Your Sales Report is a dashboard dedicated to analyzing the commercial impact of your iAdvize conversations. It gives you a clear and reliable view of how your conversations influence your online sales by comparing your performance with that of your overall website.
Prerequisites
To populate the sales report, iAdvize needs to receive data for each transaction completed on your website: the amount, the unique transaction ID, and the details of the products purchased (ID, quantity, price).
Four possible scenarios depending on your current implementation:
1. Your website already pushes a purchase event from GA4 or UA (EEC) into the dataLayer
✅ No further action is required. If your dataLayer pushes a purchase event correctly formatted according to the Google Analytics standard, iAdvize automatically detects and collects the relevant information (amount, transaction ID, and product list).
Expected format reference: Google documentation — purchase event (GA4).
2. You use the iAdvize app for Shopify
✅ No further action is required. The iAdvize app for Shopify automatically handles the transmission of transaction data, including the details of the products purchased.
3. You install iAdvize via Google Tag Manager
If you manage your iAdvize tags via GTM, refer to our installation guide to configure the iAdvize GTM transaction tag:
👉 Install iAdvize via Google Tag Manager — Transaction tag
A Google Tag Manager bug ("invalid input" error) currently prevents the transaction tag template from being updated. If you are affected, uninstall the template and then reinstall it from the GTM gallery.
4. You integrate the transaction tag directly on your website (manual integration)
If you call the transaction tag directly in JavaScript from your website, refer to the manual integration documentation:
👉 Integrate the transaction tag manually
id, quantity, price).Indicator definitions
- Data collection : this report requires details of the products purchased, which were not collected by the previous version
- Reference date : sales are dated at the close of the conversation (versus the actual transaction date in the previous version)
| Indicator | Definition | Formula |
|---|---|---|
| Revenue | Total revenue generated by products purchased after a conversation with the Shopping AI Assistant | / |
| Conversion rate | Percentage of conversations that resulted in a sale | (Number of orders ÷ Number of closed conversations) × 100 |
| Average order value | Average amount spent per transaction after a conversation with the AI Assistant | Revenue ÷ Number of orders |
| Orders | Total number of completed orders after a conversation with the AI Assistant | / |
| Revenue per conversation | Average revenue generated by each closed conversation with the AI Assistant | Revenue ÷ Number of closed conversations |
| Assisted sales | Total number of sales made following a conversation fully automated by iAdvize | / |
| Attributed sales (beta testers) | Total number of sales made following a conversation fully automated by iAdvize where at least one product purchased was mentioned or recommended. The amount counted corresponds to the total transaction value. | / |
Filters
Resolution status
This applies only to your conversations handled by your Shopping AI Assistant (fully or partially automated). Conversations handled exclusively by your human agents are not taken into account.
It allows you to analyze the quality of the responses provided by your Assistant and assess the extent to which it was able to address the visitor's request.
The level is automatically determined by artificial intelligence after each conversation is closed, based on all exchanges between your Assistant and the visitor.
Here are the possible resolution statuses:
| Status | Definition |
|---|---|
| Resolved | The Assistant responded to the visitor. The response is either complete or useful but still requires action on their part. |
| Missing information or transferred | The Assistant did not provide a solution: no matching product or information was available, or the conversation was transferred to an agent. |
| Unclassified | Everything else: a response that did not match the request or a conversation without a question or issue to address for your brand. |
Degree of automation
| Value | Definition |
|---|---|
| Fully automated | The conversation was handled entirely by an AI Assistant (or a workflow), without any human intervention. |
| Partially automated | The conversation was handled both by the Assistant (or a workflow) and by a human agent. |
| Not automated | The conversation was handled exclusively by a human agent. |
Two complementary levels of analysis
This report presents two distinct levels of analysis:
1. Global indicators
Analyze your performance at the transaction level. When a transaction is attributed (at least one product discussed was purchased), the total order amount is counted as Attributed Sale (beta).
Each indicator is clickable and displays a graph showing its evolution over time.
2. The "Top Performances" table
Analyze performance at the product level. Only the amount of the specific product discussed and purchased is counted in that product's revenue.
Example: A visitor discusses product A priced at €100, then purchases A + B + C for a total of €500.
→ In the global indicators: €500 in Attributed Sales
→ In the Product A table: €100 in revenue
Your top-performing products and categories
- Search : use the search bar to filter by name or ID (product or category)
- Sort : click the column headers to sort according to your criteria (conversations, orders, revenue or conversion rate)
- Export : export the table in CSV format (current view or the first 10,000 entries)
- Read conversations : access the conversations report to read the relevant conversations
Product
- Product name and ID
- Conversations : total number of conversations where this product was discussed
- Orders : sales volume for this discussed product
-
Revenue : total revenue generated by sales of this product. It corresponds only to the amount of the specific product discussed and purchased (unit price × quantity).
The unit price used to calculate revenue per product is the price transmitted by your website at the time of purchase : specifically, this is the price present in your e-commerce data layer when the transaction is recorded. If your prices change or differ from one website to another, the price in effect at the exact time of the transaction is always captured — whether it is a promotional price or the standard price.
If a visitor purchases several products in the same transaction, but discussed only one, the table will display only the amount of that discussed product, while the global indicators will count the total transaction amount. - Conversion rate : percentage of conversations that resulted in the purchase of this specific product
- because the visitor clicked on a starter linked to this product,
- because the AI Assistant presented it to the visitor.
Categories
The categories displayed come from the product catalog that you provided to iAdvize. They correspond to your catalog's categorization structure.
If no categories appear:
- check that your catalog includes a field for your categories (example: product_types) and that it is populated
- If your catalog is up to date, this may mean that no product belonging to a category was discussed during the analyzed period.
- Category name and ID
- Conversations : total number of conversations where a product from this category was discussed
- Orders : sales volume for a product in this category (following direct interaction)
- Revenue : total revenue generated by sales of products in this category
- Conversion rate : percentage of conversations that resulted in the purchase of a product from this category
Conversations
FAQ
- Why doesn't the revenue in the "Top Performances" table match the total Attributed Sales?
This is normal. These two indicators analyze different levels:
- Attributed Sales (global indicators): total transaction amount where at least one discussed product was purchased
- "Top Performances" table: amount of the specific products discussed and purchased
Example:
A visitor discusses Product A (€100) with your AI Assistant, then purchases :
- Product A: €100
- Product B: €200
- Product C: €200
- Total order: €500
Result:
- Attributed Sales: €500 (the entire transaction)
- Product A revenue: €100 (only the discussed product)
- Products B and C: do not appear in the table (not discussed)