As other sectors, including travel and hospitality, experienced online commerce surges, the retail market shifted at a more leisurely pace. But the grinding pressures of a lengthy global pandemic – shoppers avoiding in-store shopping where possible, lockdowns in non-critical retail verticals, parents juggling remote work and homebound kids – precipitated a rapid shift. A recent global shopper survey revealed that while 78% of shoppers often or always shopped in-store prior to COVID, by spring 2021 that had dropped by half to 39%. Even post-pandemic, only 49% anticipate they will often or always shop in-store.
It’s not surprising, then, that shopper responsiveness to digital promotional offers has increased, with 55% saying they are extremely or very likely to respond to a smartphone or mobile app offer v. 41% to an in-store flyer today. Contrast that with pre-pandemic levels of 21% for mobile offers and 31% for in-store flyers and the message to retailers is unmistakable: pricing and promotions in digital channels are of premier importance to shoppers.
Fortunately, there is well-established, proven science-based technology that enables retailers to offer carefully crafted prices, promotions and markdowns in online as well as brick-and-mortar channels. Let’s look at three ways that data science can help retailers thrive in a highly competitive multi-channel environment.
1. Deliver channel-specific prices and offers that meet shopper expectations but preserve overall margins.
Shopper studies consistently reveal different shopper price expectations for in-store v. online prices. By leveraging price optimization science, retailers get science-based price recommendations down to the item-store level that factor in shopper price expectations by channel and store zone. With its ability to detect accurate demand signals amidst the noise, data science knows which items shoppers pay most attention to pricing on, and it recommends more aggressive prices on these Key Value Items, while knowing where else in the assortment retailers can safely recover margin to meet overall goals for business health.
2. Think of pricing in the context of the full lifecycle, from everyday price through promotions and markdowns/clearance.
As online shopping becomes more entrenched globally, innovative retailers are thinking boldly about ways to build more dynamic processes and automation into their key functions, and pricing is no exception. A recent retailer study revealed that 70% of retailers say they are willing to take humans out of key processes and rely on AI-powered automation and dynamic pricing, and 60% of retailers say they are already focused on putting science-powered pricing in place. As part of the shift to a more dynamic business pace, many retailers are focused on breaking down the silos that typically generate promotional offers and markdown plans independently of the teams setting everyday prices.
Instead, many retailers are taking a holistic, flexible view of pricing, recognizing that items often move from everyday to promotional pricing and back again to everyday pricing, for example. Rather than purchasing point solutions to support price optimization, promotion optimization and markdown optimization, they are turning to vendors that support the entire lifecycle. Even those who opt for incremental adoption of optimization for various phases of the lifecycle are thinking ahead to eventual adoption across the full lifecycle. Pricing teams armed with optimization are freed up from manual, repetitive processes to do what-if scenario planning and act as trusted advisors to category managers and market team members, across all channels and lifecycle phases.
3. Pricing science continues to innovate, evolve and advance.
Unlike many other applications of machine learning and artificial intelligence (AI), price lifecycle science has been delivering proven ROI to retailers for some twenty years now. Not only is it fully road-tested under demanding real-world conditions, but the machine learning algorithms also continue to get smarter and evolve as the ingest new shopper, competitive and market data and stay close to fast-changing behaviors.
In addition, data science itself continues to advance as a field. While some price optimization solutions rely on a small number of models to cover all situations, others take a more flexible approach to incorporate a range of model and optimization toolkit options to apply the best science for each retail pricing challenge. As advances take place in the data science arena more broadly, the component-style approach enables more adaptability, efficiency and performance – to the benefit of retailers and their shoppers.
Moving forward, retailers need to not only think about leveraging science to optimize everyday, promotion or markdown price, but instead think about optimizing the lifecycle and the price/discount in parallel. For example, a change in everyday price from $2.99 to $2.49 will yield a different outcome than promoting the same product at $1.99 for two weeks. This is where it’s critical to break down silos that exist today and let the science kick into high gear.
The shock waves unleashed during the global pandemic shook retail to its core. Fortunately, change presents not only challenge but opportunity, and many retailers are embracing a more shopper-centric, automated, agile business approach that helps them structure for long-term success. Data science has a key role to play in achieving that vision for retailers in the all-important price, promotion and markdown cycle.
