Collecting Primary data for Products on E-commerce Websites

Collecting the products that the client sells is essential for enhancing the intelligence of the assistant for more personalized product recommendation.

For information collection, the software used here is an AI scraping tool named Octoparse. Octoparse has many use cases, collecting the necessary data about products on an e-commerce website is one of them.

The information collected for each product are seven:

Link.

Image.

Name.

Color (if any).

Category (if any).

Description.

Price.

The AI assistant will recommend products for the audience according to the data provided

After connecting the Airtable database with the AI assistant, the bot will be able to recognize the user's preference using an LLM (GPT-3.5 Turbo for example), search for the preferred products on the Airtable database then provide multiple products to the user according to their description.

For instance:

the user: I need a black hoodie, do you sell any?

the assistant will send the following condition to the Airtable database: AND({color} = 'Black', {category} = 'Hoodie'), then will display multiple products that have the same characteristics.

Benefits of the product recommendation feature:

Increased conversion rate: By suggesting relevant products, the bot can lead to higher conversion rates.

Time efficiency: Instead of manually browsing through numerous products, customers can receive instant recommendations from the bot.

Customer satisfaction: By helping customers find products that match their needs and preferences, the bot can contribute to higher levels of customer satisfaction and loyalty.

Personalization: A recommendation AI assistant can tailor suggestions to individual users, enhancing their shopping experience.

Scalability: As the business grows, the bot can handle increasing numbers of users simultaneously without any additional staffing requirements.