Margün Enerji
Margün Enerji's document archive became searchable at the content level
The Situation
Margün Enerji is a publicly listed energy company engaged in solar power plant investment, EPC and operations & maintenance. Those activities generate a continuously growing document archive — project files, field reports, contracts, technical drawings and official correspondence.
A significant share of that archive consisted of scanned documents. They could not be searched beyond their file names and folder structure, so reaching a contract clause or a field record depended on already knowing which folder held it. As the archive grew, this became an operational bottleneck.
Margün Enerji needed an internal archive system that did more than store documents — one that made their contents searchable and let teams reach the right document quickly while respecting access boundaries.
How We Worked
Document and Process Analysis
We examined the existing archive structure, document types and each department's access requirements.
Search behaviour is shaped by document types and by who needs access to what.
OCR Processing Pipeline
We built an automated pipeline that converts scanned documents to text, classifies them and prepares them for indexing.
Making image-format documents machine-readable was the precondition for searchability.
Search Infrastructure
We built a full-text index on Elasticsearch and added AI-assisted query interpretation for searching within document contents.
The goal was for users to type what they are looking for, not the name of a file.
Access Control and Interface
We developed role-based access control and a search interface with document preview.
Contracts and official correspondence required access boundaries by department.
Migration and Rollout
We migrated the existing archive into the system and completed user training and operational handover.
The system's value depended on the historical archive being searchable, not just new documents.
What We Delivered
Electronic Archive System
Internal platform where documents are stored and managed centrally
OCR Processing Pipeline
Automated process converting scanned documents to text and preparing them for indexing
In-Document Search
Full-text search with AI-assisted query interpretation across document contents
Role-Based Access Control
Document access managed by department and permission level
Archive Migration
Migration and classification of the existing document archive
The Outcome
Project Details
Similar Projects
Naturel Enerji's plant portfolio, sustainability and investor content brought onto one platform
Ekin Enerji's service areas, project archive and certifications brought onto a bilingual corporate site
Have a similar project in Energy?
Let's discuss how we can apply this approach to your business.