Retail managers have been facing an extraordinary time with the COVID-19 pandemic. They had to respond to changing consumer behaviors on many fronts, including how to deliver on an old marketing stalwart: seasonal sales. Over the years more seasonal consumer goods have appeared on the market. But the typical preparations for seasonal sales went out the window during the pandemic and are unlikely to go back to what they were as the United States emerges from stay-at-home mandates and the resultant online shopping.
So how should marketers conduct a seasonal analysis? If you’re asking this question, you have to understand the seasonal shopping behavior that existed before the pandemic, and what behavior has emerged over the last 15 months. I like to call these metrics Pumpkin Spice Metrics or PSMs, given the flood of pumpkin-themed products that arrive on store shelves every fall.
Pumpkin Spice Metrics and the Seasonal Effects
Pumpkin Spice Metrics indicate what online activity typically happens during a seasonal sales period. They require no extra effort in calculation as they’re the same ones chosen for typical digital marketing campaigns, it’s just a conscious choice to examine inputs with seasonal conditions in mind. The key purpose in establishing PSMs is to determine customer segment activity from data based on known seasonal influences and campaign details. Imagine cohorts, similar to those described in this post on Google Analytics, and you get the general idea of the customer segment PSMs are meant to help identify.
So the question you must answer is what is driving the PSMs and what is emerging now.
There are several techniques to gaining this answer. One simple starting point is to simply look at regional traffic. Seasonal campaigns can differ by regions and how businesses participate in the promotions. This can result in more activity from a local region related to those events.
Related Article: Meeting Customer Needs This Holiday Season
Identifying Seasonal Search Patterns
The most useful approach given the pandemic is demarking a time between a seasonal period and time leading up to the season, then using it to compare how search patterns may have changed during the pandemic compared to seasons past. Search patterns can change between a season and the weeks leading up to the season. Marking the time periods can reveal search patterns. You can then identify if any specific keywords or phrases are appearing regularly before the season.
Those search trends are an opportunity to adjust paid search ads or content ideas for a blog associated with a site. While hashtags are meant for social media, tweets and Pinterest pins can appear in search results. Thus a search report can spark ideas for hashtags as well. The end result for this exercise is to develop a list of content ideas that fits the season and the phrases that customers are collectively speaking.
Let’s take the fall retail seasons — Halloween, Thanksgiving, Back to School — as an example. You can look at search patterns related to these events during 2018 or 2019, then examine the same kind of search with last year’s pandemic in mind. You will need to incorporate the dates when stay at home mandates appeared. California was the first state to issue stay-at-home orders on March 19. Many states revised reopening plans on July 2. Mark these dates in a search comparison and see if a significant change in search volume occurred from those dates forward. The search comparisons can reveal any alterations from a past strategy. Use this information to decide where to spend a digital ad budget or if marketing resources are better spent developing a campaign around an emerging behavior.
Related Article: Preparing for a Holiday Season Like No Other
A More Advanced Analytics Approach
Businesses usually consider PSMs as a factor in modeling advanced analytics.
One advanced technique to consider is a market basket analysis or association analysis. A market basket analysis identifies the likelihood that one or more items are purchased alongside another item. A toothbrush alongside a toothpaste or mouthwash may make sense, but other, less obvious products may also be triggered when a mouthwash or toothbrush is purchased.
The seasonal sales time period discovery activity above can be a launching point for this analysis. Identifying the lift in retail depends on establishing a minimum statistical threshold for a sale, then seeing if a lift occurs during the specified season. With changes in customer demand from the pandemic, it is likely that threshold may change, or other product associations have emerged. An analysis may uncover new approaches to how products are mentioned on website pages, in search result within an app, in a demo video, or in valuable coupons.
Seasons change. So do metrics and analytics strategy. But choosing the best strategy with pumpkin spice metrics requires changes your understanding of your customer’s interests against the new timeline the pandemic has introduced in bringing seasonal products to market.
Pierre DeBois is the founder of Zimana, a small business digital analytics consultancy. He reviews data from web analytics and social media dashboard solutions, then provides recommendations and web development action that improves marketing strategy and business profitability.