6. SUMMARY: Given prioritization, I’d focus on the payment volume and location of spend metrics. If I’d like specific details on a customer, I’d focus more on averages of the customer v. totals for the entire population using Google Pay.
7. CAVEATS:
   1. Google doesn’t have any insight into Apple users. Missing that portion of the population to insight on weather changes during winter, so data doesn’t necessarily tell us everything re. all types of Wallet Spend (including Samsung Pay too). Given that Apple has a high mobile device presence in the US (based on personal observation), caveat to data should be noted.
   2. Google Pay users may also be spending outside of Google Pay on card or cash. Not part of data.
   3. Not all merchants have Google Pay acceptance.