Launch with us

Intelligent Promo Price Optimization

Launchpad's pricing and promotional AI systems can take the guesswork out of your campaign strategies revealing revenue opportunities that are overlooked.  

Schedule a Demo
dynamic pricing solutions

Launchpad Price Optimizer

Launchpad's Price Optimizer uses a surrogate model architecture that finds the optimal price for maximizing both the demand and margin of a set of products. This architecture is comprised of 2 models: a Price Sensitivity model for demand prediction, and an Optimizer for efficiently exploring the vast search space of all sets of prices for all products within a retailer's catalog.

Price Optimizer Capabilities

With Launchpad's Price Optimizer, your team can simulate pricing strategies and outcomes using real sales data from your business

Price Optimizer Dashboard

Price Optimizer's interactive dashboard allows you to select the best price performers and display a detailed breakdown of the demand change, loss opportunity and excessive inventory.

Frequently asked questions

We Are Here to Help, Just Ask

What are the requirements to use the pricing model?

To produce initial results, the model needs 2-3 years of historical purchases, promotion calendar and historical discounts, product data, traffic data and inventory.

What can we do if we don't have all the inventory data needed?

Since accurate inventory data can be tricky, we’ve found a way to curb that issue by looking at the out of stock date and summing purchases before that.

Can we use the price optimization algorithm on multiple products?

Yes, our optimizer can handle multiple products and works in such a way that no one product is highly prioritized over another - hence solving the product cannibalization issue.

What are the categories of features used by the model?

The features are primarily divided into 4 families. Macro-inherent features are date-dependent features that aid in capturing seasonality, such as US-holidays or promotional calendar; Macro-historical features describe the past-year performance, such as traffic and sales; Micro-inherent features such as product specifications and pricing; and Micro-historical features which enable the model to understand product sell-through in various time frames.

Can the pricing optimization model handle a large catalog of products at scale?

In short, Yes, our model will easily handle a larger catalog. In our implemented solution for one of our larger retail customers, there were 200 products with a price search range of $20 in $1 increments. That’s a space of 200 raised to the 20th power possible configurations.
However, even if you have a larger product catalog, we’ve got you covered as the runtime of the model has been highly optimized using techniques like directed search and early stopping.

The Next Big Idea is In Your Data

Contact us to learn how we can help you create new solutions and insights to deliver next generation experiences and grow your business.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.