Skip to main content

Ensuring Transparency in Price Optimization: Unveiling the Strengths of AI-powered Pricing Strategies

If the thought of neural networks controlling your pricing seems a little daunting, that’s understandable. After all, the stakes are high. But that’s precisely why businesses are using the power of AI. They get an incredible view of not just historical performance but also where their market is headed to better predict the most winnable future prices.

In this blog, I’ll explain two price optimization problems AI is helping businesses solve today: 1) How to ensure your segmentation is as dynamic as your business; and 2) How to proactively keep up with the market.

As you’ll see, AI-powered pricing strategies utilizing neural networks can help you optimize everything—your prices, discounts, and margins—in the most beneficial way for your company, all while providing the transparency you need to know how a price recommendation was created. Because the truth is, with AI powering your prices, you’ve never had more control over your business.

“My business is dynamic but I’m stuck in a static segmentation.”

Businesses are facing intense competitive, economic, and market pressures. Your data changes depending on how you react to those pressures, and your analysis needs to capture this. How does AI ensure that your optimization is just as dynamic as your business?

The correct attributes are used, every time, to define price. Take geography. It’s an important attribute in determining price, but is it always important in the same way? Do you break it up by state or by region? Do you lump certain states together? Is geography just as relevant for large, national accounts as it is for smaller accounts?

The PROS Platform takes all of this into account to ensure that the correct attributes are used to define pricing recommendations, every time. For example, geography is almost always used as an attribute to determine customer willingness to pay. But what if you are a distributor servicing small, mid-sized, and national account customers? Geography should probably be considered differently. For a small-end customer, the geography component should probably look and customers in a similar area – that area could even be defined at a very granular level. If you are also using price optimization for larger accounts or national accounts, you shouldn’t be forced to use attributes in the same why when the customers you sell to are not always buying based on the same patterns – this type of static segmentation that forces you to always use the same attributes can drive your optimization the wrong direction.

The attributes are flexible and dynamic. This is hugely different than traditional micro-segmentation, such as the ordering of attributes by importance in determining a price recommendation. Traditional segmentation looks at all the attributes that are significant in the business, puts them together, and shows all the peers in the micro-segment as important in determining what the price recommendation should be.

But we have found that this approach doesn’t provide the full picture of what’s going on in the market. For example, micro-segmentation doesn’t accommodate numerical attributes like annual spend. If you have 10,000 customers, you might have 10,000 unique customer annual spends. That means you end up having to split those customers into various segments, such as 0-$1 million, $1 million-$5 million, and $5 million and above. But you might end up comparing a $1.1 million annual spend customer to a to $5 million annual spend customer when they might have a lot more in common with those customers that are slightly under the $1 million dollar range.

A dynamic segmentation approach avoids this rigid approach to segmentation. Plus, as you’ll see in the next section, micro-segmentation doesn’t enable you to use current cost data and market data to project what a price recommendation should be. On the other hand, PROS looks at all your data, picks which data is relevant, and uses it at the right time to optimize pricing.

And if you want to see what attributes were used to determine a price, you simply tap into the attribute section of the PROS Platform to see everything that determined a price, including what was most significant and what was least significant.

What factors influenced baseline price, graphic

New data is incorporated dynamically as it becomes available. When you’re using a pricing application, you’re going to find and capture new bits of information, such as more information about your customers and products. You might also want to include new data feeds that are relevant for pricing such as commodity indexes or competitor intelligence. How do you make sure that you use that new data without having to constantly re-segment your business? If you're not thinking about optimization in a very dynamic sense, then it's going to require a lot of rework for you to be able to constantly update your model.

With PROS and our approach to optimization, you don't have to do a constant re-segmentation or rerun of optimization. As you gather additional data points, start selling products in a different way or have the ability to incorporate external data into your pricing process, the PROS Platform eliminates all that manual work because the application will adapt based off the data you feed into it. The days of engaging your price optimization provider to re-segment your optimization every time you have a change to your data structure is over.

“How do I proactively keep up with market changes?”

PROS AI-powered price optimization enables you to stay ahead of the competition. Here’s how:

Cost and market data guide your price optimization. A traditional segmentation approach relies solely on your historical data to produce the price recommendation. Of course, it’s important to look at the historical data, including the behavior attributes that give you insight into the customer, product, and transaction. But with the massive shifts in markets and costs today, this can result in price recommendations that are too low in a market that is rising quickly – it can also leave you quickly overpriced in a down market.

The PROS Platform looks at cost information, changes in production costs, changes in labor costs or supply, and readily available market data, such as oil prices. These are great directional triggers to tell you not only what the behavioral attributes are but where the market is heading so you can set a better predicted customer price. This has been game-changing for our customers, allowing them to have a much better, more accurate price recommendation based off the current snapshot of the market.

The analysis goes deep—you have more data than you think. What if you don’t operate in a market that has easily trackable information like the oil industry? Things like changes in inventory can be tremendously important. It doesn't even have to be the exact number. Not everyone tracks information to that level of detail, but it could just be a low/medium/high indication of changes in inventory. This was the case for a manufacturer we spoke with recently. They also incorporated a quick snapshot of some distributor inventory they had on hand so we could help them see if what they're doing on the production side is matching what's happening on the distribution side.

Plant capacity and competitive intelligence are also good market signals. Do you feel like you have certain areas that are more competitive or less competitive for a certain product line in a particular channel? You don’t have to know what the competitor's price is but directional competitive information is another great signal that our algorithm can use to provide a more accurate price recommendation.

Understanding what is driving changes in margin is another huge one. You can look at period over period, actual margin analysis to see what's driving the change, whether it’s an increase or decrease in customers or changing purchasing patterns. The PROS algorithm can look at all of this.

Optimize beyond pricing. The beauty of the PROS Platform is that we can customize it to meet the unique needs of your business. We can help you think beyond pricing optimization to also include things like discount optimization, markup optimization, and margin optimization.

If you need to understand the historical impact of pricing and price realization, we can break all of that down to show you the price paid, what the price included, how successful the price was (who is buying it at this price). With the PROS Platform, you can look at what the customer actually paid, what the salesperson requested, what optimization they saw, the final approved price, and so much more. All this information in one place has proven to be incredibly powerful for companies and their profitability.

Learn more

To learn more about Ensuring Transparency in Price Optimization, view the full webinar here and be sure to join us for our next webinar in the series. Sign up today!

About the Author

Daniel Wolf currently leads the Strategic Consulting division at PROS. With over a decade of experience in AI-driven pricing and selling optimization, his insights are shaped by a deep understanding of technological advancements in pricing strategies. Based in Nashville, Tennessee, he has been a key figure in driving forward-thinking solutions in North America, Europe, and ANZ. His approach combines technical knowledge with a keen sense of market dynamics, making him easily understandable.

Profile Photo of Daniel Wolf