Prior to the pandemic, summer was predictably a time of low illness, but overall trends as of this writing in the middle of June 2022 suggest we won’t see the traditional relief. We still have lingering flu cases, a slight uptick in unseasonal RSV cases, and omicron subvariants (BA.4 and BA.5) now represent at least 13% of new coronavirus cases in the U.S., which is likely based on undercounted case data.
For retailers and brands in the illness industry (think makers of OTC medications, tissues, soup, etc., as well as the stores that stock them), this is just one example of how illness seasonality now comes with a lot of unknowns. How much product should be shipped? When is it time to restock? Will there be too much product left over or will it fly off the shelves? With COVID continuing to add an extra layer of unpredictability, seasonal and historical data alone is falling short this summer. Combine this with supply chain challenges and rising consumer expectations, and there is a very real risk of losing customers if the right shelves aren’t stocked at the right time.
Brands all over the country rely on demand forecasting to maintain customer loyalty, but it’s time to reconsider the standard way to approach data. To better understand when, where and how much to stock, consider rethinking your CPG strategy with the following tips:
1. Use first-party data.
Most illness trend forecasting uses third-party data or big aggregate data that revolves heavily around claims-based data (data that is aggregated once someone enters the healthcare system). This data tends to rely on one reporting source, which creates a problem. Forecasting using this data is frequently dated and only paints a very general picture of illness, missing parts of the population who never seek care.
In contrast, aggregating first-party illness data is impactful because it catches consumers’ symptoms prior to them entering the health care system. First-party data allows brands and retailers to anticipate consumer behavior in real time or even in advance of illness spreading from person to person in a household.
2. Reevaluate your standard predictions.
Whether you’re a retailer or a brand, you likely have a set of standard trend models you’ve relied on for the past several years. These models look at the typical spikes that happen based on seasonality and general geography and have been fairly reliable — until now. Retailers and brands can no longer count on seasonality the way they have in the past, and therefore these models are likely out of date. Instead, it’s time to create new predictive models that incorporate more locally nuanced data and account for the uncertainty brought along by COVID. First-party data will also provide additional access to more real-time data, empowering teams to be nimble with short-term decision making and more strategic in the long-term.
Retailers and brands that continue to look to the future with the industry’s “standard” models are already stuck in the past when history isn’t repeating itself.
3. Think hyperlocal.
Much illness reporting is separated by region or state, but as our country moves into a new era of illness, it’s increasingly important to narrow down symptoms and cases to even more local levels. COVID has shown us that pockets of illness can pop up quickly and not all geographies will have the same symptoms at the same time.
This hyperlocal illness lens is key to getting the right product on the right shelves at the right time — a critical goal in a time of extreme supply chain issues. County-level reporting means getting the granularity you need to make sure distribution makes sense for where a spike is occurring or about to increase. These specific insights are also important for local retailers as they aim to avoid out-of-stocks or overstocks.
4. Pull in the experts.
Having access to data is critical, but when it comes to illness, make sure to partner with experts like epidemiologists. A layperson looking at illness trends may not know if they’re seeing a significant surge or a small influx, but an epidemiologist can help distinguish between a slight uptick in illness and a true anomaly. Epidemiologists provide important context and color to data, making it even more powerful in predicting illness. With something as important as ensuring products are on shelves, retailers and brands can’t afford to ignore the experts.
While these tips can help brands and retailers make smarter product decisions now, it’s also a great time to start forecasting for the 2023 illness season. We can’t rely on the trends of years past, but we can combine first-party data, hyperlocal insights and the input of experts to paint a clearer picture of the future. More accurate forecasting and supply chain optimization keep products stocked so that when people fall ill, we can ultimately help them feel better faster.
Inder Singh is the Founder and CEO of Kinsa. Kinsa’s mission is to stop the spread of infectious illness through earlier detection and accurate prediction. Kinsa can accurately forecast the spread of infectious illness surges weeks to months ahead of time. It uses these insights to help individuals, communities, and companies to prepare for and prevent illness spread. Notably, it detected the local presence of COVID-19 in early 2020 weeks before existing systems. Prior to founding Kinsa, Singh served as the EVP of the Clinton Health Access Initiative (CHAI). At CHAI, Singh brokered a series of agreements between 70 developing countries and 20 companies that lowered the price of treatments for AIDS and malaria, enabling millions of people to access treatment and resulting in more than $1 billion in cost savings. Prior to CHAI, Singh worked at technology startups and in the medical device industry. He holds three graduate degrees from Harvard and MIT, is a member of the Council on Foreign Relations and sits on the World Health Organization’s Roster of Experts for Digital Health.