Ecommerce ops · Production

ManoMano increased conversion rate by 20% in multiple markets after switching to Algolia

The problem

ManoMano's search team needed to help non-expert shoppers find the right DIY products even when using the wrong vocabulary, but their ElasticSearch-based solution was hard to configure, lacked basic ranking fine-tuning, and produced indexing bugs resulting in duplicate product listings.

First attempt

ElasticSearch required significant expertise to configure, lacked ranking fine-tuning capabilities, and produced indexing bugs that caused duplicate product listings.

Workflow diagram · grounded in source
1
Shopper searches with imprecise vocabulary
trigger
“help shoppers purchase the right products, even if they're using the wrong vocabulary to find them”
2
Algolia query processing
ai_action
“InstantSearch, Analytics, Query Rules, Query Categorization”
3
Non-developer ranking iteration
feedback_loop
“iterate on search and rankings, even without involving developers. For example, when ManoMano had an internal data issue that was negatively affecting the search experience, getting a fix up and running took only 50 minutes. In the past,…”
4
Conversion rate improvement
output
“conversion rates had increased up to 20%”
Reported outcome

After a two-week test across Italy, Spain, Germany, and the UK, ManoMano saw conversion rates increase up to 20%, prompting an early decision to adopt Algolia.

Reported metrics
Conversion rate increase20%
Time to implement a fix50 minutes
Reported stack
AlgoliaInstantSearchAnalyticsQuery RulesQuery Categorization
Source
https://www.algolia.com/customers/manomano
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

After a two-week test across Italy, Spain, Germany, and the UK, ManoMano saw conversion rates increase up to 20%, prompting an early decision to adopt Algolia.

What tools did this team use?

Algolia, InstantSearch, Analytics, Query Rules, Query Categorization.

What results were reported?

Conversion rate increase: 20%; Time to implement a fix: 50 minutes (source-reported, not independently verified).

What failed first in this deployment?

ElasticSearch required significant expertise to configure, lacked ranking fine-tuning capabilities, and produced indexing bugs that caused duplicate product listings.

How is this ecommerce ops AI workflow structured?

Shopper searches with imprecise vocabulary → Algolia query processing → Non-developer ranking iteration → Conversion rate improvement.

Related ecommerce ops cases