Built for the way perfume stores sell
How do you prepare a perfume store for an AI sales assistant?
Updated: · By the WebVitas AI team
What information belongs in your product catalog?
Give each product a clear name and distinguish sizes or variants consistently. Include its price and describe scent characteristics in language customers understand. “Fresh citrus with a woody base” is more useful for matching preferences than “our best fragrance.” Use descriptions supported by the actual product, not invented longevity claims.
Check common sources of confusion: two products with similar names, a discontinued size, a price copied from an old list or a bundle presented as a single bottle. Correct the catalog before asking the assistant to sell from it. Adding more prose will not repair contradictory product facts.
Should you add products manually or use a spreadsheet?
Manual entry is a reasonable starting point for a small catalog. Spreadsheet import is available on Growth and Pro. Before preparing a large upload, confirm the current accepted format in the dashboard or with support; this guide does not assume column names or an import screen that has not been verified.
Begin with a representative sample and review the result. Include different scent families and sizes so you can spot inconsistent labels. Keep the original product information available to your team for corrections and future updates.
Which buying conversations should you test?
Test preferences, uncertainty and exceptions. A customer may know the scent family they want, ask for a gift without knowing the recipient’s taste, or request a size you do not stock. Review whether the assistant uses the catalog accurately and asks a relevant follow-up instead of making up details.
Then test order capture with non-personal sample data. Check the selected product, size and quantity, followed by the contact and delivery details. Finally, test the human handoff available on your plan and make sure the responsible person knows how to continue the conversation.
What should your team review after launch?
Review questions the assistant could not answer well and determine whether the catalog needs clearer information or a person should handle that request. Update products when sizes, stock or prices change. Keep someone responsible for notifications and unusual orders.
Measure useful outcomes such as correct order details and relevant recommendations using your own reviewed records. Do not turn a small sample into a general conversion claim. A store’s results depend on its catalog, traffic, customer service and fulfillment as well as its software.
Should I upload customer data to test the catalog?
Use invented sample contact and delivery details for tests. Catalog preparation requires product information, not a customer list.
Can the assistant fix inaccurate prices automatically?
Do not assume it can. Your store is responsible for maintaining accurate product information and reviewing the results before launch.
Does this guide guarantee more orders?
No. It provides a preparation and review process. Sales results depend on the store and its customers; no order or conversion uplift is promised.
Give every fragrance question a useful answer.
See how a conversation can lead to a recommendation and an order.