Conversational Navigation
A conversational navigation pattern for a complex B2B e-commerce platform that takes users straight from intention to destination. Shipped to production in 2026.
Context
The platform served a B2B e-commerce environment where different user roles coexisted (buyers, operations, logistics, account owners), and all of them landed on the same homepage. Traditional navigation served browsing well. Users who arrived with a specific goal had to dig for it.
The starting point
Task cards in first-person language under a welcome header. Clear and easy to validate, but a card grid risks becoming another layer of navigation as the platform grows.
The quick actions pattern
Users were navigating around looking for things that sometimes are a step away, and sometimes they have to keep browsing and looking for: a previous order, a delivery reference, a page they'd used before but couldn't locate. In B2B, that friction is expensive: a buyer repeating a weekly order wants the order; a logistics manager chasing a status wants the status.
The homepage hero could work as a task entry point. The interface asks a direct question.
What would you like to do today?
The interaction
The hero asks “What would you like to do today?” and types the question out on load. That motion is the hint: it shows the headline is a field, so the entry point could stay clean and editorial: just the question, ready for an answer. Once the user starts typing, the field resolves keywords as they go and surfaces matching pages, products or order references right below, so the answer is often one tap away before the sentence is finished.
Welcome!
Welcome!
Products catalog
Browse and manage your full product listings
Wireless Headphones Pro
SKU-0042 · 14 in stock
Leather Wallet Slim
SKU-0118 · 3 in stock · Low
System logic
A lightweight layer (predictable, deterministic, at the moment based on matching keywords) recognises task language, product names and references, and routes to the most relevant destination. At MVP stage, reliability came first.
It was designed to return products, orders and other destinations like settings or security. Lately it also includes some FAQs, which makes it more valuable.
It works in combination with quick action buttons for the latest links. However, typing the order number or a product name is sometimes the quickest way to get what's needed.
Future-ready
Live on the platform today, the pattern scales to a more complex matching-keywords logic, or AI. The same entry point, suggestion layer and intent-to-destination mapping could be extended to handle more flexible or contextual queries.