General Information

Watson Analytics takes a peek into Holiday Shopping and Weather

Blog Home > Watson Analytics takes a peek into Holiday Shopping and Weather

Watson Analytics takes a peek into Holiday Shopping and Weather

Do cold temperatures or bad winter weather keep you from going out for those last minute gifts or holiday discounts? That’s the question the Weather Company asked visitors to their website in a recent survey. Most of us check the weather before going to work or school, but what about before going holiday shopping? According to Paul Walsh, Business and Weather Expert with The Weather Company, weather and forecast have a huge impact on holiday shopping, with weather being the factor in determining when, where and what people buy (

Businesses can benefit from being weather aware to help guide them on operational decisions by considering the combination of weather with people’s behavior. As Walsh says: “Businesses can use (weather information) to be better prepared with the right products at the right time to be able to serve their customers” (

For this blog, I used IBM Watson Analytics to analyze and visualize the data from the survey.

Holiday Weather Image 1

Check the weather forecast before holiday shopping?
I started out by looking at the survey results for the basic question: “How likely are you to check the weather forecast before you go holiday shopping?”. The survey recorded responses as multiple choices of “Likely”, “Somewhat Likely”, “Very Likely”, “Unlikely” …, so I tried different visualizations to look at these results.

To get going, I used Explore to type in the column name for this survey question and see the related starting points.

Holiday Weather Image 2

The first starting point provided a bar chart visualization, which I sorted in descending order and excluded blank survey responses.

Holiday Weather Image 3

Next, I switched the bar chart from column-style to row-style.

Visualization Tip: You can switch between column- and row-style bar charts using the “Display as column chart” option.

Holiday Weather Image 4

The row-style bar chart gave another perspective into the same results.

Holiday Weather Image5

Changing the visualization type to a tree map also gave another way to look at the break down of the results.

Holiday Weather Image 6

Then I started thinking about how this breaks down by regions of the country, so I used the Rows feature for the tree map to organize the visualization by region.

Holiday Weather Image 7

Ultimately, I was curious to see the “Do you check the weather forecast” question as a simple “yes” and “no”. To do this, I used the data group feature in Explore to create a new column that organized the responses into “yes” and “no” groupings. A basic pie chart summed it up best: 65% of people said they are likely or very likely to check the weather before holiday shopping.

Holiday Weather Image 8

One last visualization for this survey question was to see yes/no by regions of the country. Looks like everyone checks the weather forecast before holiday shopping except for the western part of the U.S.

Holiday Weather Image 9

Too cold outside to go holiday shopping?
What’s your cut-off point for heading out to buy those holiday gifts? According to the survey, “baby it’s cold outside” has a different meaning (temperature level) depending on where you live.

Overall, the lower cold temperature thresholds were in the north and northeast of the U.S., where people are more used to the cold weather. Survey respondents from California said “no” to heading outdoors to holiday shop when the temp drops below 32.95 degrees F. But if you live in North Dakota, you’re good for shopping all the way down to 3.63 degrees F. Wow, that’s cold!

Holiday Weather Image 10

I created this map visualization in Explore and then moved it into Assemble using the collections feature so I could add some callouts. The visualization shows the average cutoff temperature for going out shopping by state, where darker color means colder temperatures.

Did You Know? In Assemble, you can change the color order for heatmap values (dark to light or light to dark). In Edit mode, select the visualization, then click Properties > Visualization details > Color order.

Holiday Weather Image 11

What type of weather keeps you from going out holiday shopping?
Some people go out holiday shopping no matter what the weather is. But for most of us, bad winter weather of more than 5 inches of snow or icy conditions means stay home.

For this visualization, I organized the results from multiple yes/no columns of the survey to see the overall feedback on the type of weather that keeps people from going out to holiday shop. For example, flurries are OK, but ice storms are not; they have the most significant impact on shopping with 76% of people staying home in this type of weather event.

Holiday Weather Image 12

Weather and the forecast definitely have a big impact on holiday shopping plans, causing a majority of people to check the forecast and even change their shopping plans. Businesses can also combine weather and forecast information to better prepare for winter weather and customer behavior. Watson Analytics gives you the tools to dive in, visualize and get insights from survey and weather data like this. You could even extend the analysis by analyzing related comments on social media websites using the new Watson Analytics for Social Media.

Survey data and related videos
This analysis was done on data from the 2015 Forecast Factor Holiday Shopping Survey on, conducted October 16-19, 2015. The survey included 5238 respondents, but not all records were complete and were not used in this analysis. For more information, see the related article and videos.

How Cold Is Too Cold For Holiday Shopping? (Ada Carr,, Nov. 2015)

Holiday Shopping and Weather Survey Results (Forecast Factor, Nov 8, 2015)

Weather Impact on Black Friday Sales (Forecast Factor, Nov. 29, 2015)

More General Information Stories

General Information

VIDEO: Navigating the spiral in IBM Watson Analytics

With IBM Watson Analytics you can navigate your data from different perspectives using the Spiral so that you can discover any unknown relationships. To do this with the Spiral: You can modify the view of the spiral using the level selector You can add two fields to expose more predictors You can add additional fields using combination (this exposes the most predictive information) Watson Analytics also lets you navigate your data sets with the ability to view associations within the data. You can use the Data Tray at the bottom of the page to change the view of the spiral. Simply click and drag a data field to the centre of the spiral. In this video, you can see the spiral in action with sample employee retention data. For more information about Spirals, please see the documentation.

General Information

Complimentary virtual event, Outthink Disruption: Innovate New Possibilities with Analytics

See and hear what’s brand new with Watson Analytics& Planning Analytics on-demand. During this broadcast, you’ll hear how IBM’s newly improved smart data discovery and enterprise planning cloud products make working lives radically simpler with agile, self-service capabilities. We share the latest advances in interactive analytics such as Expert Storybooks and planning accelerators that can bring know-how and best practices to your next initiative. During the Customer Panel segment, representatives from Hot Topic and Revelwood share their performance management success stories. Watch it here on-demand.

General Information

Everyone has an opinion - Starting a Conversation in IBM Watson Analytics

Have you ever seen something in one of your Watson Analytics reports that raises more questions than answers? That's where the Conversation functionality in Watson Analytics comes in. With Conversations, you can collaborate with other users inside of Watson Analytics by leaving comments and even creating polls that are saved with an exploration, prediction, or view. The conversations you create are only available for that specific asset. As with most social media services, within your comments you can also use the @ symbol to mention other users and the # symbol for keywords. For more information on using the Conversation functionality and getting the conversation going, see the How to add comments to analysis in Watson Analytics video on YouTube. For more information on creating a poll, see the How to create a poll in a conversation video on Youtube. Remember to subscribe to the IBM Watson Analytics channel on YouTube!