Retail / Apparel
H&M: Shifting onsite search to an intent driven model
- Company
- H&M
- Country
- Sweden
- Adoption stage
- In operation
- Source published
- Date basis
- The date the source was published. It can differ from the date adoption started.
- How the source was checked
- Read the full source text
The work problem
According to the Voyado customer case page, H&M serves millions of customers across more than 80 markets. The page lists H&M's challenges as understanding customer intent across markets, moving beyond keyword matching to interpret context, and reducing dead ends in the customer journey. The source says ranking that relies heavily on clicks makes it difficult to account for stock availability, product newness, revenue performance and changing customer intent. H&M needed an approach that could operate at global scale while letting its teams curate experiences, adapt rankings and respond quickly to changing customer behaviour and inventory levels.
Technology and data
After evaluating a range of solutions, H&M selected Voyado, and the two companies deployed the search across markets. Rankings combine purchases, add-to-bag behaviour, revenue performance, product newness, stock availability and customer intent signals, and update in real time. H&M teams decide which products to curate and how to adjust rankings. Lars Gezelius, Acting Global Head of Digital at H&M, said some technically impressive solutions lacked flexibility, and that ease of use mattered because up to 60 markets need to work in the tool.
Results
According to the Voyado customer case page, zero-hit searches fell by 50%, exits by 30% and soft exits by 80%. Earlier, the trade publication Retail Technology Innovation Hub reported that H&M had overhauled onsite search across more than 60 markets, with exits down 30% and soft exits down 80%. H&M is now testing search results and product rankings that adapt to customer behaviour during a session, and is exploring conversational commerce in natural language.
Limits and open questions
The figures appear on the customer case page of the search vendor Voyado, which gives no measurement period or baseline. This review did not find material published by H&M itself.
Sources
- H&M overhauls its onsite search across more than 60 markets in retail tech partnership with Voyadoretailtechinnovationhub.com, Accessed
- How H&M reimagined discovery and reduced zero-hit search by 50% with Voyadovoyado.com, Accessed
Compiled from public sources. These are not results from ATF Works customers.