Suggest products that fit, instead of showing a list of bestsellers

Customers rarely describe an item. They describe a situation: a present, a problem, a budget. The bot searches your catalogue for what's described and shows the results as product cards with image, price and buy button.

With the Shopware or Shopify connection it searches the catalogue live. In Shopware it also knows what's in stock right now and won't suggest an item that has been sold out for weeks.

If a question is too open, the bot asks a follow-up, for example about size, intended use or price range. For typical buying situations you can also set this up in advance: in the Flow Builder you build a short sequence with buttons, such as “Who is it for?” and “What's your budget?”. After that, the AI handles open questions again.

Follow-up questions about a suggested product are answered from your product data and descriptions. Whether accessories are needed, how an item fits, what's in the box.

The bot only recommends from what's in your catalogue and texts. If the description doesn't say what branch thickness a pair of secateurs is made for, it can't know. It says so rather than making something up.

Whether a spare part really fits a particular appliance is best confirmed by someone who knows. In those cases the bot hands over to your team.

The analytics show which questions come up most and which ones the bot couldn't answer. If people keep asking about maximum cutting capacity and it isn't stated anywhere, you know which product text needs one more line. Every addition makes the next recommendation better.