Pricing optimization explained: what it is and what it needs from your data

Pricing optimization is the use of your own sales data to find the price most likely to meet a goal, usually the most gross profit you can earn without losing deals you want. In retail it means estimating how demand changes with price. In B2B distribution it means learning, from past quotes, how the chance of winning changes with the price for each kind of customer and item, then recommending the price with the best expected result.

How pricing optimization works

1. Gather the history

   Invoice lines, quotes won and lost, costs, customers and items.

2. Group what is alike

   Segments of customers and items that behave the same way on price.

3. Model the response

   How the chance of winning, or the quantity sold, moves with price.

4. Recommend and learn

   Suggest a price per line, record what happened, and update.

The goal decides which price wins. A business optimizing for margin picks a different price than one optimizing for revenue or share, from exactly the same data.

Retail optimization vs B2B optimization

Classic elasticity models need many sales of the same item at different prices to the same market. Distributors rarely have that: each customer buys a different mix, at negotiated prices, a few times a year. Win and loss on quotes is the signal distributors do have in volume, and it answers the question a rep actually asks: at this price, will I get the order?

A worked example

A rep is about to quote 300 6205-2RS bearings to a contractor. Landed cost is $11.60. Past quotes for this kind of account and item show how the chance of winning falls as the price rises.

Across 300 units the higher price is worth about $94 more in expected profit on this one quote, even though it wins slightly less often. Repeated across thousands of quote lines, that is the margin pricing optimization finds. It also works the other way: on lines where a small cut raises the win chance a lot, the optimizer recommends the cut. To see how strongly one product reacts to price, plot its history in the price elasticity calculator; to estimate what optimization is worth across your revenue, use the pricing ROI calculator.

What pricing optimization needs from your data

• Invoice lines

  Two years of sales with price, quantity, cost and customer.

• Lost quotes

  Quotes that did not convert, with the price that was offered.

• Costs

  Landed cost history, as well as today's standard cost.

• Customers

  Segment, size, region, terms and price level.

• Items

  Category, brand, substitutes and how often they sell.

• Market

  Competitor prices where you can match items reliably.

Most ERPs hold the first, third, fourth and fifth already. Lost quotes are the gap. Many teams delete a quote that did not convert, or never enter it, which leaves the model looking only at wins. Starting to record losses, with the price offered, is the single most useful step before any optimization project.

Common data problems

• Units of measure that change between purchase and sale, so a box price looks like a unit price.
• Costs that were never updated after a supplier increase, so margins look better than they were.
• One-off deals and stock clearances mixed in with normal sales, which teach the model the wrong price.
• Customer records duplicated across branches, so one account looks like three small ones.

None of these needs a perfect fix before you start. A good optimization tool flags them, leaves the doubtful lines out, and gets better as the data gets cleaner.

Common misconceptions

• "You need millions of rows." A mid-sized distributor's two years of history is enough to begin, because segments pool similar customers and items.
• "It replaces reps." It gives reps a starting price and a reason. The rep still owns the customer.
• "It is a black box." It should not be. A recommendation without a reason gets overridden, so the reasons are part of the product.
• "It always raises prices." It finds the price with the best expected result, which is sometimes lower.

The difference between this and dynamic pricing is covered in pricing optimization vs dynamic pricing. Pricing optimization software brings both together: it learns from your won and lost quotes and recommends a price per line, with the reasons, inside the margin floor you set.