For Escandinavia del Plata, we developed a virtual assistant for WhatsApp designed to answer incoming messages automatically, support users throughout the conversation, and route them to the right contact or department depending on what they need. The goal of the project was to create a faster, more available, and more scalable support channel capable of handling both general inquiries and more specific requests related to products, services, or commercial contacts.
The solution was implemented by combining Make for automations, ChatGPT as the AI engine, and an API hosted on the website itself as a structured information source to generate more accurate responses. Through this integration, the bot can understand user intent, retrieve relevant business information, and produce responses aligned with the client’s real context, while also applying escalation logic when human intervention or specialized assistance is needed.
In addition to the initial development, we also provide maintenance, support, and reporting for everything that happens within the system. The API used by the bot allows contacts to be stored in a spreadsheet for later analysis and reporting, while the backend also keeps contact-related data and message counts updated. The result is a tool that not only improves automated support, but also adds traceability and valuable operational data for commercial follow-up and decision-making.
Conversational automation with AI, data integration, and traceability
This project was designed as an automated support solution capable of integrating naturally with Escandinavia del Plata’s commercial operation. Rather than being a simple rule-based bot, it was built as a conversational assistant able to interpret requests, respond using real business information, and direct each interaction to the appropriate destination according to the user’s specific need.
The architecture combines Make as the automation layer, ChatGPT as the language and reasoning engine, and a custom API connected to the client’s digital ecosystem as the data source for both answering and logging interactions. This approach made it possible to build a more flexible and maintainable flow, where the information used by the bot can evolve together with the website and the business systems, avoiding static responses disconnected from actual operations.
Beyond answering and routing conversations, the solution was also designed to provide full traceability for each generated contact. Information is stored in spreadsheets to support reporting and follow-up, while also being synchronized with the backend to keep contact history and message volume updated. As a result, the project delivers value not only in customer support automation, but also in analytics, operational control, and continuous improvement of the conversational channel.