Price Optimization Models: How to Set Profitable Prices

Different price optimization models solve different challenges. Retailers typically combine several approaches, instead of relying on a single model.

Demand-based models

Demand-based models are price optimization techniques that recommend prices to meet customer readiness to buy and maximize profits, revenue, or volume. The foundation is price elasticity, which measures how price changes influence customer demand.

In practice, you’ll get three concrete outputs to help make pricing decisions, which are:

• Elasticity curves showing how demand shifts across a defined price interval.
• Forecasted sales, revenue, and profit for a specific price change, before it goes live.
• A ranked list of which competitors actually influence sales, not every name that shows up in a scrape.

Competitive pricing models

Competitive pricing optimization determines how SKUs should be positioned against competitor pricing. This works well in categories with well-known price leaders, since customers in those categories often compare listed prices before buying.

It’s a part of dynamic pricing that updates continuously, factoring in multiple market signals, including:

• Competitor price movements and target price index.
• Product and category role within an assortment.
• Customer switching behavior.

Cost-based models

Retail price optimization with cost-based models means adding a fixed dollar amount or percentage margin on top of a product’s cost to establish the minimum acceptable price. This is a common price optimization model because it’s easy to calculate and doesn’t need deep market data.

However, there can be weaknesses, such as:

• It assumes customers care about your margin structure when they don’t.
• Customers respond to what a product is worth to them, not what it costs the retailers to stock it.
• It gives away margin when customers are very price-sensitive or have competitive alternatives.

AI and machine learning (ML) models

AI and machine learning models continuously analyze large volumes of pricing and demand data, helping retailers respond more quickly to changing customer behavior and market conditions while identifying patterns that would otherwise be difficult to detect.

The factors typically include:

• Seasonality and promotion cycles.
• Inventory availability and demand levels.
• Competitor prices and customer behavior.

AI-driven price optimization software solutions like Competera can process more than 20 demand-impacting factors to refine price recommendations. Retailers can change the cost of each SKU individually, while accounting for competitor activity and their own goals. 

Portfolio optimization vs. individual product pricing

Portfolio price optimization gauges how products influence one another, while individual product pricing treats each SKU as separate pricing decisions.

Changing the price of one product triggers a chain reaction across a group of neighboring products among customers. This makes the fine-tuning of portfolio pricing a difficult task, given the thousands of latent relationships between product sales.

Portfolio optimization helps retailers answer questions that individual product pricing cannot, such as:

• How will changing the price of one product affect sales of related products?

• Which products should absorb losses while protecting overall basket value?
• Should two competing products move together or occupy different price positions?
• How will changes to a KVI influence customer price perception across the category?

Individual product pricing works at a narrower scale, identifying price points that predict how many units of a specific product customers will buy, and it maximizes the chosen parameter for that one item. Getting this right requires knowing exactly which categories are in high demand and which products within them should actually be repriced, not just which ones are easiest to adjust.

How price optimization models work in practice

Price optimization models turn raw sales and pricing data into visual patterns that retailers can act on. Retailers will be able to identify pricing opportunities that influence purchasing decisions while supporting business objectives.

Price partitioning and segmentation

Price partitioning groups products with similar value and features into a buyer perception of clusters. Building this view requires sales and median price data by product, tracked weekly or monthly, over the past year.

Customers judge those products as a set, and they are specific about the minimum, average, and maximum price they're ready to pay for each segment. Plotting SKUs on a scatterplot shows exactly where a cluster’s boundaries sit and which price range is safe within that segment.

Without partitioning and segmentation, it’s difficult to predict if customers will perceive the price as justified, and easy to set either too high or too low a price. In both cases, the potential profit is lost.

Psychological price points 

A price point is a retail price at which a product sells well, while psychological price points are threshold prices within a category where sales peak or drop sharply. Identifying them lets enterprise retailers avoid raising prices too high or discounting products that don’t really need it.

Buyers classify products into price segments or categories with clear thresholds based on subjective value. Sales are maximized around the center of these segments and strive towards zero on their borders.

Ranking the top-selling SKUs in a category by price and sales volume reveals where those thresholds sit, so pricing teams can set the right price for the right product rather than guessing. This also helps retailers avoid the trap of following their competitors into a "dead zone" and losing sales.

Optimal price intervals 

Optimal price intervals are the price ranges in which a retailer can position its prices to remain competitive without racing to the bottom. This gives retailers greater flexibility while keeping pricing decisions profitable.

Retailers use this range to set prices that can neutralize the competitors or milk their weaknesses, instead of reacting to every competitor pricing activity. Customers are generally willing to pay the same price for similar products, and the most frequently observed price from a category leader is the benchmark from which to build the optimal price.

With optimal price intervals, you’ll get a price ladder chart that visualizes:

• The current price points.
• The price points to play.
• Your price positioning against competitors.
• Typical discounts in the category.

Promotional pressure optimization 

Promotional pressure optimization measures sales volume moving through discounts across a category, so retailers can ensure promotion effectiveness without conditioning customers to wait for discounts.

Without this data, retailers may start “promo wars,” which usually lead to a loss in sales or market share. It’s impossible to determine the pricing model to use at the beginning: whether to activate or deactivate deep discounts, or offer low prices every day.

Plotting the share of sales volume on the deal against regular sales volume highlights which products are overselling on discount and which have room for more promotional cadence. After reviewing a chart like this, retailers should choose the moderate path that protects both volume and margin.

What-if simulation and scenario testing

Before setting prices, retailers can test and compare pricing scenarios using what-if pricing simulations. This approach allows them to evaluate predicted impacts on revenue, margin, and volume, and select the strategy that best supports their goals.

Changes in customer demand, competitor reactions, and inventory levels all influence results. Testing these variables before execution reduces pricing risk and improves confidence.

Price optimization software like Competera supports this process through AI-driven what-if simulations that project outcomes before prices go live. Pricing teams remain in control of the final decision while gaining greater visibility into the likely business impact of each scenario.

Why enterprise retailers need price optimization models

Enterprise retailers need price optimization models because manual pricing can’t keep pace with the number of decisions required across large assortments — and blanket discounts aren’t working anymore.

Maciej Kraus, partner at Movens Capital and Stanford guest lecturer in pricing, says that price optimization is the right tool for retailers to find a balance between profitability and long-term growth.