Sponsored content
Food retailers need to test and experiment to meet the changing needs of shoppers
by Amichay Oren 
The pandemic is still with us, but its impact on marketing is different as we move into 2022.
In the early days of the pandemic, marketing took a back seat to more urgent matters. Food retailers mainly focused on keeping products in stock so consumers could purchase essential items.
Now, nearly two years into the pandemic, retailers and consumers are gradually returning to more normal ways – albeit a new normal. Marketing is back to being important, but in some ways it’s harder to do than before. That is because shopper purchasing behaviors are harder to figure out in this new normal.
Retailers can gain some clues by analyzing purchase patterns from the past two years. However, much of that data can’t predict the future. After all, some consumers became top shoppers by buying massive amounts of paper goods such as toilet paper – a trend that likely won’t continue.
Many of the buying patterns were temporary and specific to the pandemic. At the same time, other behaviors emerged that have the potential to endure – especially if retailers reinforce and nurture them.
So how do retailers tell the difference between the temporary and the long-lasting as they map out marketing campaigns? How do they get a handle on who are their top shoppers going forward, and what are the key behaviors worth focusing on to drive loyalty and retention?
Interpreting pandemic data
It starts by analyzing retail data sets from the pandemic — even if the findings need to be taken with a grain of salt. The insights should be melded together with pre-pandemic data to gain the most accurate picture of what the future is likely to bring.
This exercise is easier said than done and needs high-level analytical capabilities and support. AppCard specializes in leveraging machine learning to support this analysis. It views data sets through a new lens and leverages the expertise of its team.
Experiments point the way forward
While smart data analysis can lead to important consumer insights, it’s only a starting point. The next step is to use these insights to test hypothesis in real time. This amounts to a three-step process: predict, experiment and refine.
Consider the following example. An independent grocer’s data shows that its pandemic shoppers made purchases in more store departments than before as their basket sizes increased. Many of the shoppers frequented a range of store departments for the first time.
This grocer hypothesizes that its pandemic shoppers avoided secondary outlets, such as mass market supercenters, in order to consolidate purchases — a trend that the retailer feels may now be reversing. So the grocer wants to test a prediction: that with the right incentives shoppers can be convinced to continue buying at these higher levels across departments.
This leads to an experiment in which the retailer unveils highly targeted promotions to incentivize shopper behaviors. The goal is to see how shoppers react. Here are some questions the experiment is geared to answer:
- Can the grocer maintain and build on pandemic buying behaviors?
- Does it understand which levers to pull to impact behaviors?
- Can marketing efforts reverse a recent outflow of spending?
- If the test doesn’t drive higher sales, does it mean products aren’t strong enough? Or prices aren’t low enough? Or some other factors to consider?
Even a failed experiment relays important findings — on what needs to be done differently.
A need for sophisticated approaches
A word of caution about this exercise: not all experiments are well constructed or statistically significant. It’s easy to let assumptions and biases get in the way. Too often experiments are designed merely to prove theories rather than to research in an open-minded way.
Retail marketers, like all human beings, can rely on their instincts for certain things, but not for the kinds of complex decision-making involved in these experiments. Automation can help streamline the journey by running hundreds of tests and calculations simultaneously.
Succeeding in 2022 and beyond
AppCard works with a wide range of retailers to analyze their data and drive testing that leads to marketing success.
Retailers need this kind of sophistication more than ever to answer key questions. They are wondering, for example, how to benefit from the increased pandemic trial of private label items. How to successfully cater to shoppers’ growing focus on health and wellness. And so on.
The answers may be different for each grocer. Those that stay open minded, analyze their data sets and embrace experimentation will position themselves for success. It’s not about survival of the largest retailers, but rather of the smartest and most agile.
