Reimagining Leadership Reporting with AI.

The leadership team of a fast-growing AdTech organisation was operating without a reliable reporting infrastructure. Every week, department heads compiled their Progress, Plans and Problems manually. Updates were typed into spreadsheets, narrated in weekly meetings, discussed, then updated again after the retrospective. The cycle was repetitive, time-consuming and produced information that was already outdated by the time it reached the CEO.

There was no single source of truth. Data lived in inboxes, chat threads and individual files. Leadership was spending more time collecting information than acting on it.

Across 235 employees, the manual reporting cycle consumed approximately two hours per person per week. That is over 24,000 hours of organisational capacity lost annually to an administrative process that added no analytical value. Senior leaders were making decisions based on fragmented, manually compiled data with no consistent structure or timing.

The cost was not just operational. The absence of real time visibility meant that problems surfaced late, escalations were reactive and strategic alignment across departments was inconsistent.

The requirement came directly from the CEO and Chief of Staff. Weekly reporting was consuming the organisation and producing information that was already stale by the time it reached leadership. The question was not whether to fix it. It was how.

Existing systems were evaluated first. The organisation already operated on the Zoho ecosystem. Each application was assessed against the reporting requirement. None could fulfil it without significant customisation that would still fall short of what was needed. The decision to build a custom solution was made on that basis.

The design brief was defined with the CEO and CoS: automate the collection, synthesis and presentation of weekly Progress, Plans and Problems across all departments, and deliver a structured dashboard that leadership could access in real time. No unnecessary features. No external vendor. No expensive licensing.

The build was owned and delivered as a solo engagement. Architectural guidance was sought from three members of the Engineering and Product teams at the outset. The development itself was done independently. The prototype was built in Google AI Studio to validate the concept with the CEO. Once approved, the build was moved to Claude and completed. IT Infrastructure deployed the system to production.

The platform was built on a full-stack architecture. React and TypeScript on the frontend, Node.js on the backend, MongoDB Atlas as the database, and Google Gemini as the AI engine. Integrations were built across Gmail, Google Calendar, and the Zoho ecosystem. The system was deployed on Kubernetes with GitLab CI/CD pipelines and hosted on AWS CloudFront and S3. The architecture was deliberately kept lightweight so the platform could be built, deployed and operated without introducing a new enterprise software dependency.

The entire build took 30 days. Total development cost: under USD 20.