Using a GPT-based natural language model, iFood transformed the grocery shopping experience via WhatsApp, achieving 3-4x more conversion than traditional button-based systems.
14 de janeiro de 2026
8 min read
Source: institucional.ifood.com.br
3-4x
Conversion vs Buttons
20+
Retail Partners
AI
Natural Language
iFood, Brazil's largest food delivery platform, launched Compr.AI — a generative AI-based solution that allows consumers to do their grocery shopping via WhatsApp in a completely natural and conversational way. Unlike traditional chatbots that rely on menus and buttons, Compr.AI understands slang, abbreviations, emojis, and even incomplete shopping lists.
The result was a 3 to 4x increase in conversion compared to traditional button-based systems, proving that the naturalness of conversational interaction is a decisive factor for conversational commerce via messaging.
3-4x
Conversion vs Buttons
20+
Retail Partners
AI
Natural Language
Grocery shopping via digital channels still faced significant friction: traditional apps and chatbots required consumers to navigate complex menus, search for products one by one, and deal with rigid interfaces that didn't understand how people naturally communicate.
iFood identified that the average Brazilian consumer prefers to send messages like "2 kg of tomatoes, a pack of Nescafé, and whatever is good for a barbecue" rather than clicking through product categories. The challenge was to create an AI capable of interpreting this natural language and converting it into a precise and complete shopping cart.
Compr.AI was developed with GPT-based models customized for the Brazilian food retail context. The system understands:
- Informal language and abbreviations ("pct" for package, "dz" for dozen) - Emojis as product references (🍎 for apples, 🥩 for meat) - Context-dependent requests ("something for a barbecue" generates contextual suggestions) - Incomplete lists with intelligent auto-completion
The AI connects directly to partner retailer POS systems, checking real-time availability and prices. Currently, it operates with over 20 retail partners, covering supermarkets and wholesalers.
The experience is fully conversational: the consumer sends the list, the AI organizes by categories, suggests substitutions when needed, and completes the order — all within WhatsApp.
The main impact was a 3 to 4x increase in conversion compared to traditional button-based chat systems. Additional results include:
- Significant reduction in cart abandonment rate - Increase in average ticket due to contextual suggestions - Greater recurring engagement (consumers repeat the experience weekly) - Expansion to 20+ retail partners in the iFood grocery network
As highlighted by iFood: "Sales conversion is three to four times higher compared to button-based systems."
Catalisa offers a WhatsApp integration platform with native conversational AI agents capable of understanding natural language, connecting to POS and inventory systems, and managing the complete purchase flow — from request interpretation to payment.
With Catalisa's Building Blocks, integration with product catalog systems, inventory management, and payment gateways is simplified, enabling any retailer to create conversational shopping experiences similar to Compr.AI without the need for large engineering teams.
IFOOD INSTITUCIONAL. “Compr.AI: compra de mercado pelo WhatsApp”. institucional.ifood.com.br. Disponível em: https://institucional.ifood.com.br/inovacao/compr-ai-compra-de-mercado-pelo-whatsapp/. Acesso em: 27 fev. 2026.
Schedule a demo and see how Catalisa can transform your business
Analysis of how WhatsApp evolved from a messaging channel to an end-to-end sales platform, with data from Magalu, iFood, and Gupshup/Forrester.
19 de jan. de 2026
·
10 min read
Analysis of how the food-tech sector became a pioneer in conversational AI with three fronts — shopping, management, and mentoring — with cases from iFood and Carrefour.
17 de fev. de 2026
·
9 min read
Analysis of how AI agents on WhatsApp evolved from reactive bots to proactive business consultants, with cases from iFood (Gerente, Cris) and Magazine Luiza.
11 de jan. de 2026
·
9 min read
© 2026 Catalisa. Todos os direitos reservados.
Conforme LGPD
Meta WhatsApp Business
Dados no Brasil
We use cookies for analytics and load external resources (images) to improve your experience. By clicking "Accept", you consent to the use of cookies and third-party resources in accordance with our Privacy Policy.