Repricing AI is an additional strategy available in the Repricing module. You can create multiple AI rules, assign products, and set price limits - the algorithm then handles the rest automatically without further user action.
✅ Before you start – requirements
To use Repricing AI, make sure that you have:
At least 500 active products in Dealavo.
Integration with GA4 (Google Analytics 4) or another source of transactional data, so the algorithm can analyze sales data.
⚙️ How it works
1. Assortment segmentation
The algorithm divides products into three categories based on analytics and sales data:
Traffic Builder – products that generate high website traffic.
Potential Traffic Builder – products with the potential to generate traffic.
Long-tail – products usually purchased together with others.
2. Dedicated rules for each group
Each group gets a tailored pricing rule designed to maximize its role in your business.
3. Focusing on key competitors
The algorithm only compares your prices with competitors who actually influence your sales, not every competitor in the market.
4. Testing and scaling
Start with a small group of products to test the strategy.
If results are positive, gradually expand it to the entire assortment.
5. Continuous monitoring and improvement
The algorithm constantly monitors performance and automatically adjusts actions whenever the data shows it’s needed.
Channels supported by Repricing AI
Main use case: your e-shop – this is where sales and traffic data are collected.
Prices can also be sent to price comparison sites, like Google Shopping or Idealo.
Not supported: marketplaces (e.g., Amazon, Allegro). The AI strategy does not manage marketplace pricing.
Setting up Repricing AI – step by step
If you work with different brands or product groups that have different minimum price limits, create separate AI rules for each group.
Example: One rule for Brand A with a minimum margin of 10%, another rule for Brand B with 15%.
Step 1. Open the Reprice tab
Go to the Reprice rab in Dealavo.
Click Add new pricing rule.
Step 2. Name, priority, and channels
Enter the rule name – choose something descriptive, e.g., “AI Rule – Electronics”.
Set rule priority – the system uses this to decide which rule applies if multiple rules overlap.
Choose AI rule as the strategy type.
Select channels:
Your e-shop,
Price comparison sites.
Do NOT select marketplaces – AI rule doesn’t support them.
Step 3. (Optional) enable test mode
If you want to see suggested prices without applying them:
Enable Test AI Rule.
This lets you preview recommendations without sending them to your system.
⚠️ Important: An inactive AI rule will not generate suggestions, unlike standard rules.
Step 4. Add products
Select products to include in this rule.
This step is identical to adding products in standard Dealavo rules.
Step 5. Set minimum and maximum limits
You can:
set minimum price – the lowest price you are willing to sell at,
set maximum price – the highest allowed price.
The algorithm will work only within this range.
You cannot:
choose ranking goals,
decide behavior when there’s no competitor,
manually pick competitors,
set price endings.
The algorithm decides these automatically.
Step 6. Review simulations
Once the rule is saved and in test mode or activated, go to the Simulation tab:
Here you will see suggested prices calculated by the AI.
Not every product will have a simulation at first – the algorithm learns over time.
How it works:
If the AI rule has highest priority, and there’s a simulation for a product:
That price becomes the recommended price.
If the rule is active, it will be sent to your system.
If there’s no simulation for a product:
Dealavo will look at lower-priority rules and apply them instead.
Where you will see the results
In the Dashboard Tab, two dedicated charts will show you:
changes in average margin,
sales volume,
total margin value,
other analytics like product views and add-to-cart actions.
This helps you track whether the strategy is improving your key metrics.
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article