PowerUnits platform
European power market intelligence for serious analysts
Unit-level generation assets, historical market context (M-track), and forecast-readiness evidence (F-track)—with methodology-backed trust boundaries. Not live forecast scores or trading signals.
The public workspace is a bounded product surface—coverage and depth vary by country and layer. We state limits in-app; this page is the same contract in short form.
Who it is for
Built for teams that need defensible European power context
PowerUnits is for professionals who care about provenance and claim hygiene as much as charts. The same semantic split you see in the product—assets, M-track context, F-track readiness, methodology—starts here.
Market & power analysts
Orient on European generation assets, then layer historical market and driver context where the pipeline exposes it—without conflating that with forecast skill.
Trading & strategy desks
Use structured asset and market context as inputs to internal workflows—not as automated signals. F-track readiness is evidence and caveats, not live alpha.
Utilities & asset developers
See fleet structure and geography at unit level, alongside documented data semantics when you compare markets or plan studies.
Corporate & policy research
Ground narratives in methodology-aware boundaries: what the stack measures, what it does not, and how M-track differs from F-track.
Product surfaces
One workspace, four intentional surfaces
Each route answers a different analyst question. We keep M-track (what happened) and F-track (readiness and evidence on a fixed snapshot) explicitly separate so claims do not leak across tracks.
Assets
Generation fleetInteractive map and unit-level GEM-backed generation assets across European markets—spatial exploration and attributes, not price prediction.
Open in workspace →Market Analysis
M-trackHistorical, explanatory context from normalised hourly market layers and drivers where the join is populated—bounded snapshots, not forward scores.
Open in workspace →Forecast Readiness
F-trackVersioned readiness snapshot: ex-ante input semantics, offline compare evidence, country-level trust labels and caveats—not live forecast quality or trading signals.
Open in workspace →Methodology
Trust framingHow we separate raw from normalised data, hourly semantics, and the M-track vs F-track contract—so interpretation stays aligned with engineering reality.
Open in workspace →Trust & methodology
Honest boundaries are part of the product
We treat methodology as user-facing: what each layer can support, what it cannot, and how to read F-track readiness without mistaking it for production forecast performance.
M-track vs F-track
- M-track — Historical and explanatory: markets and drivers as observed through documented transforms.
- F-track — Forecast readiness and evidence on a bounded snapshot: inputs, gating, and conservative country guidance—not a leaderboard and not live model skill.
Where to read more
The in-app Methodology page summarises layer semantics for analysts; engineering ADRs remain the source of implementation truth. Start from the workspace when you need the same language the UI uses.
Open Methodology in workspace →Analyst outcomes
Outcomes we design for—not buzzwords
These are the practical reasons teams adopt a bounded intelligence workspace: speed with provenance, and fewer wrong conclusions from mixed track types.
Faster orientation on European generation
Move from country view to unit-level assets with consistent attributes—still the backbone of the product.
Ground market stories in documented layers
When M-track snapshots are available, anchor demand, price, and driver context to normalised hourly semantics—not ad-hoc spreadsheets.
Separate readiness from hype
Use Forecast Readiness for evidence-aware, country-level caveats on a fixed snapshot—explicitly not trading signals or live forecast quality.
Reduce cross-team misinterpretation
Shared vocabulary across desks: methodology-first framing reduces M/F confusion when slides leave the trading floor.
Differentiation
More than a plant list—still grounded in real units
We started from rigorous European generation asset data. The product has grown into an analyst workspace that keeps that foundation and adds market context and readiness views without pretending they are the same thing.
Asset-grade core
Unit-level records, map exploration, and export-minded structure—so spatial and fleet questions stay first-class.
Market context layer
M-track surfaces join hourly market and driver layers where data aligns—useful for past behaviour, not packaged as forecasts.
Readiness without overclaim
F-track readiness is evidence-aware and versioned, with explicit non-goals: no live quality score, no signal product, no AI theatre.
See the workspace on your own data questions
Request access for a walkthrough, or open the public analytics workspace to validate semantics and coverage before you commit.