How Musique Dépôt uses Hyvä and Doofinder to bring AI-guided service online
Project Snapshot
- URL: musiquedepot.ca
- Location: Saint-Jean-sur-Richelieu, Quebec, Canada, bilingual storefront (French & English)
- Industry: Musical Instruments & Accessories
- Business Model: B2C
- SKU: 5,000+ products spanning instruments, accessories, and gear from major brands
Technology used
- Platform: Magento Open Source
- Hyvä Products Used: Hyvä Theme
- Tech stack: Doofinder AI Assistant + Hyvä frontend
Results
- Conversion Rate: 41% of AI assistant conversations result in a product click
About Musique Dépôt: Quebec's Home for Musicians
Founded in 2006, Musique Dépôt has become one of the most trusted music retailers in Canada. Their flagship store in Saint-Jean-sur-Richelieu — 7,500 sq ft of instruments, gear, and accessories — draws thousands of customers a year from across the South Shore of Montreal and beyond.
But Musique Dépôt is more than a shop. They run a music school offering lessons seven days a week across guitar, piano, bass, drums, and ukulele. They have a lutherie workshop. They do rentals. Their staff are musicians, and the culture of the store reflects it — knowledgeable, passionate, and genuinely invested in helping customers find the right instrument for their situation.
That identity is easy to deliver in person. The challenge is translating it online, across a catalogue of thousands of SKUs, to customers who may not know exactly what they're looking for.
The Challenge: complex catalogue, consultative customers
Buying a musical instrument is not a straightforward transaction. Unlike categories where customers arrive knowing the product they want, music retail is full of people who arrive with a feeling — a budget, a genre, a skill level — and need guidance to get from there to a specific product.
- "I want to start learning guitar for around $200."
- "Something my kid can practice on quietly at home."
- "I'm a jazz bassist, looking to upgrade my amp."
Standard keyword search struggles with this. It's built for customers who already know what they want. For everyone else, the experience breaks down quickly — irrelevant results, confusing category structures, dead ends. On a site with 5,000+ products across a deeply varied catalogue, that's a real conversion problem.
Musique Dépôt's answer was to build a stack that could handle both: a fast, modern frontend in Hyvä, and a conversational AI layer in Doofinder that could do what their in-store team does — listen, interpret, and guide.
The stack: Hyvä + Doofinder
Hyvä Theme gives Musique Dépôt's Magento store a frontend that is genuinely fast and clean. Across Hyvä implementations, merchants have seen up to 6x increases in page speed and up to 200% improvements in website performance scores — and faster pages don't just feel better to browse, they rank better organically, bringing in more qualified traffic before a single ad is spent. The mobile experience holds up too, which matters for a customer base that increasingly browses on their phone. And because Hyvä's architecture keeps the codebase lean, integrating third-party tools doesn't come at a performance cost — the gains don't get quietly eaten by the next plugin in the stack.
That last point matters for how Doofinder fits in. The AI Assistant runs as a lightweight overlay on the Hyvä storefront — no heavy scripts, no framework conflicts, no drag on page speed. From the customer's perspective, it feels native, not added on.
Doofinder handles the bilingual reality of Musique Dépôt's market naturally — queries in French and English are both understood, and the assistant surfaces results from the same indexed catalogue regardless of language. For a Quebec retailer serving a mixed-language customer base, that's not a nice-to-have.
What it looks like in practice
A customer lands on musiquedepot.ca. They don't know exactly what they want. Instead of navigating through category menus or typing brand names they half-remember, they open the assistant and describe their situation.
Example conversation
👤 Customer: I'm looking for a first electric guitar for my daughter, she's 13. The budget is around $300.
🤖 Assistant: For a beginner at that age and budget, a short-scale guitar is usually a great fit — easier to play and more comfortable for smaller hands. We have a few options in that range. Are you looking for just the guitar, or would a starter pack with an amp be more useful?
👤 Customer: A pack would be better, yeah.
🤖 Assistant: In that case, here are two bundles that work well at that budget — both include a practice amp and everything she needs to get started. [Product A] and [Product B]. Want me to show you the difference between them?
This is the conversation that happens every day at the Musique Dépôt counter. The assistant brings it online — available at any hour, in either language, across the full catalogue.
The numbers back it up
In its first months live, the Doofinder AI Assistant on musiquedepot.ca delivered a 41% click-through rate — meaning that of every conversation the assistant had with a shopper, nearly half ended with a product click. Customers weren't just chatting; they were being guided to specific products and following through.
For a physical-first retailer where a significant portion of final purchases still happen in store, that engagement signal is meaningful. It shows the assistant is doing its job — moving customers from "I'm not sure what I need" to "this looks right".
Why this pairing works for speciality retail
Musique Dépôt is a useful illustration of a wider pattern. Speciality retailers — stores where the product knowledge gap between staff and customer is significant, and where buying decisions are consultative rather than transactional — tend to underperform online relative to their in-store conversion rates. The store experience is their competitive advantage, and standard eCommerce tooling doesn't replicate it.
What Hyvä and Doofinder address together are the two sides of that problem. Hyvä makes sure the store is fast and trustworthy — a slow or clunky storefront loses the customer before the conversation even starts. Doofinder makes sure that once the customer is there, the experience feels intelligent and guided rather than self-service and opaque.
The commercial logic compounds: faster pages rank better organically, which brings in more qualified traffic. A higher-converting on-site experience then turns more of that traffic into revenue. The two effects multiply — and unlike paid acquisition, neither erodes gross margin.
Neither solves the problem alone. Together, they close the gap between what a speciality retailer can offer in person and what they can offer online.
About Doofinder
Doofinder is a search and discovery platform built for eCommerce. Its AI Assistant helps shoppers find what they're looking for through natural-language conversation — reducing friction, improving conversion, and extending the kind of personalised guidance that used to live exclusively in physical retail. Doofinder integrates natively with Magento 2 and is a Hyvä Technology Partner.
