Enerlytica

Enerlytica’s value sits in dense, high-value energy market information: report archives, PDFs, data tables and recurring data flows built up over years of market coverage.

That material was valuable but hard to work with. Knowledge was locked inside documents rather than held in a structure that could be queried, reused or built on. The opportunity was not simply to “add AI” on top; it was to create a more reliable information layer underneath.

Before any build started, we ran a discovery workshop with the report desk to understand why the existing process was fragile in the first place: undocumented data flows, no shared schema, manual handoffs between spreadsheets and PDFs, and no clear picture of what governance existed around who could access which sources. Diagnosing that root cause came first; the database and search layer were the answer to a specific, understood problem, not a guess.

Part of that early work was advisory rather than technical: reviewing Enerlytica’s data governance and the data-sharing agreements it holds with its own sources, so the structure we built reflected what Enerlytica was actually allowed to do with the data, not just what was technically possible.

From there we started with the foundation rather than the interface: document ingestion, table extraction from PDFs and reports, and a foundational database design that gave the material a schema it never had.

On top of that we added search capability that understands the content rather than matching keywords, API-connected workflows so the layer could be used by other systems, and improvements to the reliability of the recurring data processes the business already depended on.

The relationship kept going from there. In the years since, we have automated Enerlytica’s daily market reports end-to-end (generated from the database, published to a content system, emailed to subscribers and shown on a rebuilt website), replacing the manual spreadsheet and PDF workflow the report desk depended on before.

That relationship now also runs alongside ongoing advice on Enerlytica’s AI and data strategy: helping decide what to prioritise next as new data sources, tools and AI capabilities become available, not just maintaining what has already been built.

Five-plus years of reporting, held across many different formats, became one accessible and reusable layer. Historical reports, market commentary and tabular data are now a foundation to build on rather than an archive to dig through.

That foundation now also produces Enerlytica’s daily reports automatically, with a rebuilt website to present them on, work that has continued year over year since the first engagement in 2023.