With conversational commerce, customers buy through conversation. A chatbot asks about their needs, recommends matching products from your product data, and adds them to the shopping cart. This works in the chat on your website, in WhatsApp, and on Instagram. After the purchase, the same chatbot answers questions about orders, deliveries, and returns. This way, your store provides round-the-clock advice just like an expert in-store salesperson.
This article explains how conversational commerce works in an online store. At the end, you will find a checklist for choosing the right platform.
What does conversational commerce mean?
Conversational commerce means selling through dialogue. The customer describes their need in their own words, and the chatbot translates it into product attributes. If information is missing, the chatbot asks follow-up questions. The result is a recommendation that fits their requirements.
This is especially helpful for product ranges where customers need guidance. These include products with many variants, technical products, and goods where a wrong purchase leads to an expensive return. The page for E-Commerce shows how Mercury.ai supports online stores in this process.
Which steps of the purchase journey does the chatbot assist with?
The chatbot accompanies the customer from the very first visit to after the purchase. The table shows what it does at each step.
Step | What the customer wants | What the chatbot does |
|---|---|---|
Orientation | figure out what suits them | engages them on the product page with a relevant prompt |
Consultation | find the right product | asks about their needs and filters by product attributes |
Comparison | weigh variants against each other | displays matching variants as product cards or a carousel |
Purchase | order quickly | adds the product to the shopping cart, including reorders |
Post-purchase | check status and arrange returns | provides the shipping status and registers returns |
Special case | speak with a human | hands over the conversation with the complete history to your team |
The chatbot proactively engages visitors on the website. It uses prompts and starter messages tailored to the respective page.
How does the chatbot guide a customer all the way to the shopping cart?

The customer describes what she is looking for. The chatbot asks a clarifying question because essential information is missing. It then searches your product data, filtering precisely by fields such as flavor, ingredients, or dimensions. It displays the matching product as a card with an image and price. A single click adds it to the shopping cart.
To do this, the chatbot exchanges data with your website. It understands the context of the page and can trigger actions on it. TeeGschwendner uses this approach to advise customers in their Shopware store until the right tea is added to the shopping cart.
The chatbot only recommends products from your data. It only responds with information from your sources. If an attribute is missing in the product data, the system flags the gap. Your team can then add the missing information.
Why does high-quality consultation depend on your product data?
A chatbot's advice is only as good as the data it has access to. Mercury.ai retrieves your product data directly from your shop system, PIM, or ERP. XML shopping feeds also serve as a source. As a result, prices and inventory levels are always up to date. Logged-in B2B customers see their personalized pricing.
You don't need to build a new database for this. Mercury.ai starts with the data you already maintain today and enriches it until the chatbot can work with it effectively. Mercury.ai offers ready-made connectors for Shopware, Shopify, and Magento. The guide on Connecting an AI Chatbot to Shopware details the integration process.
On which channels do customers buy through conversation?
The chat on your website is the natural starting point. There, the chatbot sees which page the customer is currently on. It can therefore provide targeted advice on the product the customer is viewing.
Mercury.ai also integrates with WhatsApp, Instagram, and Facebook Messenger. On Instagram, the chatbot responds to direct messages and mentions in stories. In WhatsApp, it messages customers using templates pre-approved by Meta. All channels use the same knowledge base and the same workflows. The article on the Omnichannel Chatbot explains how this works.
What happens after the purchase?
After the purchase, the sales assistant turns into a service agent. The chatbot provides the shipping status, registers returns, and logs complaints. Before doing so, the customer authenticates themselves in a standardized step within the dialogue. The article on Customer Service with Order Data describes these use cases in more detail.
Some cases require a human. The chatbot hands them over with the complete history to Zendesk, Userlike, or the Agent Desk from Mercury.ai. In B2B stores, the chatbot can also turn a consultation into a request for quotation (RFQ). The article on Quote Requests from Shop Chat shows how these land in your CRM with all relevant details.
How do you measure the success of conversational commerce?
Analytics & Insights shows you what happens in conversations. You can see conversation volume, automation rates, and drop-off points. You also see which products customers ask about most frequently. UTM parameters show which campaign brought a visitor to the chat.
Moreover, the conversations reveal how your customers speak. Your marketing team can use these terms for product copy and search engines.
How do you identify a good conversational commerce platform?
The following questions will help you compare providers.
Does the platform read product data, prices, and inventory live from your shop system, PIM, or ERP?
Does the chatbot filter precisely by product attributes?
Does the chatbot ask clarifying questions when information is missing?
Can the chatbot add products to the shopping cart?
Does the chatbot respond exclusively using your own sources?
Does a single chatbot serve your website, WhatsApp, and Instagram with the same knowledge?
Does the chatbot hand over conversations to your existing live chat tool?
Where does the platform process your customer data?
For an overview of the possibilities in e-commerce, see the article on AI Chatbots in E-Commerce.
Frequently Asked Questions
Does conversational commerce also work in B2B stores?
Yes. The chatbot recognizes logged-in B2B customers and provides their personalized pricing. Customers can also place repeat orders via chat, for example using a list of part numbers or a CSV file. Böllhoff uses chat to advise its customers on a broad range of fastening technology. The team receives qualified inquiries from this on a daily basis.
Which shop systems can be integrated?
Mercury.ai provides pre-built connectors for Shopware, Shopify, and Magento. Additional systems can be connected via the open REST API. XML shopping feeds can also serve as a source.
Does the chatbot on the website require a cookie banner?
The Chat Widget is also available in a cookie-free variant. In this variant, you do not need a cookie banner for the widget.
Where does Mercury.ai process data?
Mercury.ai processes all conversations at AWS Frankfurt, within Germany. Mercury.ai hosts the language models itself and does not train them on your customer data.
Would you like to find out where conversational commerce has the greatest leverage in your store? Let's discuss this in a 20-minute consultation. Talk to us.

About the Author
Mirco Schmidt
Chief Revenue Officer @Mercury.ai
Mirco Schmidt is Chief Revenue Officer at Mercury.ai. He has over ten years of experience in international and leadership positions, including at the Volkswagen Group, Club Med, and EQS Group, and holds a degree in Marketing Management from the FH des Mittelstands in Bielefeld. His core areas of expertise are project management, marketing and sales, and communications.





