Pricing shapes demand and demand shapes pricing.

Traditionally, the two were much more separate discussions. Today, the power of AI creates an opportunity to blend price optimization strategy with demand forecasting to create a value multiplier for distributors.

Episode 11 Introduction

Cliff Isaacson, EVP Product Strategy joins our show today to discuss the role of AI in pricing strategy. He shares how distributors are using Price Optimization solutions that combine pricing with supply chain planning to navigate the uber-complex B2B sales channel.

In fact, according to Gartner, nearly 1,000 companies had deployed Price Optimization and Management (PO&M) solutions at the end of 2018 — a 30% increase from 2017. Furthermore, Gartner estimates that as many as 10,000 B2B companies globally might benefit from a PO&M deployment.

when you combine Price optimization solutions with demand forecasting and ai, you get a continuum of intelligence that just keeps improving over time.

How AI-Enabled Price Optimization Works

AI-enabled Price Optimization solutions bring smart analysis to both price segmentation and customer segmentation across a broad assortment of items. So distribution businesses can get closer to actual demand BEFORE they make a risky price adjustment.

These systems apply AI and Machine Learning to merge big data with computational power to deliver deep confidence in your demand forecasts. You know the financial outcome will be a good one. And that’s a massive competitive differentiator.

Unlike B2C retail businesses where demand is much more obvious, the science is different in B2B. You have volume plays, payment terms, supplier negotiations. Even the item mix is much broader and should be analyzed with a different kind of intelligence.

A fully integrated Price Optimization solution quickly identifies under-priced and over-priced items, calculates max price, and identifies your top price-sensitive customers. So you can respond with agility and impact gross profits. Contact us after you listen to the show to get ROI examples or just share your pricing challenges.

Episode 11 Show Notes

Todd: Earlier we talked with Rod Daugherty about AI and machine learning in supply chain planning. Now we’re going to explore the role AI plays in your pricing strategy. We are joined by Cliff Isaacson, Director of Pricing Solutions. Cliff, Welcome!

Cliff: Thanks for having me, Todd. Yes, I want to start by saying that the Art of Pricing in an AI-driven world involves a great deal of science.

There are several points I want to make today:

• You have to look at price sensitivity, which is very difficult to determine and ultimately the customer’s willingness to pay.
• It’s even more difficult for certain items and industries such as distributors and manufacturers selling on B2B channels.
• If you want to get better at determining price elasticity, you HAVE to have Machine Learning.
• Pricing segmentation best practices vary by industry, but all can benefit from Machine Learning.
• And finally, pricing strategy approaches depend on customers and industry, but are gaining early benefits from AI.

Todd: I’ve heard you talk a lot about retail vs. industrial price optimization. How does the science differ in each of those applications?

Cliff: Science in B2C pricing: elasticity measurements, pricing segmentation, inference, competitive sensitivity:

1. Robust price changes, demand data, quick response
2. Highly price sensitivity
3. Direct competitive substitutes

Science in distribution and manufacturing:

1. Larger assortments, lower demand at long-tail
2. Fewer direct competitive substitutes
3. Demand obfuscated by B2B selling: discounting layers, payment terms, sales quoting, volume discounts
4. Demand less driven by price change; forward buying

Pricing segmentation

1. B2C science
2. B2B pricing art: attributes, value-based pricing
3. New item pricing, Machine Learning

Customer segmentation:

1. B2C: Customer personas, zoning, cannibalization & affinity, price sensitivity by customer segment
2. B2B: price waterfall, margin leakage, tiering, customer price sensitivity by product

Pricing strategy aligned to business goals:

1. B2C: segmentation, alignment across groups
2. B2B: customer discounting and segmentation, which items to apply Temporary Price Reductions/Pulses
3. Both: use science to identify under-priced and over-priced products

Todd: Let’s dive into the science of finding “the perfect price.” Is there a magic formula?

Cliff: The hypothesis is that there’s a “perfect price” for any given item, location or channel, time of year, day of week, time of day, given customer, etc. It’s a matter of whether this perfect price can be determined with certainty & confidence, and how frequently?

Similar to demand forecasting and replenishment: remove seasonality, day of week, macroeconomic, and micro-economic external factors like events, promotions to get to baseline demand then understand pricing influence.

Cliff: Price elasticity is determined by…

• Traditional methods using de-seasonalized, baseline demand
• Advanced techniques to determine price sensitivity through Bayesian inferences, direct vs. indirect, negotiated volume pricing

Todd: Are some calculated perfect prices more or less accurate?

Cliff: [explains]

Todd: Alright so let’s get to the “meat” here. The reason we’re here… How can Machine Learning help in price determination?

Cliff: Some applications for Machine Learning used in price optimization include…

Gamification – use expertise of pricing team to capture where science recommendations are overridden, learn from them to improve over time

Todd: Tell us about pricing segmentation. What are the B2C and B2B applications?

Cliff: [explains B2C approach]…

Overlay price sensitivity over traditional product groupings, shared attributes to distinguish where more competitive influence should be applied

Todd: So how do we apply this knowledge?

Cliff: [explains]

Todd: So how would we turn this into a pricing strategy for retail customers?

Cliff: [explains]

Price sensitivity by customer, willingness to pay for each product they purchase

List price determination first before determining discounting to get to net-pricing programs; even MORE important if the program will be in place for extended periods

Todd: Alright, Cliff, help us tie all of this together. What do we need to know right now to leverage Machine Learning in our pricing strategy?

Cliff: [explains]

• By market
• By Channel
• By customer segment
• By pricing segment
• Use low-risk, science-driven approaches to inform; improve each
• Leverage a natural feedback loop where more price changes mean more confidence in financial outcomes

Todd: What new Machine Learning tools should we keep our eye on?

Cliff: AutoML will change how businesses use Machine Learning in pricing through REST APIs… [explain]

New tools, but also new process like gamification, which allows learning from organizational pricing expertise (learn more here).

Summary

1. Price elasticity is difficult to determine.
2. It is even more difficult for certain items and industries. Manufacturers must rely on art more than black box science approaches.
3. Better price elasticity determination requires Machine Learning.
4. Pricing segmentation best practices vary by industry, but all can benefit from Machine Learning.
5. Pricing strategy approaches depend on customers and industry, but are gaining early benefits from AI.

Todd: [thank you and closing]

Where to Learn More About AI-Based Planning and Pricing Optimization Solutions

Merging AI into your pricing strategy and demand forecasts creates a value multiplier for your business. If this isn’t something you have thought of, now is the time. You’d be surprised at the fast ROI possible.