Niche perfumery customers do not ask if you sell fragrance. They ask whether the Le Labo Santal 33 is available in the 50ml extrait, whether your Maison Francis Kurkdjian Baccarat Rouge 540 is the original or the 2014 reformulation, how the Amouage Reflection 45 differs from the Reflection Man, and whether the Frederic Malle Carnal Flower would suit a humid Singapore evening. Those are house-specific, batch-aware, concentration-conscious questions that generic chatbots cannot answer.

SleekAI reads the WooCommerce structure most niche perfumeries already use. House (Le Labo, Frederic Malle, Amouage, Maison Francis Kurkdjian, Diptyque), concentration (EDT, EDP, extrait, parfum, oil), bottle size (15ml, 50ml, 100ml, 250ml), and note categories all live as variation attributes in wp_term_taxonomy. Stock and price per variation live in wp_wc_product_meta_lookup. SleekAI's data mapper reads the active variant at answer time, so a Santal 33 50ml question quotes the right in-stock count and current price.

The other half of perfumery support is matching scent to person and occasion. Which fragrances suit a humid climate, which performers last on dry skin, which compositions feel office-appropriate. Those answers tie to longevity postmeta, projection notes, and your in-house bottle reviews. SleekAI exposes them so the bot recommends based on your tested experience rather than generic Fragrantica aggregate scores.

Niche fragrance buying is a high-trust transaction. Customers spend $200 to $500 on a single bottle and rely on the shop to know what is in it, how it performs, and which batch it is from. A buyer asking whether your Baccarat Rouge 540 is the original 2014 release or the current formulation is testing whether you know your stock.

A buyer asking how Carnal Flower behaves in humid weather wants the answer of someone who has worn it on tropical evenings. A generic chatbot trained on aggregate Fragrantica scores gives surface answers that signal the shop does not know the catalog. SleekAI changes that calculus by reading the same variation attributes, postmeta fields, and tested longevity notes the fragrance consultant uses to advise in person.

The bot answers concentration questions with the actual variant. It quotes the batch date when collectors ask. It recommends climate-appropriate scents from your tested experience.

It explains note pyramids using the vocabulary the perfumer would use. The result is a chat experience that feels like the consultant is in the booth at 11pm. The shop sees a log of which houses, notes, and questions drive the most pre-sale chats.

That feeds discovery set composition, the next stock buy, and the next sample distribution. High-value customers are won one informed answer at a time, and a niche perfumery is built on those wins.