Rising inflation has put fashion companies in a difficult position. They can typically raise prices and risk losing buyers, or absorb the higher costs and let their margins take the hit.
Retailers who manage to walk the narrow path between these scenarios are amply rewarded. Levi’s, for example, said in its recent successful quarter that it was able to increase its average selling prices for the period by 10 percent without seeing a drop in demand, allowing it to mitigate rising costs of raw materials and manufacturing. logistics.
Levi’s relied on more than instinct and market research to find the right spot to sell their jeans.
“These decisions are driven by powerful proprietary technologies and analytics, including artificial intelligence and methodical price elasticity analysis,” CEO Chip Bergh told analysts and investors in an April 6 phone call to discuss. the results.
Levi’s first applied artificial intelligence to pricing at the start of the pandemic to drive promotions, a practice it has now implemented in 26 countries. He realized he didn’t have to discount as much as his competitors, or in some cases not at all. The company credits AI for contributing to the growth of its margins over the past year and a half. In the last quarter, gross margins hit 59.3 percent of net revenues, up from 55.7 percent pre-pandemic, as Levi’s recorded more direct-to-consumer sales, as well as “lower promotions, higher share of sales. at full price and price increases, “he said.
A growing number of companies across all types of industries are building pricing models that proponents say are more accurate and adaptable. At the extreme, some food and beverage companies, and the most famous, Amazon, use AI to enable dynamic pricing, where the cost of an item changes frequently in response to market conditions.
The fashion brands are not there yet. But more brands are, like Levi’s, exploring what’s beyond standard over-cost markup formulas and standard pricing software. While these efforts go back years in some retailers, they are receiving new attention as major economies from the US to Germany to China grapple with high inflation.
Getting results with AI, however, involves collecting and cleaning up large volumes of data, which can be challenging and time-consuming. Forecasts also become more unstable as they stretch into the future, part of the reason AI models tend to be used primarily for short-term decisions in fashion, such as determining end-of-season discounts.
Businesses may find the returns worthwhile, though. McKinsey noted in late 2018 that some fashion companies using advanced pricing analysis saw a three to six percentage point increase in margin and sales.
The price is right
Probably the biggest difference between the old and the new pricing methods is the volume and variety of data they use.
Measuring how much consumer demand changes relative to price changes is called price elasticity. Traditional pricing software estimates it using a simple, rules-based approach, according to Michael Orr, product marketing director at Blue Yonder, who developed his own AI software used by fashion retailers such as Orsay, Bon Prix, and BestSecret. . If the cost of a commodity increases by a certain percentage and you want to keep your margin, you increase the price by a certain amount. Or you can price competitively and say that a competitor has lowered the price by a certain percentage, so if you want to match or beat it to keep the sell-through, reduce your price by that as well.
But AI can also consider other types of data, such as detailed weather forecasts, to produce more comprehensive price elasticity models. Orr said Blue Yonder’s AI incorporates about 20 separate weather factors, such as dew points, predicted maximum and minimum temperatures, and sunrise and sunset times.
Levi’s, which developed its own artificial intelligence, found that the slight temperature difference between Rome and Milan is enough to influence shopping behavior in those cities, said Katia Walsh, senior vice president and head of strategy and artificial intelligence. of the company.
The company’s traditional approach to pricing was based on data from competitive intelligence and market research and “was still based on intuition and consumer surveys,” he explained. Merchants and planners are still involved, but now Levi’s can insert thousands of data points into AI models that allow them “to predict the optimal price at which a consumer would purchase each of our thousands of products in our portfolio around the world,” according to Walsh.
“They’re also specific to fit and finish,” he said. “So our iconic 501 classics, for example, we know what the optimal price is not only for the 501, but also for a specific dark finish, a specific fit for the 501 in various parts of the world.” (Levi’s now uses AI for pricing and promotions in 26 countries.)
The data can include standard elements, such as what Levi’s has charged in the past for an item and its sales history, but also weather conditions, economic outlook, consumer sentiment, and social media trends. Some of these data sources are more predictive than others, Walsh said, but added that the ability to combine disparate sources into a model is what makes it effective.
A problem recognized by both Walsh and Orr is that these large volumes of data must be cleaned of problems such as errors and inconsistencies. One model will provide very different suggestions for discounting a tank top if you confuse degrees Celsius and Fahrenheit in the weather forecast.
The pricing decisions fashion retailers make with AI still tend to be shorter-term, such as whether or how much to cut items for clearance sales, not so much for initial pricing. This is because AI models are more accurate when they predict short-term scenarios, according to Orr. Over a longer timeline, variables can change repeatedly, although he pointed out that traditional pricing software suffers from the same shortcoming.
In the first half of 2022, however, Levi’s also started using AI to set initial prices.
A potential benefit of AI’s rapid short-term forecasts, however, is that they can allow companies to respond more quickly to changing market dynamics. It is conceivable that fashion companies may even someday use AI for dynamic pricing.
“I think companies are looking into this,” said Simeon Siegel, chief executive of equity research at BMO Capital Markets. “We are in the very early stages of trying to figure out how to take advantage of dynamic prices without causing backlash.”
According to Orr, there is resistance from retailers who don’t think consumers would accept it, even if they’re already used to the prices of Uber rides, hotel rooms, and Amazon items changing over the course of the day. In fashion, retailers generally don’t even charge store-specific prices, she pointed out, noting the added wrinkle that clothes sold in brick and mortar stores tend to have price tags attached that should also be changed.
On the other hand, Siegel said that price doesn’t exist in a vacuum: buyers and retailers alike already allow the same item to be priced differently when sold in an outlet.
The idea of a smoother pricing, which allows retailers to adapt to market conditions and even potentially different buyers, is appealing to companies who envision the perfectly optimized combination of margin and sell-through.
“Are we still there in any meaningful way? No, “said Siegel.” Are we moving in that direction? I think so. “