Warm Weather, Cool Responses

Posted by Guy Hanson 

Europe is now at the tail end of its warmest summer since 2006. However, the sunshine hasn’t been good news for everyone. For instance:

  • Ryanair announced that one reason for reduced summer bookings was the heat wave in Northern Europe, which encouraged more people to enjoy their holidays at home.
  • Conversely, before the heat wave fired up, SABMiller – the world’s second-biggest brewer – reported that “quarterly sales unexpectedly declined as cold, rainy weather caused customers in Europe and China to drink less beer than anticipated.”
  • Even Carpetright saw its first quarter results fall by 1.9% “after the UK saw significantly warmer weather in July.”

Previously, we have considered the impact that factors such as seasonality, offer size, and even the use of sex have on email responsiveness. So I started wondering – does the weather also have an impact on email subscriber engagement? In an attempt to answer this question, I extracted a cut of data from Return Path’s Inbox Insight competitor intelligence solution for the UK’s casual dining sector. The date range that I selected ran from mid-March to mid-July of 2013, starting with the tail-end of a bitterly cold winter and finishing with the first few weeks of the sweltering heat wave.

I then overlaid average weather data so the impact of external variables such as minimum and maximum temperature, and rainfall could be evaluated. The values represent UK averages for each day of the review period, with the data segmented into 3 ranges – “Low”, “Mid”, and “High” – as follows:

Data Range

Minimum Temp

Maximum Temp


Low Range

-4.0°C to 1.7°C

3.3°C to 11.1°C

0.0” – 0.1”


1.8°C to 7.5°C

11.2°C to 19.0°C

0.2” – 0.3”

High Range

7.6°C to 13.3°C

19.1°C to 26.7°C

0.4” – 0.6”

This enabled me to establish the following correlations:

  • Maximum Temperature:


Email Read rates for the casual dining sector showed little sensitivity to variances in average maximum temperature. However User-Marked Spam rates increased sharply as thermometers climbed, suggesting that while subscribers remain receptive to offers, they are less relevant when the weather is good and there are other activities competing for their attention.

  • Minimum Temperature:

image003 (1)

Read rates showed a much stronger correlation with minimum temperatures. There is a distinct sense of the nation being more disposed toward comfort eating when nights (when minimum temperatures are usually recorded) are dark and cold. Again, note the same upward trend for User-Marked Spam rates as conditions became less inclement.

  • Precipitation:


Levels of rain and cloud corresponded with the temperature trends. Subscribers were more likely to respond positively to restaurant emails during periods of heavy rain (“High-Range”), while User-Marked Spam rates were significantly lower during the same periods.

  • Eat-in vs. Take-out:

Emails from brands operating a “Take-out” service model (as opposed to “Eat-in”) showed a slightly different subscriber engagement profile:


Email Read Rates were highest when temperatures occupied the “Mid-Range” category. There was a definite sense of subscribers being less inclined to order take-outs when temperatures are at either extreme. The highest  ISP-Marked Spam rates – a good measure of negative overall sentiment – bias toward the “Low-Range” category.

Using the confirmation activity from a high-volume email program such as Dominos Pizza, it was possible to validate this trend in terms of actual orders placed.


Here the data is presented as an index, where 100 represents average orders per day over the period analysed. The blue columns are particularly illustrative, showing that order activity significantly under-indexes during periods of high temperatures, and over-indexes during extremely wet and cloudy periods.

What else did we learn from casual dining email engagement metrics?

  • 94% of broadcast activity took place between Monday and Friday, but the best average Read Rates were seen at weekends – 30.1% versus 25.0% for weekdays.
  • “Payday poverty” appeared to be an important driver of User-Generated Spam complaints, which were three times higher in the weeks furthest from payday.
  • There also appeared to be a small but identifiable relationship between the phase of the moon and levels of spam complaints!

This article has considered an external stimulus that clearly impacts on the effectiveness of some email programs. There are others – buying cycles, school holidays, and special events all exert varying degrees of influence. Weather clearly has an important say in the overall effectiveness of the casual dining sector’s email marketing initiatives. Brands smart enough to harness this insight to their advantage will increase subscriber engagement and maximise attributable ROI.

Every cloud does indeed have a silver lining!

Author Image

About Guy Hanson

Guy is a passionate advocate for intelligent use of customer data to drive responsive email programs. With a knowledge base spanning the best part of two decades, he is globally recognized as an email expert and thought leader. Most recently at Return Path, Guy led a global consulting team, and worked with a broad range of world-famous clients across 5 continents to improve their email delivery, subscriber engagement and revenue. Now blazing a new trail as an independent consultant, Guy is continuing to explore his passion for email and data – and sharing his knowledge – through appearances at flagship events, providing training, and regular fresh new thought leadership Outside of work, Guy has had long-term involvement with both the DMA and IAB email councils, been a regular contributor to the industry press, and is a 3-time finalist as data storyteller of the year!

Author Archive