Day-ahead price forecasts at 15-minute resolution, PV production forecasting, and a financial model your lender will accept — DSCR, IRR, NPV, built on P90, not on optimism.
Prices effective September 2026. Changes are announced 60 days ahead and do not affect contracts already running. Every product carries its state — available, pilot or in development: we do not sell a capability before it can be executed, verified and delivered within the promised time.
Prices are net, in euro. Payment is by proforma invoice and bank transfer; the invoice is issued in RON at the National Bank rate on the order date, plus VAT.
For the credit officer, adviser or lawyer who received a document from someone else.
For the owner of a plant in operation who already has a forecaster, a BRP and a trader. We replace none of them.
For the developer who needs a price assumption the lender can verify.
For the bank or adviser whose data does not leave the perimeter.
For the adviser who keeps their own financial model and already has the level. We do not redistribute market data — realised prices belong to ENTSO-E, FX rates to the National Bank, and storage spreads we already publish free, in full.
What nobody has and we do produce is the shape: captured price and value factor by technology and zone, the cannibalisation profile over the horizon. We do not transfer the data, we transfer the result computed from it — the inputs are public anyway, and we attribute them. The level is bought from Aurora or AleaSoft; the shape is bought from nowhere.
Not sold until published coverage reaches target on the series being sold. That is a calibration matter, not a calendar one. And a client's settlement data never enters this — it is theirs, with its own threshold, at 15 plants.
Forecasting and nomination support, per MW installed. It runs in parallel and with no effect: the model writes what it would have proposed, to a journal — it submits nothing, zero operational risk. We publish no price for something we have not demonstrated.
What we know so far, measured: over 16 months of backtest the effect came out between −8,209 and +30,169 RON — an interval that crosses zero. The rule we follow is that the decision threshold is written before the run, not after the result; that threshold is not written yet, and we say so rather than imply it exists.
And there is a gate ahead of the measurement that the measurement cannot jump: the lever that produced the largest backtest gain needs the plant's output data close to current. In operation it arrives with the monthly settlement note — 46 days old on average. Until a plant has a near-real-time output source, the backtest gain is not realisable, and a measurement run now would assess a system with its main lever switched off.
The subscription buys availability, not a number of deliveries. The fee covers verification of up to 12 settlement months in a year, one per month. A month you do not send us a note is not owed back and does not roll over — it is an unused right, not an outstanding delivery. The settlement note is issued per account holder and covers every plant, so one Report covers them all.
The registry is priced per account holder, not per plant and not per MW. The settlement note is issued to the account holder named in the BRP contract and covers all their plants in a single document — so the price follows the unit the document arrives in, not the plant count. A power threshold would exclude the numerical majority of Romania's plants.
We do not index the price to your imbalance cost. A subscription computed as a percentage of a problem we are trying to shrink rises when the customer does badly and falls when the product works. The measured cost stays the justification, never the formula.
The price rests on our published error. Every month we publish how often our band covers the realised price, per zone and per horizon, including the months we miss the target. See the calibration.
One-off deliverables, with a methodology note. They are not tiers of the same product.
We anchor the physical model to your energy report (PVsyst, RatedPower), keeping the hourly shape of production, and issue the methodology note. What changes is the annual level; what stays is the profile. It is not a certification and we do not call it one.
Request calibration →At your offices or online: how to read a P10–P90 band, what published coverage means, DSCR across scenarios and sensitivities. For investment, development or credit teams.
Book a session →Because its value comes from the number of assets tracked consistently, not from the size of each. A threshold would exclude the numerical majority of Romania's plants — precisely the cases where imbalance cost matters most relative to revenue.
If you already have a formal energy yield report (PVsyst, RatedPower), Voltaraj adjusts the annual yield to match the exact figure in that report, while preserving the hourly physical shape of production. The result: a financial model built on the number your lender already recognises.
The platform runs on your servers. The financial engine receives only configuration objects and revenue figures, and returns DSCR / IRR / NPV — it never touches or sees the raw market data warehouse. Your data never leaves the institution's perimeter.
Because we are not selling them yet. A published price is a promise that we can execute, verify and deliver within the stated time. For Operations, the measurement that would justify a price is not finished; until it is, it runs in parallel with no promise attached. We would rather leave a gap on the page than print a figure we cannot defend.
Coupled European spot markets — Romania, Bulgaria and Hungary live. Prices come entirely from the ENTSO-E Transparency Platform (as reported by the market operator) — a source with commercial re-use rights — in each market's own currency, converted to RON at the official central bank rate for Romania. No scraped data, no negotiated licensing dependencies. Financial modelling currently runs on the Romanian price series.
Payment is by bank transfer, against a proforma. A fiscal invoice is issued for every payment.
Start with the published calibration, for the zone you care about. If what you see there is useful to you, let's talk.
See the published calibrationOr see the published comparison first — how much the P10–P90 band actually covered against the realised price, day by day, across the whole record.