A shop chat generates a qualified inquiry with full specifications when it combines consultation, targeted follow-up questions, and CRM integration. The chatbot advises based on your product data and asks questions until the application, quantity, and requirements are clear. It then creates the inquiry with all details in your CRM, such as Salesforce, HubSpot, or Pipedrive. As a result, your sales team works with complete specifications right from the start.
Mercury.ai combines these capabilities for B2B shops and online retailers with complex product ranges. The article on lead quality in B2B explains how a website chatbot prepares leads in general.
What must a shop chat be able to do to generate qualified inquiries?
A shop chat delivers a qualified inquiry to your sales team when it masters four things.
Capability | How to recognize it |
|---|---|
Consultation based on product data | The chat references attributes from the shop, PIM, or data sheet. It asks follow-up questions if details are missing. |
Structured follow-up questions | The chat saves each detail as a dedicated field. It validates inputs such as email addresses or dates. |
CRM integration | The chat automatically creates the inquiry with all fields in the CRM. |
Handoff to humans | The chat transfers the conversation along with the full history to your team whenever the customer wants to speak with someone. |
Your sales team ends up with another free-text email if any of these capabilities are missing. They then have to follow up on every single detail by phone.
When should the shop chat trigger a request for quote?
Customers order standard items in standard quantities directly via the shopping cart. For some orders, however, the shopping cart is not the right fit. This applies to large volumes, special requests such as custom printing, and products that need to be configured first. Without a chat, customers write an email or call in these cases.
A shop chat offers the request for quote within the very same conversation. Your team defines in the No-Code Studio when the chatbot does this. Typical rules include a quantity threshold, a custom request, or an item without a list price.
The chatbot continues to add standard items to the shopping cart automatically. This also applies to reorders via item number lists or CSV files. An online retailer with a consultation-intensive product range uses Mercury.ai in its Shopware shop. The chatbot provides guidance there until the right item is in the cart.
What details belong in the specification?
A specification is complete when your sales team can prepare a quote without needing a callback. For most product ranges, six details are sufficient.
Detail | Why your sales team needs it |
|---|---|
Product and variant | To calculate the correct item, for example based on material, dimensions, or standards. |
Application | To verify whether the product fits the use case. |
Quantity and schedule | To calculate volume tiers and plan partial deliveries. |
Custom requests | To clarify printing, packaging, or in-house customization. |
Delivery date and location | To check availability and shipping. |
Company and contact person | To assign the inquiry to the right customer. |
The chatbot only asks for information that hasn't already been covered in the conversation. A customer asking for an acid-resistant glove has already described the application. Customers who don't know the technical terms describe their requirements in their own words. The chatbot translates this into product attributes and suggests, for instance, a nitrile glove compliant with EN ISO 374. It validates the email address, phone number, and date directly upon entry. This ensures that email addresses and phone numbers reach the CRM in a valid format.
Your customers receive answers directly from your own product data.
Your sales team can only work with a specification if the preceding consultation was accurate. If the chatbot states a feature incorrectly, sales has to correct the inquiry later. That is why the Mercury.ai chatbot only answers using sources you have approved. These include product data from the shop, PIM, and ERP, XML shopping feeds, as well as data sheets and manuals.
You don't need to build a new database for this. Mercury.ai starts with your existing data and expands it until the chatbot can advise effectively with it. Mercury Intelligence validates every answer against these sources before outputting it. This significantly reduces the risk of incorrect information. The system flags a gap if an attribute is missing in the catalog. The chatbot then switches to a predefined workflow or hands off to your team.
The request for quote itself runs as a fixed workflow according to your rules. The guide on connecting an AI chatbot to Shopware demonstrates how product data flows using a shop example.
How does the inquiry reach the CRM?
The chatbot writes the inquiry directly to your CRM while the conversation is still ongoing. Standard connectors are available for Salesforce, HubSpot, and Pipedrive. Mercury.ai integrates other systems via open REST API, webhooks, or Zapier. You can find all supported systems on the Integrations page.
Each detail from the specification is mapped to its own field. Your team determines beforehand which field receives which information. This is supplemented by the product page where the conversation started and the campaign from UTM tracking. The guide to live chat with CRM integration shows what such a field mapping plan looks like.
The chatbot recognizes logged-in customers via their shop account. It does not need to ask for the company or contact person again. In addition, the chatbot responds to confirmation from the CRM. The customer receives a confirmation directly in the chat. You can also have an email containing all details sent to them automatically.
Your team takes over as soon as the customer wants to talk.
Some customers prefer to speak with a human before submitting an inquiry. The chatbot then transfers the conversation to the Mercury.ai Agent Desk. Your team sees the full conversation history and all details the chatbot has already captured. The Agent Desk routes conversations by department, language, or workload. After the live chat, the conversation returns to a defined point in the chatbot.
The chatbot can also hand off to Zendesk or Userlike, passing along the full conversation context. The Agent Desk features bidirectional integration with Salesforce. Learn more on the Agent Desk page.
Böllhoff generates qualified leads every day from chat conversations.
Böllhoff is a specialist in fastening technology and offers one of the broadest product ranges in the industry. The company has been advising its customers with a chatbot from Mercury.ai since 2019. Customers find the right product and resolve open questions within the same conversation. These conversations generate qualified leads every day, which are routed to sales via the Agent Desk.
Johanna Neumann, Head of Marketing Europe at Böllhoff Group, says:
> "With our chatbot, our customers can quickly and easily find the right product and clarify open questions directly without having to switch channels. This allows us to provide seamless, efficient service right where our customers need it."
Your sales team provides the blueprint for the chat workflow.
Your sales team knows best what belongs in a high-quality inquiry. Compile the details they currently have to ask for by phone or email for every inquiry. This list serves as the field plan for the chat. Next, check which information is already available in the shop, PIM, or ERP. The chatbot will not need to ask for these details again. Finally, define the point at which the chat should offer a request for quote.
Let's discuss in a conversation how this workflow can look in your shop.
Frequently Asked Questions
Does the chatbot also provide pricing?
The chatbot provides logged-in B2B customers with their specific prices and tiered pricing, retrieving them live from the shop or ERP. For bulk quantities and custom requests, your sales team prepares the quote. The chatbot prepares it with the complete specification.
Can customers without an account also submit a request for quote?
Yes. The chatbot requests company name, contact person, and contact details during the conversation. Your sales team receives the inquiry with these details in the CRM.
Can customers also submit inquiries via WhatsApp?
Yes. The chatbot provides consultation via the chat widget, WhatsApp, Instagram, and Facebook Messenger. All channels utilize the same dialogs, meaning your team only needs to maintain the request for quote workflow once.
Where is the conversation data processed?
Mercury.ai processes all data at AWS Frankfurt and therefore exclusively in Germany. No data is transferred to third countries. Mercury.ai hosts the language models itself and does not call external AI providers. Your product and customer data are never used to train models. Mercury.ai signs a data processing agreement with you in accordance with Art. 28 GDPR.
How quickly can a shop chat with CRM integration go live?
A shop chat typically goes live within four to six weeks. During this period, Mercury.ai reviews your data sources, builds the dialogs, and integrates the shop and CRM. A standard connector for Salesforce, HubSpot, or Pipedrive can be ready in just a few days.
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.
