Battery revenue study, from your site's own data
Upload a year of hourly load and PV data. We simulate a battery hour by hour against real market prices and network tariffs, and return the revenue and sizing figures your financial model and your lender will ask for.
For developers, EPCs, ESCOs and owners deciding whether a battery pays, how big it should be, and what it should do once it runs.
- Any data formatCSV, SCADA exports, meter files, bills
- Five battery sizesvalue per kWh, cycles, revenue by stream
- Quote in one working daystudy in two, reviewed in five
A battery revenue study simulates how a battery would have been dispatched, hour by hour, against a site's real load and generation and the real market prices and tariffs of a given year, to estimate annual revenue by stream and the value of each battery size. It replaces the spread assumption and the cycles-a-day rule of thumb with the site's own numbers.
Why the rule of thumb fails
The number everyone needs and nobody wants to guess
Every battery project starts with the same question: what will it earn? The answer decides the size, the financing and whether the project happens at all. Most teams answer it with a spreadsheet, a spread assumption and a rule of thumb about cycles a day.
That worked when prices peaked at midday. They don't anymore. In the Iberian market the cheapest hour of 2025 was 13:00 at about €30/MWh and the most expensive was 20:00 at about €110, with around 200 hours of negative prices in the year. Solar surplus now sells into the trough. A battery that charges only from solar at a site with steady load can sit largely empty. The value moved, and the rules of thumb didn't move with it.
A revenue estimate you can put in front of a lender has to be built from the site's actual load, the actual PV profile, the actual price curve and the actual rules on metering, grid charging, export caps and tariff periods. And it has to show its assumptions.
Get the number for your siteSolar surplus sells into the trough. The value moved to the evening. A study built on the actual curve tells you what your battery earns from that move.
Get the number for your site.
One site, one study. Send the details and we reply within one working day with a quote and an upload link.
Spreadsheet, forecast platform, or a study from your own data
Three ways to get the number
Each has a job. A market forecast platform tells you where prices may go. A spreadsheet tells you what your team assumed. A study from your site's data tells you what this battery, at this site, would have earned.
| Spreadsheet rule of thumb | Market forecast platform | DynVolt site-data study | |
|---|---|---|---|
| Starts from your metered load and PV | Sometimes, if someone types it in | No — a generic asset in a market | Yes — the study is built from it |
| Hour-by-hour dispatch against real prices and tariffs | No — a spread and a cycles-a-day rule | Yes, against a market forecast | Yes, against the real year, haircut for forecast error |
| Battery sizes compared | One | One per run | Five, your proposal marked |
| Assumptions register with sources | No | Rarely | Table and JSON, every flagged default |
| Method card a lender's adviser can read | No | Methodology page | Ships with every study; signed in the Reviewed tier |
| Forward price view to 2040 | No | Yes — this is what they sell | No — pair the study with a forecast provider |
| Time to a number | Days of your own team | Minutes to hours | Quote in one working day; study in two; reviewed in five |
| Cost known before you start | Your team's time | Subscription | Fixed quote per site, credited against an operating contract |
From upload to study in three steps
How a study is produced
- 1
Upload what you have
CSV, XLSX, SCADA exports, inverter portal downloads, meter files, PDF bills. Any format, any structure, several files. We normalise it to hourly, find the gaps and tell you what we found before we model anything.
- 2
Answer six questions
Own meter or behind the site meter. Grid charging permitted. Import and export caps. Indexed or fixed-period tariff. Market route. Revenue streams in scope. Every answer goes into an assumptions register with its source. Skip any of them and we apply a default and flag it.
- 3
Receive the study
Hour-by-hour dispatch optimisation against real day-ahead prices and published network tariffs, across five battery sizes, with every flagged assumption toggled. Revenue by stream, cycles, value per kWh installed, what is still open, and what we need to make the figures firm.
Eight deliverables, one file your lender can open
What's in the study
Findings, not a dashboard
Three findings about your site, each with the evidence and the consequence for the business case.
Size sweep
Annual value, value per kWh installed and cycles per year for five configurations, your proposal marked.
Revenue by stream
Self-consumption shifting, arbitrage, peak shaving, and where metering and market route allow, aFRR, mFRR and FCR, each shown separately with its activation assumption.
What is still open
Every assumption that moves the answer, with the euro swing and who needs to confirm it.
Assumptions register
Table and JSON: value, unit, source, confidence, confirmation status. Drops into your own financial model.
Results workbook
Hourly dispatch, monthly summary, size sweep and a finance sheet with live formulas your team can audit.
Method card
Engine version, price and tariff data versions, model description and limitations. The document a technical adviser asks for.
Follow-up and re-runs
Ask "what if grid charging is not permitted" or "run 1.5 MW / 3 MWh". The result is appended to your study.
Numbers that drop into your model
The assumptions register ships as a table and as JSON. The results workbook keeps live formulas. Your team audits it; nobody retypes it.
Request a quoteBuilt for a financing file
Method you can hand to a lender
Deterministic engine
Revenue comes from a linear-programme dispatch optimisation at hourly or 15-minute resolution. No black box, no averages of averages.
Real prices, real tariffs
Full-year day-ahead prices at native resolution, regulator-published network access tariffs by period, ancillary services results where in scope.
Realistic, not perfect foresight
The base case is haircut using the measured day-ahead error of our own production forecasting system, not perfect hindsight.
Versioned and reproducible
Every study is stamped with engine, price-data and tariff-data versions. Re-run it in a year and get the same numbers.
Independently checked
Every figure in the report is matched to a source cell before it is published, and the narrative is read the way a lender's adviser would read it.
Operated, not just modelled
The same optimiser dispatches assets live on the DynVolt platform. We see what predictions do when they meet a real meter.
The optimiser behind the study is the one that runs BESS optimization on the DynVolt platform: warranty-bounded dispatch across day-ahead, intraday and ancillary markets, live on real batteries.
Read the method before you buy the study
The method card, in your inbox
Three pages: inputs and their sources, the dispatch method, the forecast-error haircut, what comes out, the checks before release, the limitations stated plainly, and the assumptions-register schema. It is the document that ships with every study, blank. Hand it to your lender's technical adviser first and see if they have questions we have not already answered.
Pick the depth you need
Ways to work
One site, one study. Portfolio arrangements for developers running several sites a quarter.
Study fees are credited in full against a DynVolt operating contract signed within twelve months. Quotes are per site and sent within one working day of your request.
Request a quote
Start with one site
Tell us about the site and the kind of study you need. We reply within one working day with a quote and an upload link. Work starts when the quote is confirmed and the data is in. Data is used only for your study and deleted on request.
Coverage: any market with a public day-ahead price and a published network tariff. The engine is market-agnostic; loading a new market's rulebook takes a few days.
Fourteen questions, answered plainly
Questions
Is the study bankable?
The study is built to be used in a financing file: a deterministic dispatch method, real market and tariff data, an assumptions register with sources, and a method card. Whether a lender accepts it depends on the lender and on which assumptions are confirmed. The Reviewed tier includes a signed method card and a walkthrough with your lender's technical adviser.
What data do I need?
Ideally a year of hourly or 15-minute site consumption and PV production. Any format works. If PV is planned rather than measured, we synthesise a profile from satellite irradiance for the site. If you only have monthly bills, we can still screen the site, with the confidence stated.
How do you get market prices?
We maintain full-year day-ahead prices at native resolution for each covered market, plus ancillary services results and regulator-published network tariffs. Each study records which dataset versions were used.
Do you model ancillary services?
Yes, where the battery's metering and market route make them accessible. Each stream is shown separately with its activation assumption so you can include or exclude it in your own model. We do not fold speculative ancillary revenue into a single headline number.
What about degradation and warranty?
Degradation is modelled as a cost per MWh of throughput, and cycles per year are reported for every configuration so you can check them against the warranty envelope of the system you are considering.
How long does it take?
Screening the same day the data is in. A full study in two working days. A reviewed study in five.
Who is DynVolt?
DynVolt builds and operates software for solar and battery assets: SCADA, production forecasting, market operations and battery dispatch. The study engine is the same optimiser that dispatches assets on the platform.
Is this a BESS feasibility study?
It is the revenue and sizing part of one. A feasibility study also covers grid connection, permitting, technology selection and capex; this study gives that file its hour-by-hour revenue model, size sweep and assumptions register, and states what it does not cover.
Is this a BESS revenue forecast for future years?
No. The study simulates one full historical year of real prices, tariffs and site data, so every number is traceable to something that happened. For a forward price view to 2035 or 2040, pair the study with a forecast provider; the assumptions register is built to accept their curves.
Can I use the results in my own battery storage financial model?
Yes. The assumptions register ships as a table and as JSON, and the results workbook keeps live formulas for hourly dispatch, monthly summary, size sweep and the finance sheet. Nothing needs to be retyped.
Which markets do you cover?
Any market with a public day-ahead price and a published network tariff. The engine is market-agnostic; loading a new market's price data and tariff rulebook takes a few days.
How is the study priced?
Per site, with a fixed quote sent within one working day of your request, before any work starts. Study fees are credited in full against a DynVolt operating contract signed within twelve months.
Is my site data confidential?
Yes. Data is used only for your study, is not pooled with other customers' data, and is deleted on request. The Reviewed tier can be delivered under your NDA.
Do you also operate the battery afterwards?
We can. The same optimiser that produced the study dispatches batteries live on the DynVolt platform, which is why study fees are credited against an operating contract. The study stands on its own if you choose another operator.
Still deciding which study fits?
Tell us what the number is for and we will suggest the depth. Quotes are per site and carry no obligation.
Studies are indicative and prepared for discussion. They are not investment advice.