GoCoffee — Bulgaria

GoCoffee, a vending-machine network operating in Bulgaria, worked with Awara IT to deploy a Dynamics 365–centered solution that combines Customer Engagement capabilities with Azure AI to create a conversational “digital barista.” The project targeted improved in-situ customer interactions, higher repeat visits, and faster feedback collection — while keeping customer data unified in a single CRM backbone.
Awara IT implemented Dynamics 365 Customer Engagement and Power Platform components integrated with Azure OpenAI, Speech Studio and a TTS custom avatar. The result was a measurable uplift in on-site engagement, higher conversion on promoted products and a new feedback channel feeding operational and marketing workflows in Dynamics 365 and Power BI.
GoCoffee is a retail operator specializing in a vending machine network in Bulgaria (Eastern Europe). Machines average about 30 cups sold per day; the company holds exclusive agreements with Dutch and Italian product manufacturers and operates at a regional scale that favors rapid pilot-to-rollout projects.
GoCoffee faced low engagement at point-of-sale: most customers completed a single transaction with minimal opportunity for product education, upsell or feedback capture. The physical interface limited conversational support and did not capture structured customer preferences linked to individual transactions. This made it difficult to measure campaign impact, prioritize product stocking and react to recurring friction points.
Operationally, field teams lacked timely, machine-level feedback to diagnose user issues or apply targeted promotions. Marketing could not reliably target customers or personalize offers because transaction and interaction data lived in disparate systems. The company wanted an interactive, low-friction solution to increase repeat visits, raise average transactions per customer over time, and create a reliable feedback loop into sales and operations.
GoCoffee required a CRM and engagement platform that would act as the single source of truth for customer interactions, transactions and campaign outcomes. Dynamics 365 Customer Engagement (with Dataverse) matched that need: it enabled persistent customer profiles, campaign orchestration and service-case tracking while providing native connectivity to Microsoft Power Platform and Azure services used by Awara IT.
Dynamics 365 also provided governance, role-based access and compliance capabilities important for handling customer data in the EU. Its extensibility—Power Virtual Agents for conversation orchestration, Power Automate for workflow, and seamless integration with Azure AI—made it possible to combine advanced speech, natural language and avatar rendering with familiar CRM-driven business processes.
The implementation accounted for Bulgaria/EU regulatory and language requirements. All customer-facing interactions support Bulgarian and English; dynamic receipts and interaction logs respect Bulgarian VAT treatment for retail sales and integrate with the operator’s financial export processes. Personal data is handled in line with GDPR: customer profiles and interaction recordings are stored in Azure regions compliant with EU data residency expectations, with retention policies and role-based access controls configured in Dynamics 365 and Dataverse.
By centralizing customer interactions in Dynamics 365 and augmenting them with Azure AI, GoCoffee created a low-touch conversational channel that improved product discovery, increased uptake of promoted SKUs, and generated structured feedback for operations and marketing. The digital barista reduced the need for in-person guidance and produced behavioral data used to refine stocking and campaign decisions.
Beyond immediate sales uplift, the project supplied a reusable pattern for combining Dynamics 365 with Azure cognitive services so GoCoffee can scale the avatar across additional machines and geographies with consistent CRM governance and analytics.
Awara IT delivered a practical integration of Dynamics 365 and Azure AI that quickly turned passersby into repeat customers and gave us a usable feedback stream tied directly to sales and operations. — Ivan Petrov, CTO