Overview of the Smart Pricing app

The Smart Pricing app uses machine learning models to generate data-driven pricing recommendations, helping you find the best prices for your products and optimize your store's profit. You can review and apply immediate markdown and markup recommendations based on your store's sales and inventory data, or run A/B pricing experiments to test price adjustments against your current prices before committing to permanent changes. Any pricing changes are applied by you, so you remain in control of your product prices at all times.

Requirements for using the Smart Pricing app

The features of the Smart Pricing app have different requirements. Refer to the following sections for a complete list of requirements for each feature:

Role permissions for Smart Pricing

In order to use Smart Pricing app, a staff member must have at least the following role permissions:

Products > View > View cost
Products > Create and edit > Edit cost
Products > Create and edit > Edit price
All Catalogs permissions
All Markets permissions
App permissions to Smart Pricing

Learn more about store permissions for user roles.

Considerations before using the Smart Pricing app

Before using the Smart Pricing app, review the following considerations:

Your store is eligible to receive only 1 type of AI-driven pricing tip. If you have access to the AI-driven price tips in A/B pricing experiments, then you're not eligible for markdown and markup price tips. Learn more about the different types of AI-driven price tips.
The Smart Pricing app isn't compatible with third-party personalization apps.

Considerations about AI-driven price tips

If you want to implement pricing changes based on AI-powered price tips, then review the following additional considerations:

The Smart Pricing app optimizes for profit at the store level, rather than the product level. This means that in a single experiment, some price tips are increases and some are decreases, rather than only increases. Learn more about cross-price elasticity.
The Smart Pricing app doesn't adjust your product prices in real time for customers based on demand. Insights are presented to you, generating a single stable price to use over time.
The Smart Pricing app typically generates small adjustments, as opposed to larger price swings.
Only 1 price tip is generated per product. This means that for products with variants that have different prices (for example, different prices for a small, medium, and large variant sizes of a product), only 1 price tip is generated which applies to all variants during an experiment. You can choose to exclude these products when creating an experiment.
Price tips aren't generated for different customers or customer segments based on any specific characteristics. Price tips are always generated to apply to your entire customer base and, during an experiment, are randomly assigned to 50% of the traffic.

Types of AI-driven pricing tips available in Smart Pricing

Using data based on market trends, product conversion rates, and product price history, Smart Pricing generates price tips that you can choose to implement to optimize your product prices to increase store profit. The Smart Pricing app offers the following types of pricing tips:

Markdowns and markups: These pricing tips are based on your store's inventory and sales, adjusted for seasonality and trends. These price tips can be immediately implemented, with no A/B experiment or trial period required. Learn more about markdowns and markups.
(Early access) Pricing tips for A/B price experiments: You can create price experiments with AI-driven price tips to help you test and implement product price optimizations. Smart Pricing generates price adjustments and then allows you to run A/B experiments to test the impact of pricing adjustments over a trial period. You can then implement the lessons learned during the experiment to choose the best prices for your products on your online store.

How AI-driven price tips are generated

The Smart Pricing models work by learning product demand dynamics from your product performance over time, your product and shop characteristics, as well as broader market trends.

When generating a price tip for markdowns or markups, tips are based on your store's sales and inventory, adjusted for seasonality and product trends. These types of price tips refresh on a weekly basis. Learn more about how price markdowns and markups are generated.

When generating a price tip for an A/B experiment, the key data points are the product's sales and conversion rates at different prices over time, focusing on changes in average selling price and units sold. Metrics from other products in your store are also considered. Business category and industry data are important, but less influential to the model. The model isn't trained on non-public price information, geographically localized information, or personally identifiable information (PII) about customers, and Smart Pricing doesn't use such information to generate price tips.

There's no direct benchmarking against other brands or products. The Smart Pricing model takes into account products across multiple categories in order for the model to learn larger market trends more broadly.

New price tips are generated daily until an experiment is created and is actively running. Some products might not have an associated price tip, depending on whether the model has enough information over a sufficient period of time to generate an actionable price tip.

In order for a price tip to generate, a product must meet the following requirements:

The product was created more than 30 days ago.
The product must have sufficient sales.
The product's price isn't already optimal, according to the Smart Pricing model.
The product's price or cost per item values haven't been changed in the last 30 days.
The product hasn't been included in an experiment within the past 3 months.

Data handling and privacy with Smart Pricing

To learn how individual products will perform at different price points, millions of data points are aggregated and anonymized, and then used in the Smart Pricing model, based on hundreds of factors.

Your app use, feedback, and results help the model make better predictions over time.

Shopify doesn't use personally identifiable information (PII) to train machine learning models for or create price tips in the Smart Pricing app. All data is aggregated and anonymized to ensure no confidential information from either you or your customers or other merchants enters our models. No non-public price information, geographically localized information, or PII about customers is used in our models or to create price tips.

Shopify also doesn't share your confidential, non-public, or proprietary information or data with other merchants.

Learn more about privacy.

Submitting feedback about Smart Pricing

You can provide feedback regarding your experience using the Smart Pricing app. Sharing your thoughts is optional, but with your input, the app can grow and improve.

Steps:

From your Shopify admin, go to Apps > Shopify Smart Pricing.
Click Give feedback.
In the Share feedback dialog, include a summary of how Shopify can improve the Smart Pricing app.
Optional: If you'd accept being contacted by Shopify about your feedback, then select I'm open to a follow-up conversation about my feedback.
Click Share feedback.