Ecommerce ops · Production

How Sary Used Algolia to Improve Search & Better Understand Its Customers

The problem

Sary's early open source search solution created problems with discovery quality, keyword search accuracy, and understanding customer intent, with Arabic-language keyword support being a particularly acute gap.

First attempt

The simple open source search solution Sary used in its early days lacked adequate discovery quality, keyword search accuracy, and customer intent recognition, and could not deliver good results from Arabic keywords.

Workflow diagram · grounded in source
1
Customer search or browse
trigger
“being used for product-specific searches powered by keyword search capabilities, category searches to improve product discovery”
2
AI keyword search with Arabic support
ai_action
“This was one of the biggest problems we had before, and we found that Algolia had phenomenal Arabic support. This gave us an immense uplift across the board for all our Arabic customers.”
3
Recommend-powered product suggestions
ai_action
“in just four months since the adoption of Recommend, which is also being used both on the web and mobile apps, Binghannam is seeing strong results”
4
Backend brand and category boosting
ai_action
“marketing and category management teams have been using Algolia's AI capabilities to boost specific brands, partners, and categories”
5
A/B Testing and behavior feedback
feedback_loop
“I believe we've had a more fundamental, maybe cultural change at the organization since we implemented Algolia. We found ourselves able to change, control and experiment with search and discovery in an effortless and immediate manner.”
Reported outcome

After adopting Algolia, Sary achieved a 500x improvement in search speed and a 300% boost in click-through rates, with customers frequently buying recommended products and the team gaining the ability to experiment with and understand customer behavior through A/B Testing.

Reported metrics
Search speed improvement500x
Click-through rate improvement300 percent
Reported stack
AlgoliaRecommendA/B Testing
Source
https://www.algolia.com/customers/sary
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

After adopting Algolia, Sary achieved a 500x improvement in search speed and a 300% boost in click-through rates, with customers frequently buying recommended products and the team gaining the ability to experiment wi…

What tools did this team use?

Algolia, Recommend, A/B Testing.

What results were reported?

Search speed improvement: 500x; Click-through rate improvement: 300 percent (source-reported, not independently verified).

What failed first in this deployment?

The simple open source search solution Sary used in its early days lacked adequate discovery quality, keyword search accuracy, and customer intent recognition, and could not deliver good results from Arabic keywords.

How is this ecommerce ops AI workflow structured?

Customer search or browse → AI keyword search with Arabic support → Recommend-powered product suggestions → Backend brand and category boosting → A/B Testing and behavior feedback.

Related ecommerce ops cases