Why The Timeseries Refinery

How we compare

These are the tools we hear most when talking with energy, trading and industrial teams. The Timeseries Refinery is a complete platform — it replaces what other teams assemble from multiple tools.

		Data platforms		Spreadsheet	Databases		Visualisation
	The Timeseries Refinery	Databricks	Dataiku	Excel	InfluxDB	TimescaleDB	Grafana visualisation only	Power BI visualisation only
Built for energy & trading	Designed specifically for energy, trading and industrial time series teams	General-purpose data lakehouse — no domain specificity	General-purpose data science platform — no domain specificity	General-purpose spreadsheet — adapted by users for energy work	Built for IoT monitoring and DevOps — not energy or trading	General-purpose time series database — no domain specificity	General-purpose dashboarding — no energy or trading focus	General-purpose BI tool — no energy or trading focus
Traceable formula engine	Every calculation is versioned and auditable — from dashboard to primary source	Delta Lake provides data versioning and Unity Catalog tracks data lineage — but there is no formula-level traceability for time series calculations	Visual pipelines and recipes — not a time series formula engine	Formulas exist but are not versioned, auditable or reproducible across teams	Flux queries — no formula traceability	SQL only — no formula engine	—	—
Versioned series storage	Every stored series is fully versioned — any past version is accessible and comparable	Delta Lake provides table-level time travel — but not per-series versioning with a named catalog	No native versioning of time series — depends on the underlying storage layer	No versioning — manual file copies are the only way to keep history	Append-only storage — no version history per series	Append-only by design — no built-in version history per series	—	—
Usable by non-coders	Analysts can build formulas, manage series and publish dashboards entirely through the UI — no coding required	Designed for data engineers and data scientists — not accessible to non-technical analysts without support	Visual interfaces reduce the need to code — but initial setup and data preparation require technical expertise	Fully accessible to non-coders — but limited to what one person can manage in a spreadsheet	Flux or InfluxQL required for queries — not accessible to non-coders	SQL required for all operations — not accessible to non-coders	Dashboard creation is accessible — but data queries require InfluxQL, Flux or another query language	Accessible to non-coders for dashboarding — but underlying data preparation and modelling require technical skills
Operational dashboards	Real-time dashboards with direct access to the versioned time series catalog	Basic dashboards — not designed for real-time operational time series monitoring	Basic dashboards — not built for real-time operational use	Static charts — no real-time updates or shared operational views	—	—	Strong real-time dashboards — one of its core strengths, but no link to a versioned catalog or formula engine	Strong dashboarding — but no native time series catalog or formula traceability
API & data access	Full Python library and REST API, no SQL — every operation accessible via code or formula language	Strong Python and REST APIs — but SQL required to model and expose data to analysts	Python and REST APIs — visual recipes reduce SQL, but complex preparation still requires technical expertise	No native API — no server-side computation or shared catalog	HTTP API and client libraries — Flux or InfluxQL required for all queries	Via SQLAlchemy or psycopg2 — SQL required for every interaction	REST and provisioning APIs available — primarily read-oriented	REST API available, primarily read-oriented — DAX for report-level calculations, data preparation requires a separate platform
Ready to deploy	Docker-based — operational in hours, no data engineering team required	Available as a managed cloud service — but requires data engineering expertise to model and expose time series data to analysts	Available as SaaS or on-premise — but significant configuration needed before analysts can work with time series data	Immediate — but does not scale beyond a team or a few hundred series	Self-hostable or InfluxDB Cloud — but requires a separate dashboarding tool	Database only — additional tools needed for dashboarding, computation and access management	Self-hostable or Grafana Cloud — but requires a separate data source and storage layer	SaaS — fast to deploy as a visualisation layer, but requires a separate data platform for storage and computation