Science

THE RIGHT APPROACH TO RETAIL

Price optimization has come a long way since the early days of “black box” functionality and models with inflexible business rules and hard-edged modeling constraints. While the core concepts based on price elasticities and item interaction effects remain valid, the business best practices have come a long way.

Clear Demand Regular Pricing is a third-generation solution, refined with newly-patented tech that positions retailers to meet next-level challenges. The approach is informed by market experience, iterative improvements to the core science, and the application of Machine Learning to analyze large, dynamic data sets.

Clear Demand Regular Pricing is a third-generation solution, refined with newly-patented tech that positions retailers to meet next-level challenges. The approach is informed by market experience, iterative improvements to the core science, and the application of Machine Learning to analyze large, dynamic data sets.

Responsive Results Powered by Leading Science

We do Retail Science the Right Way

Every price optimization decision comes down to one question: “What is the effective price?”

In classical economics, it is understood that a higher price drives down sales volume. We think of this in terms of a proportionate purchasing response called “elasticity.” The relationship is seldom linear, however. Some items are highly elastic, like flat screen TVs. Others like, shoe polish, are nearly unaffected.

Demand models look for opportunities to increase margin on low-elasticity items while maintaining competitive price points on highly-sensitive items that are important to the retailer’s price image and shopper loyalty. Simple in concept, but complex in practice. Numerous interaction effects happen when a change in demand for one item affects the demand for another item in a positive (affinity or halo) or negative (cannibalization) direction.

For the regular pricing model to perform, attention must also be paid to the relative prices of items within each category – such as national brands, budget store brand, organic and premium private label, as well as differing pack sizes. Rule sets are developed to ensure that these price gaps are applied in a consistent manner that will make sense to shoppers.

The Evolution of Pricing Science

While the first-generation pricing science was a breakthrough in its day, it delivered a flood of recommendations that required many hours of sifting and intervention by human experts. In the end, the effectiveness of these solutions were therefore limited by the ability of retailers to implement numerous price recommendations.

Later solutions tried a more user-friendly approach that relied heavily on heuristics, like competitive price matching and margin formulae, to simplify and speed the process for pricing professionals. This method sacrificed mathematical rigor due to the approximation resulting from “soft” rules. The results seemed to come more easily, but they just weren’t consistently reliable.

Machine Learning Provides Trusted Prices

The application of advanced machine learning to the price optimization challenge changes the game by harnessing untiring computing power to maintain a model that is continuously refined as new data are incorporated. Pricing “hygiene” problems – where the line structure falls out of alignment – become a thing of the past.

Price professionals quickly learn to trust the vast majority of everyday price recommendations and focus their attention on reviewing a much smaller number of exceptions that are flagged by the system.

Understanding Pricing Rules

Retail pricing rules are more important today than ever to enforce a retailer’s price strategy, compete effectively and deliver a consistent shopping experience. How rules are handled inside optimization solutions is critical. Optimization cannot be solved first and rules solved second – as it has been done for years – otherwise rules violations will proliferate; pricing inconsistencies will occur and same-store sales will erode.

Using Clear Demand, every pricing rule defines:

  • The upper/lower bound on the price
  • Cost associated with the rule
  • Degree of “force” that should be applied
  • Degree of confidence in the elasticity

In an assortment of 40,000 items or more, the relative priority of the rules is calculated and adjusted using machine learning. In contrast, “weighting” schemes used by other solutions depend on fixed settings made by pricing analysts.

Download Whitepaper

Latest Content on Insights & Support

What Sets Us Apart?

Competitive Pricing. Clear Demand automates competitive price comparisons and alerts merchants when existing prices deviate from competitive pricing rules. Our unique competitive cross-elasticity model enables dynamic competitive pricing.

Merchandise Analytics. Clear Demand applies an evolved optimization which constrains prices by applying retailers’ rules for competition, margin, and product-line relationships. This minimizes or eliminates pricing rules violations.

Compliant Optimization. Clear Demand uses an integrated business intelligence capability from Tableau which provides visibility into sensitive items, competitive pricing and revenue opportunities, and other business insights with interactive and drill-down tools.

Big Data Platform. Clear Demand’s platform is architected a from the ground up so retailers can quickly capture and analyze data from any source – store, online, mobile, social media, loyalty programs – to weave a consistent and competitive pricing strategy.

Single Enterprise Rules Architecture. Clear Demand pioneered a single Omnichannel (mobile, online, in-store) rules architecture. Retailers can project a unified pricing environment so shoppers perceive “one company” across digital and physical store domains.