The quality of the chatbot’s recommendations depends directly on how well you train it. The chatbot relies on the shop’s product feed, but you can significantly improve its effectiveness by providing it with additional knowledge. In this article, we’ll show you what to upload to the chatbot’s knowledge base and how to do it.
Where does the chatbot get its knowledge from?
The chatbot uses two sources:
Product feed – product data (names, descriptions, prices, attributes)
Knowledge base – documents and information that you upload yourself
The richer the knowledge base, the better the answers and recommendations.
What should you upload to the knowledge base?
1. Shop policies and terms and conditions
Upload documents describing:
Returns and complaints policy
Delivery times and costs
Available payment methods
Warranty terms
This will allow the chatbot to answer customer queries such as "Can I return this product?" without having to redirect them to customer service.
2. Service descriptions and shop policies
If you offer additional services (installation, customisation, subscription), upload their descriptions. The chatbot will be able to actively recommend them.
3. Guides and buying guides
This is the most important part of training. Upload materials such as:
"How to choose the right size X?"
"Comparison of models A, B, C"
"Who do we recommend product X to?"
The chatbot will translate this knowledge into personalised recommendations during the conversation.
4. URL of the FAQ page or blog
You can provide a URL from which the chatbot will automatically extract information. Useful if you have:
An FAQ section on your website
A blog with how-to articles
An ‘About Us’ page with a brand description
5. PDF and text files
Accepted formats are PDF and text files. You can upload, for example:
Product data sheets
User manuals
Internal sales scripts
How to ensure data quality?
Take care of your product feed:
Ensure that product descriptions are detailed and complete
Add product attributes (size, colour, material, use) – the chatbot uses them to filter recommendations
Update the feed regularly so that the chatbot does not recommend unavailable products
Update the knowledge base:
After changing the returns policy – update the document
When adding new services – upload a description
After seasonal changes to the range – add relevant guides
What to avoid?
❌ Do not upload very long, unstructured documents – the chatbot handles divided, structured content better
❌ Don’t leave the knowledge base empty – the chatbot will then be limited to information from the feed
❌ Don’t forget to update – out-of-date information can confuse the customer