Behind every smart recommendation lies a continuous learning process. We reveal how Ziadah's system collects signals and transforms them into suggestions that increase your sales, and what this means for you as a merchant.
The Data Journey: From Click to Recommendation
Every time a customer visits your store, they leave digital footprints. The AI system captures these footprints and turns them into knowledge.
Phase One: Collection
What is collected?
Pages they visited
Products they added to favorites
Products they added to cart then removed
Filters they used in search
Time spent on each page
Products they ultimately purchased
Phase Two: Analysis
The system compares this customer's behavior to thousands of previous customers. It finds: "73% of customers who viewed these three products also bought the fourth product."
Phase Three: Suggestion
Based on the analysis, the system displays the recommendation at the most appropriate time and optimal placement.
Types of Patterns AI Discovers
Complementary Patterns
Customer buys a phone case → likely needs a screen protector and earphones.
Cyclical Patterns
Customer bought coffee 30 days ago → likely running out now.
Seasonal Patterns
In February → increased demand for couples' gifts.
Segment Patterns
Customers over 35 in Riyadh prefer products from well-known brands.
Data Privacy and Protection
What we never collect:
Full name or ID number
Payment information
Personal contact details
What we collect:
Anonymous behavioral data
Purchase patterns without linking to a specific person
How we protect data:
Full encryption of all data
We never sell any data to third parties
Compliance with Saudi data protection regulations
What Does This Mean for You as a Merchant?
In the First Weeks
Recommendations will be general — based on common patterns in your sector. Don't expect instant miracles.
After One Month
The algorithm will start learning the characteristics of your store's customers specifically. You'll notice an increase in recommendation click rates.
After Three Months
Very high accuracy. Recommendations will be customized for each segment of your customers, and you'll see a clear improvement in average order value.
Tips for Getting the Best Results
Ensure your product categories are accurate — AI relies on them
Don't add too many exception rules — let it learn freely
Review the recommendations report weekly — to ensure suggestions make sense
During seasons, manually review the strategy — then return control to AI
