.jpg)
Managing travel requests in medium-sized companies often becomes a maze of emails, spreadsheets, and manual approvals. In Las Palmas de Gran Canaria, a company operating with more than twenty people in administrative tasks decided to break with that dynamic. Instead of patching its legacy system, it sought a real transformation. The solution involved building a corporate intranet with a fully automated travel workflow, integrated with its existing management tools and powered by artificial intelligence. This article analyzes how that goal was achieved from a technical and business perspective, highlighting lessons that any organization can apply.
The main challenge was not the lack of technology, but the disconnection between systems. The operations team used an ERP, a CRM, SharePoint, and Microsoft Teams, but each travel request required filling out forms on multiple platforms, attaching scanned receipts, and waiting for manual approval from several managers. Duplication errors, validation delays, and the lack of real-time visibility generated increasing operational costs. Management needed a unified dashboard to monitor travel spending, but the data was scattered. After evaluating several options, the company opted for a strategy of custom applications that would allow the system to be adapted exactly to its processes, without giving up integration with its technological ecosystem.
The architecture of the new intranet combined several key elements. First, a customized web portal was developed that centralized all travel requests, from the initial request to expense settlement. This portal acted as a single interface for employees and approvers, eliminating the need to access different applications. Underneath, a process automation engine was implemented with n8n, which orchestrated repetitive tasks: sending notifications, updating records in the ERP and CRM, and generating automatic reports. Artificial intelligence was incorporated through language models with RAG (Retrieval-Augmented Generation) to process attached documents, extract relevant data, and suggest accounting classifications. All of this was hosted on a cloud infrastructure with AWS and Azure cloud services, ensuring scalability and GDPR compliance through encryption and VPN tunnels.
One of the most innovative aspects was the inclusion of AI agents that acted as virtual assistants within the workflow. These agents were responsible for automatically validating travel policies, detecting anomalies in expenses, and escalating exceptions to human supervisors. Unlike traditional chatbots, these agents operated proactively, learning from each interaction and adjusting their criteria without constant manual intervention. The company was thus able to maintain human control over critical decisions while delegating routine tasks to the machine. Integration with business intelligence services also allowed managers to monitor in real time indicators such as average approval time, budget consumed by department, and policy compliance rate, all from a Power BI dashboard.
Cybersecurity was another fundamental pillar. Given that the system handled sensitive employee and supplier data (passports, corporate credit cards, itineraries), role-based access controls, detailed audit logging, and human checkpoints for high-value approvals were implemented. The company did not have to replace its ERP or CRM; the intranet connected through custom APIs that respected existing business rules. This approach drastically reduced manual errors and improved the accuracy of the expense settlement process, going from 78% to 92% accuracy in the first months.
The quantitative results confirmed the success of the decision. The company managed to reduce repetitive manual work in back-office tasks by more than 40%, accelerate approval cycles by approximately one third, and reduce operational costs associated with travel management by 28% during the first half of the year. The return on investment was achieved before nine months, well below the planned timeframe. These data are consistent with those observed in other organizations that have integrated artificial intelligence into their core workflows, rather than using it in isolated experiments. In fact, industry studies indicate that companies that integrate AI into core processes obtain up to five times more impact than those that use it only in pilot initiatives.
Beyond the numbers, the most valuable lesson was methodological. Defining key performance indicators before starting development turned out to be the best predictor of measurable ROI. Integration with existing systems weighed more than choosing the newest language model. The phased design —discovery, MVP, controlled deployment, and continuous optimization— minimized operational disruptions and facilitated team adoption. The company also valued the autonomy gained: a web management portal allowed business users to configure prompts, adjust thresholds, and monitor costs without depending on the IT department for each change.
This case in Las Palmas de Gran Canaria demonstrates that modernizing the intranet with a travel request workflow is not just about installing packaged software, but about designing a custom solution that combines custom software, artificial intelligence, process automation, and a solid integration strategy. For companies looking for a technology partner with experience in this type of transformation, Q2BSTUDIO has established itself as a benchmark in the area, offering everything from initial consulting to implementation and ongoing support. Its focus on AI for businesses and AI agents allows organizations not only to optimize processes, but also to build a scalable foundation for future innovations. The key is to start with a clear diagnosis, set realistic expectations, and prioritize integration. That is the path to an intranet that truly transforms daily operations.
