Forecasting Business Data From Weather Analytics

Paul Walsh, a former Air Force meteorologist, has spent the past 14 years giving advice on business decisions that take weather into account. Since August 2011, he's been doing so as vice president of weather analytics at The Weather Channel, where he's tasked with showing companies how they can use weather data to target their advertising more effectively.

“We’re using Walsh’s background and data brain to figure out how to bring something to market that would be extremely useful to clients,” Beth Lawrence, a sales exec at The Weather Channel, said in an Ad Age interview. The benefit for retailers is that knowing who to target when could help them promote their products more effectively.

The article offers this illustration: “If it’s set to be unusually cold in Phoenix next week, The Weather Channel could use Walsh’s predictions to tell sweater makers when to start advertising.” But that example offers little beyond the idea of telling people to advertise ice melt and snow shovels a couple of days before a snowstorm is due to arrive. There is nothing new about using those types of forecasts in marketing. Stores have done this for as long as news outlets have provided weather forecasts.

Walsh’s approach incorporates much more lead time, is based on long-term predictions, and plans for a whole season. In a recent presentation he gave for the Global Interdependence Center titled “Rethinking the Weather,” Walsh explains how businesses can optimize their planning by taking advantage of the advances of weather forecasting based on technology and big data analytics. A segment of the presentation is available on YouTube:

Walsh says The Weather Channel brought him in to create ways to use long-term weather data for “better long-range planning decisions.” It concentrates on retailers that use data to “optimize the way they plan and then market and then distribute products.”

Weather knowledge really is power, Walsh says. He cites his Air Force background and says the military has long used weather information for tactical and strategic planning -- in the case of the D-Day invasion, the Allies had better weather information than the Germans did.

As for the private sector, “weather is important because it has a significant impact on the economy,” which amounts to 30 percent, or $485 billion, a year for “normal weather.” For our current state of weather, Walsh shows a map dated at the end of December 2011. It shows a three-month prediction of patterns that include some cold snaps but generally mild weather for January and a colder March in the North. For seasonal apparel retailers, this pattern is far from ideal. They like a colder winter and a warmer start to spring.

In the video, Walsh does not offer possible solutions. However, we can infer that retailers that know the season will not match what they normally anticipate will adjust their strategies accordingly. For example, in anticipation of a cool March, retailers in the North could put off promotions for patio furniture and barbecues, because people usually need the prompt of warmer weather to start considering outdoor entertaining. Instead, the retailers could push blankets or sweaters in new spring colors to offer something seasonal that corresponds to the weather.

Everyone can continue to talk about the weather, but analytics can give retailers the tools to do something about it and make better business decisions.

Ariella Brown,

Ariella Brown is a social media consultant, editor, and freelance writer who frequently writes about the application of technology to business. She holds a PhD in English from the City University of New York. Her Twitter handle is @AriellaBrown.

Hiring Trend: Machine Learning Reduces Bias, Increases Applicant Pool

Unilever is using machine learning tools to increase the pool of entry level applicants and then test and evaluates candidates in the pool to cull those who are the best fits. Adding machine learning tools reduces bias, increases diversity, and makes the whole process more efficient, according to Unilever.

Analytics and School Attendance: A Laundry Story

Appliance maker Whirlpool suspected there was a correlation between student attendance and access to clean clothes. Here's the story of how the company placed washers and dryers in schools and tracked the difference it made in student performance.

Re: Are Retailers finally getting smart ?
  • 3/19/2012 12:16:06 PM

@ckelly1 Hello, and welcome to the board. I've noticed that retail establishments, in general, project a couple of months ahead. You don't just start seeing Christmas merchandise merchandise in October, you start seeing Easter merchandise in February.And spring clothes go on display already in January, though some of it passes as "cruisewear" for vacation. 

Are Retailers finally getting smart ?
  • 3/19/2012 11:35:36 AM

Ever wonder what happened to all the winter clothes in Feb? I recently found out you can't buy a space heater either at the big's already summer there and fans and air conditioners have taken their place.  

I'm not sure that they can react on a a change if the weather within a couple of weeks.  I guess that is why it is so important to maintain and support smaller hardware and clothing shops that can easily pull out appropriate gear for the change in the weather pattern.  Buy local

Re: How'd Weather Channel do ...
  • 3/14/2012 1:31:16 PM

Enjoy the day, Beth! We're having very similar weather in NY, and I hope to have a chance to get out on a bike today. 

Re: How'd Weather Channel do ...
  • 3/14/2012 12:46:49 PM

@Ariella, since we're talking weather ... I just have to toss in here, it's mid March in Chicago and the temperature is a balmy 69 degrees right now and headed toward the mid 70s by late afternoon. I think historical weather trends analysis would show that's out of whack with the norm. (Think I'll grab my laptop and work from my yard now!)

Re: How'd Weather Channel do ...
  • 3/13/2012 6:40:42 PM

@Beth, I did the same for my daughter. What can you do? You don't want your kids stuck if it does snow.

Re: How'd Weather Channel do ...
  • 3/13/2012 6:03:40 PM

LOL Ariella, yes, indeed, it was a pleasant surprise to have a mild winter. But thank goodness I don't have responsibility for purchasing & managing retail inventory, etc.! I was angry enough I bought my son a new pair of snow boots that he wore only once!


Re: How'd Weather Channel do ...
  • 3/13/2012 5:38:20 PM

@Beth Interesting, as you can see in the video, Paul Walsh already had the data in for a very mild winter by the beginning of December, which did prove accurate. I have to admit, I don't pay much attention to weather outside my own area, so I looked up the Chicago report and saw this:

The post garnered comments like this one: "Above average temps in Chicago this week, no snow. Epic fail, Accuweather. Epic fail."

"so far our weather has proven to be the most MILD start of winter I can remember - the grass is still green and no sign of snow. What's up with that AccuWeather?"

That forecast was revised in late December as follows: "'"It does not appear that the winter season will replicate last year's cold and snowy winter," the weather service said in a special statement. "Instead, odds favor this winter to continue to be on the warm and wetter side of average.'"

The weather data everyone uses would likely all come from the same source, the National Weather Service (whose radar maps are also used by scientists to track bat movements.)  But the key to predictive anlaytics is to accurately guess the movements of the forces that contribute to snow, etc. It would seem that whoever did that for Chicago specifically erred. But it must have been a pleasant surprise to have such a mild winter, particularly in a region that usually has pretty tough ones. 


How'd Weather Channel do ...
  • 3/13/2012 5:13:14 PM

@Ariella, all this talk about predicting weather has me wondering how accurate or off The Weather Channel's predictions for this past winter season were -- and whether advertisers could rely on it. All I know is all the local forecasters in Chicago had been warning us that we'd have a horrendous winter and it turned out to be the mildest  -- almost frighteningly so when you think about the implications -- I've ever experienced here (and I've lived here just about all my life). I don't remember ever checking with the local forecasters were saying against TWC forecasters but I'm guessing everybody was more or less saying the same thing?!

Re: Sunny forecast ...
  • 3/13/2012 8:57:19 AM

@Daniel Yes, and the ramifications extend even to produce and grocery prices. Weather changes have a huge impact on crops. For example, last year's harsh winter resulted in a diminished tomato harvest and an escalation in their prices: In Feburary 2011, NBC reported "With so many crops killed off by the cold, produce shop owners are seeing tomato prices triple. Cases that usually cost $12-15 are up to $40."  Other crops were similarly affected. There's not much the farmers or grocers can do in that situation, but those who can use canned tomatos in their products may decide to stock up before prices would rise if they anticipate the long term weather effects. 

Sunny forecast ...
  • 3/13/2012 2:14:24 AM
1 saves

Ariella, seasonal based marketing can make use of weather forecasting. Weather forecasting or seasonal predictions like average/above rain falls for a particular time frame, intensity of cold or snow fall etc can help the companies to either increase or decreases the production based on predictions. I think most of the umbrella and cold jerkin manufactures are making use of datas from weather forecasting department.

Page 1 / 2   >   >>