{{ searchResult.published_at | date:'d MMMM yyyy' }}

Loading ...
Loading ...

Enter a search term such as “mobile analytics” or browse our content using the filters above.


That’s not only a poor Scrabble score but we also couldn’t find any results matching “”.
Check your spelling or try broadening your search.


Sorry about this, there is a problem with our search at the moment.
Please try again later.

The weather in 2010 made front-page news across much of the planet. Washington's storms caused power outages, NYC endured insufferable heat, and the UK was brought to its knees by snow.

With retailers reflecting on their 2010 performance, the effect of the weather on online sales and how marketers can respond is worthy of investigation.

The first step is to attempt to define what constitutes a change in the weather. Broadly, there are two different types of weather change we need to be concerned about – we’ll call these seasonal and abnormal.

Seasonal weather changes

 Seasonal changes in the weather are those that are largely expected and cyclical in nature. In the UK for example, it is typically cold in the winter, hotter in the summer, rains most in January and December, and experiences on average four times more sunshine in July compared to January.

These facts don’t really change, and consumer demand therefore can most often be anticipated based on these seasonal generalisations.

For instance, a greater number of umbrellas, scarves and raincoats are sold in Q4 of any given year; more fans, air conditioners, barbeques, and garden furniture are sold in anticipation of the summer sun.

Abnormal weather changes

 Conversely, abnormal weather changes act to disrupt or alter these predictably seasonal trends, manifesting in less predictable peaks and troughs in consumer demand during the course of a given year. There are three key types of disruption at work here. We’ll call these purchase timing, direct response, and substitution.

Purchase timing

A departure from a normal, expected weather condition will often result in a consumer postponing their purchases or bringing planned future purchases forward.

For example, an unexpected bout of heavy rain will compel consumers to buy umbrellas and raincoats earlier than planned. A cold snap, particularly one including some snow, will result in earlier Christmas shopping activity as people get into the Christmas spirit earlier than they would have done without the initial excitement of snow.

These purchases would have still been made during the course of the year, but were brought forward or postponed due to abnormalities in the expected weather at any given moment.

Direct response

Direct response on the other hand refers to those purchases that would simply not have taken place at all if it were not for a departure from the normal, expected weather conditions.

For example, an uncharacteristically hot period in Shanghai resulted in a fourfold increase in the number of watermelons sold compared to any other year in that city preceding it, and directly linked to the abnormally high temperature and something that would rarely have ordinarily occurred.


There are instances where the weather doesn’t change the timings of consumer demand or create a new demand; it can also result in the expected demand for one product being substituted for the demand for another.

For instance, several weeks of unexpected heavy rainfall will result in an increase in the sale of DVDs and books (i.e. products consumed at home, away from the rain) whilst restaurant reservations and alcohol consumption will drop (as these are typically activities that require you to leave your home and people are generally averse to getting wet!).

Extreme weather

Within all three of those types of purchasing behaviour in response to abnormal weather conditions, extreme weather obviously has the most dramatic affects.

Extreme weather would be conditions which make leaving the house dangerous – be it heat, wind, rain, or snow. In those scenarios the impact on logistics becomes a debilitating concern. Heatstroke, roads being blocked by snow, flooded warehouses; all result in an immense difficulty in getting a product from A to B.

  • In those situations there would be huge deviations in purchase timings as people simply can’t leave the house to leisurely buy something that they had planned to buy.

    Also, delivery schedule uncertainty would reduce online transactions, most notably at Christmas due to the unlikelihood of presents arriving on time, but also for essentials like grocery deliveries of staple items.

  • Customers' direct response behaviour would reflect the specific demands of the weather conditions and not much else. For example, DIY equipment to fix the window that the tree flew into during gale force winds, air conditioners to combat a heat wave, etc.
  • Substitution takes a more extreme turn in these scenarios too as problems with receiving online purchases will compel people to make digital purchases that can be purchased and consumed immediately, e.g. downloadable movies, TV series, and music from Apple’s iTunes, or electronic books for their Kindle or iPad.

    The above clearly also affects the phenomenon Yahoo! calls ‘research online, buy offline’ (ROBO) as extreme weather conditions will likely change the usual balance between how much people research a purchase online and whether they make the purchase online or offline.

The online marketer’s response

Whilst the weather is a formidable force, its effects on consumer buying preferences can be managed and even exploited by a marketer willing to factor it into their planning and execution strategies.

More specifically, a marketer needs to ensure that its key online channels are doing their bit for the common good of the sales pipe – everything from natural search to affiliates.

Natural search

Natural search takes a while to have an effect so it is really only particularly useful in its plans for anticipated seasonal weather changes, i.e. those that can be broadly predicted and adequately prepared for.

On that basis, SEO campaigns that target the products or services of a particular season need to be executed around three months before the weather is expected to change consumer buying preferences.

For instance, if you want to rank for the search term ‘gloves’ then you need to have achieved a ranking by October as it is at that point that the number of searches that are made by consumers for ‘gloves’ begins to rocket, so an SEO campaign therefore needs to start in July to target that term.

Paid search

Paid search can operate on the same basis, but has a particular advantage over most other channels and methods in that it can also respond to abnormal weather changes too, and do so immediately.

For example, an online department store can exploit an unexpected heat wave by purchasing more paid search ad impressions for its fans and mobile conditioning units to reflect this unexpected demand.

It could do this incredibly quickly, assuming it has the stock levels and/or supply chain that can deliver on that demand at a time when supply-side forces are at their most unfavourable to competitors and consumers.

Paid search, in addition, has the benefit of being capable of presenting messaging and promotions that tie in directly with the nature of the demand, given that the abnormal conditions would provide you with unique insight into what the potential consumer is looking for and why.

Affiliate marketing

Your affiliate activity can also be manipulated to reflect these weather changes. For example, by offering preferential rates or particular bounties to your affiliates to compel them to push particular products or services that you expect to see a greater demand for because of the weather, and again those that you know you have the stock and supply chain effectiveness to satisfy.

The channel is better aligned with seasonal weather changes as opposed to abnormal changes as you really don’t want to be pressuring your affiliates to keep changing what they are promoting on your behalf as soon as the weather changes and most merchants simply do not have strong enough relationships with their affiliates to do so anyway.

Display & contextual advertising

Display advertising should be media planned to reflect seasonal changes in the weather but is less useful in responding to adverse weather affects, given that display advertising will need some creative work to be done.

There may be value however, depending on your sector and resources, to create a number of display ads that reflect the five or so major abnormal weather conditions (severe gales, heavy snow, heavy rain, ice, heat wave) and keep those on stand-by to be rolled out quickly and effortlessly if the weather takes a major, unexpected turn.

Newspapers use this approach for stories all the time, particularly obituaries, which are written well ahead of a public figures death and frequently updated in preparation for the absolute inevitability of the event at some stage in the future.

Social media

Social media plays a role here too. For example, by monitoring the web’s conversations around your key products and services you’ll identify what kind of demand for what kind of weather-instigated products and services you should be pushing more aggressively.

Also, you’ll discover the problems people might be having with finding various products and determine different ways of targeting through your other channels.

For example, you may discover a forum where people are discussing snow related car issues and find that people are referring to ‘winter tyres’ when you only use the term ‘snow tyres’ on your site, which then reduces the performance of all your keyword-targeted marketing activities (SEO, PPC, contextual, etc).

Email marketing

Marketing to your existing database is critical when it comes to responding to weather changes. A database segmented by consumer location will allow you to ensure that you are offering them great deals on products and services via email that have a higher likelihood of converting to sales given that you would have communicated with your clientele at the right time with the right product, satisfying a highly likely need.

In conclusion...

Roger Miller once said ‘Some people walk in the rain, others just get wet’, and he was right even in this context - the weather is going to change, it will be unexpected, it will throw you unwelcome curve balls all the time – but it’s better to embrace this reality and prepare for it, as opposed to assuming that you’re simply beholden to the whim of the universe.

Andreas Pouros

Published 14 January, 2011 by Andreas Pouros

Andreas Pouros is COO at Greenlight and a cotnributor to Econsultancy. You can connect with him on Google+.

15 more posts from this author

Comments (3)

Peter O'Neill

Peter O'Neill, Founder & Lead Consultant at L3 Analytics

Marketers and website owners should be aware too of the general impact of the weather on internet and website traffic levels.  When the weather is nice, people are outside and not on their computers using the internet.  Conversely, traffic levels are above normal when it is raining (or snowing) and people are stuck indoors.  The impact appears to be higher the first and likely last warm weekend of the year as everyone takes advantage of it to get outside - expect to come in under forecast for those days.

almost 6 years ago

Tim Watson

Tim Watson, Email Marketing Consultant at Zettasphere

Nice breakdown and categorisation of the different elements to weather! Even extreme weather can be planned for. We won't know when it happens but we can expect from time to it to happen and be prepared. I recently posted via the DMA an example of an email campaign taking advantage of weather. See http://dmaemailblog.com/2010/11/30/using-external-events-for-email-relevance/ for creative and analysis.

almost 6 years ago

Marco Botticelli

Marco Botticelli, Projects Director at Kenshoo

Great post! It looks like an opportunity to take on board for dynamic creative optimisation purposes within the DSPs/RTB space, especially in retail. Think weather XML feed plus Location data working toghether on the same creative and based on the user purchase intent/interest for a specific product category.. Does anyone have learnings to share?

almost 6 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Daily Pulse newsletter. Each weekday, you ll receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.