Garden retail: How to prepare for unpredictable weather
Searches for Gardening and DIY terms have grown by 53% from May 2017 to May 2019. With increasing weather anomalies occurring, being prepared for the things you can’t plan is essential. We’ve put together a 3 step guide for doing this, looking at search data, sales numbers and weather data.
If we side step the sinister truth of why the uplift in temperature might be (rising temperatures as a result of climate change getting us all excited for the great outdoors, or at least the patch of grass behind our homes…), we can instead focus on how garden retailers can prepare for such unpredictable uplifts.
Gardening trends: searches peaking in May
Date range: July 2016 – June 2019
We’ve tracked 2,000+ Garden and DIY terms in our Market Intelligence platform to reveal searches for these terms now peaking consistently in May each year.
We’ve seen some particularly warm weather each May since 2016, surpassing mean temperature: the chart below shows the mean temperature (the 0.0 line) between 1981-2010. The bars that extend beneath the 0.0 line represents colder temperatures to the mean, and those above, hotter.
Mean temperature, 1981 – 2010 anomaly
In May 2019, the temperature was actually colder than the mean…
…but, as we can see in the former search volume trend chart, the British public took to the web in even higher numbers than previous years.
The warmer months at the beginning of 2019 may have also lulled consumers into a false sense of summery security, detracting from the reality of a slightly cooler May.
Why should garden retailers care?
As an industry, garden retail feels the full impact of the recent erratic weather. Being prepared for weather anomalies is essential for understanding fluctuating consumer demand.
1.Combine data sets
Avoid relying on shaky seasonal expectations; instead, view your sales figures alongside search and weather data to build a more complete prediction.
- Check search data to identify certain gardening trends and seasonal peaks
- Look to your own sales figures
- Then explore the correlation the above have with weather data
2. Identify eventuality models
Developing a process where you combine data sets allows you to prepare for every eventuality.
For example, if we look again at the Garden and DIY terms search volume chart, at the beginning of 2019 you may have expected the trajectory of the May peak to look a little different:
In reality, the peak was much higher than expected – garden retailers may not have been prepared for this surge in interest.
If we look at the weather data, it was in fact -0.2 degrees cooler than the mean temperature in May.
Why, then, was search demand significantly higher? Looking at the preceding months, we can see that temperatures between February and April were on average 1.2 degrees hotter than usual. This may not seem like a lot, but can make a big difference on consumer behaviour and sales demand. The early 2019 heatwave is something that was heavily reported in the news, with Feb being touted as the “Hottest on record in the UK”. With all of this information to hand, we can begin to build a picture of why consumers felt the sudden urge to splurge on a sun lounger, despite the cooler May temperature.
If we also take a look at sales, we can build a full eventuality model or ‘Use case’ to refer back to in future years. According to Horticulture Week, Garden Centres reported sales down by a modest 13%, but an uplift of 13% on original targets during the May bank holiday weekend, with one Director commenting “[This was] A great performance given the amazing weekend of weather last year and the cold this time.”
Following this huge peak in May, June saw garden sales fall well below 2018 levels as the reality of the cooler Spring month hit consumers.
The eventuality model in this instance:
- When both temperatures and search volumes are significantly higher in the months leading up to what is seasonally regarded as ‘Peak demand’, sales can surpass expectations – regardless of whether the actual ‘Peak’ month meets seasonal predictions.
- However, sales may struggle in the months following this ‘Peak demand’.
Of course, this eventuality model may not be true every year, and while you could make this same assumption without any data, having search intelligence, weather and sales data on record can verify your theories, and facilitate future pattern identification.
3. Use this data as your shield and weapon
There is no getting away from the fact that this data is all retrospective. As a retailer, you may even be thinking ‘Why should I care? How is this going to help me if it’s all already happened?”
But having this kind of insight to hand, can help ensure you don’t make the same mistake twice.
Eventuality models show what to do if the same situation arises again. And, the learnings you take from anomalies can help you to avoid complacency or relying on seasonal expectations, including ‘We always sell the most sun loungers in May” or ‘We expect x% revenue growth in July YoY.’
Rather than using weather as post-justification for poor performance, these models can also help you manage sales expectations prior to peak sales. I.e. if you know that previous months in the lead up to peak demand are colder than average, and that search volumes are lower, then it’s fair to say that sales will match this.
In this instance, you can put together reactive contingency plans, such as applying promotions to drum up demand or reducing stock levels, where possible.