How Wattklok works.

01 · Your data & privacy

Your data stays with you.

Wattklok is free and needs no account with us. Reading your file, building a generated profile and every simulation happen in your browser. Your history, settings and results are saved in this browser’s storage (IndexedDB and localStorage), not on our servers. They stay there until you clear them and don’t sync to other devices.

Some steps need the internet. Tibber requires sign-in and data requests to go through a server, so ours passes your homes, history, production and prices on to your browser without keeping a copy; sign-in tokens are kept in HTTP-only cookies. Two small preference cookies remember your language and whether you use a file or generated data, so pages open in the right shape; neither holds any energy data. Spot-price lookups also go through our server, with only your price area and dates. Weather and location search go directly from your browser to Open-Meteo and OpenStreetMap, which receive your location and the dates requested. Comparing weather years fetches Open-Meteo weather since 2015 for your location and spot prices per year for your price area in the same way. We count anonymous page views with Vercel Analytics, without cookies; your energy data is never included.

You stay in control. Change data replaces your active source. Start fresh removes local data and results. Clear cached Tibber data downloads your history again. Logging out also removes saved file and generated data, as the confirmation explains. Because everything is kept in the browser, use a browser profile you trust. Local processing doesn’t mean the whole app works offline.

02 · Data sources

Start with your home.

Tibber provides measured hourly grid import (consumption), solar export to the grid if you have panels, your home’s location and hourly electricity prices. Tibber data shows that measured history with current and published prices. Calculator and Explore data show simulation results, not live meter readings.

CSV and Excel imports use consumption history exported from your grid operator or supplier; supported formats are listed during import. You choose a location and either historical Nordic spot prices or a fixed price, which replaces any prices in the file. A file that also reports export to the grid, or export and panel generation, marks your home as having solar and can be used as measured solar, like Tibber's export; generation without export can't be separated from your use and isn't used. Negative consumption is rejected; zero consumption and negative prices are accepted.

Generated profiles estimate hourly use from a household type, annual consumption, and heating and EV choices. With electric heating or a heat pump, heating follows the real daily temperatures at your location, so cold years use more than mild ones, while the average year matches the annual consumption you entered. They are model inputs, not measured history. The public example uses a separate Stockholm household with the same calculation engine, and leaves your own home, settings and results untouched.

How much history is used is up to you. With Tibber you choose 1–5 years, or all available (up to 8 years); 3 years is the default. An imported file is used in full, and generated profiles cover 1–5 years. More years average over more weather and price conditions, but older years can carry unusual prices or describe your household before changes such as a heat pump or an EV.

03 · Solar & battery model

One hour at a time.

Solar generation combines historical Open-Meteo sunlight (irradiance) for your location with panel capacity, orientation and tilt, matched hour by hour to your consumption. An 80% performance ratio covers losses such as wiring, dirt and the inverter. Panels also lose output as they warm up: each hour's cell temperature is estimated from the air temperature and the sunlight, and every degree above 25 °C costs about 0.4%, while cold, bright hours gain a little. Panel efficiency is already reflected in the rated kWp. For an existing system you can use Tibber's measured export instead. That is what your meter sent to the grid, not what the panels generated: solar used at home never reaches the meter. The battery is then modelled on measured export and import, and no generation or self-consumption figure is shown. Hours without an export reading count as no export, and the results say how many there are.

Wattklok simulates solar only, solar with a battery, and battery-only arbitrage. A battery has a capacity, charge and discharge limits, a round-trip efficiency (90% by default, for the whole system rather than the 95–97% quoted for the battery alone) and a minimum reserve, and it starts empty. It charges from surplus solar, or from the grid in the day’s cheapest hours when the energy is worth more later, after fees and charging losses. It discharges to cover household use, in the heaviest-use hours, for export in expensive hours, or in a mix: household use first, then selling what’s left in expensive hours when that earns more than buying it back would cost.

The battery runs by one of six strategies, each a combination of how it charges and what it discharges for. Store own solar and Solar for peak hours charge only from surplus solar, so they are offered only when you simulate solar. Buy cheap, use at home; Buy cheap, cover peak use; Use at home, sell surplus; and Trade with the grid also buy from the grid in the day’s cheapest hours, but only on days when the price spread covers charging losses and fees. Which one is best depends on your prices, export compensation and usage, so the results compare them all.

These are look-backs, not forecasts: they show what a system would have done with your past prices, weather and usage. Future prices, weather and usage will differ. The model also leaves out some site conditions, tariff details such as power-based grid fees, maintenance and equipment replacement.

04 · Yearly results

Read the result in context.

Every hour of your history is simulated in order, with its own weather, price and usage. The results are then split into 12-month periods counted back from your latest data, and the headline is the average of the full years: annual savings, costs with and without the system, energy and battery cycles. A period counts as a full year when at least 95% of its hours are recorded; yearly figures are scaled to 8,760 hours, so a gap counts as an average hour of that year. The range shows the lowest and highest year, because savings can move a lot with prices and weather. Simple payback divides the investment by average annual savings; when savings are zero or negative there is no payback, and a zero investment is shown separately. Costs, savings and export revenue can be negative.

When full years differ a lot (the best saves at least 1.5 times as much as the worst, or a year saves nothing), the calculator says so. Every full year counts equally in the average, so a year with unusual prices, such as 2022, or an older household can pull it up or down.

Months outside the full years, such as the first three months of a 15-month history, are left out of the averages but still shown in the explorer. With less than a year of data, yearly figures are scaled up from the hours you have and monthly charts show only the months you have; seasons missing from the data aren’t represented. Monthly charts in the calculator show the average for each month of the year.

Solar self-consumption is the share of solar generation used in your home, either directly or later from the battery. Battery output is split between stored solar and stored grid energy in proportion to what went in. Exports, conversion losses and energy left in the battery don’t count. This is different from the share of your consumption covered by solar.

05 · Heating & weather years

How the weather shapes the result.

From your own history, day by day, Wattklok works out how your use rises as it gets colder: a base level (weekdays and weekends), the outdoor temperature where heating starts, and the extra kWh a day for each degree below it. The temperature is smoothed over a few days, because a house stores heat. From that come the share of your electricity that goes to heating and, for each full year, how much colder or milder it was than a normal year at your location (the average weather since 2015, from Open-Meteo), with what you would have used in a normal year. It is shown only when the pattern is clear: temperature must explain at least 60% of the day-to-day variation over at least 120 days. With existing solar, only October to March is used, when grid import is close to what the home uses, so the heating share is an estimate.

What if the weather had been different? After your results are ready, your latest year is replayed with the actual weather and electricity prices of every full year since 2015, and simulated with the same engine. Each day keeps its own use and gets the heating the same date needed in that year; solar comes from that year's sunshine. Generated profiles are generated again with each year's temperatures. Weather and prices always come from the same year, so a cold, still winter comes with its own price spikes. Years without complete spot prices for your area are left out rather than estimated. Your habits stay like your latest year, and your measured years remain the main result.

Spot prices are turned into what you pay from how your own prices followed the spot price over your latest year (supplier markup, tax and VAT); grid fees and export compensation apply as in your normal results. By default each year's prices are scaled to your latest year's average level, keeping which hours were cheap or expensive; you can also view them as they were that year, when expensive years like 2022 show much larger savings. With a fixed price every year uses that price and only the weather varies. With existing solar the use is modelled from winter days, so its base level is an estimate. The results show the lowest, typical and highest year and explain the coldest and mildest one.

06 · Prices, fees & export

What a kWh is worth.

Every kWh you buy costs more than the electricity itself: your grid operator’s transfer fee and energy tax are added per kWh, and every kWh you produce and use yourself avoids them. Wattklok adds them to each hour’s import price (typical 2026 values for your country, which you can change to match your bill). They are not added to export compensation, which is automatic (90% of the hourly price), fixed, a percentage, or the hourly price plus an adjustment. Manual prices include VAT; regional VAT is added to spot prices. Check Your system before comparing results.

07 · Is it worth it?

Over the equipment’s life.

The average yearly savings are replayed over the equipment’s life, starting from what it costs: panels last 30 years and lose 0.4% of their output a year, as current warranties assume (no inverter replacement is included), a battery lasts 12 years with no replacement, and later years aren’t discounted unless you add a return the money could earn elsewhere. With a loan (Financing in Your system), only the part you don’t borrow is paid on the day you buy; the monthly payments of an annuity loan, less the interest deduction for your country, then come off the savings until the loan is repaid, so the interest counts against the result. Prices and usage stay at your past levels, so this is a look-back rather than a forecast. The band shows the result if every year were like your worst or best one. The verdict is “likely worth it” when the system ends at least 20% of its cost ahead and still comes out ahead in years like your worst; “unlikely” when it falls more than 10% short of earning back its cost; and “borderline” in between. With solar and a battery, the battery is judged on what it adds on top of the same panels. All of these assumptions can be changed.

If you already have solar, the baseline keeps that solar without a battery. The investment is the new battery plus any solar upgrade cost you enter. Money already spent on panels isn’t counted, and an upgrade cost doesn’t add generation by itself.

08 · Strategies & sizes

What else could work.

How should the battery run? After your calculation, your system is also run with every other battery strategy that fits it, on the same history. The comparison shows savings per year, the difference from your strategy, and the net result and payback over the equipment’s life, worked out the same way as in Is it worth it?. Use and recalculate switches to a strategy and reruns everything with it.

Which size makes sense? After your calculation, the same history is rerun with half to twice your solar size and half to one and a half times your battery (5 kWh or more), plus no battery and, when you simulate a battery, no solar. Your own quote is used as entered; other sizes keep the fixed part of the price for your country (see the table below, at most 40% of your quote) and scale the rest by size. Your subsidy rules are then applied to each size, so a battery rule that requires solar doesn’t apply to the options without solar. Real quotes vary, so check the best option with an installer. The size table uses your strategy and can switch to another one; that table is calculated when you pick it and saved with your results.

09 · Costs & subsidies

Starting points for costs and subsidies.

Subsidies and deductions are applied openly. Installation costs are entered before them; subsidies are a list of rules in Your system, and you can add as many as apply (a national deduction and a local grant, say). Each rule has a part for solar and one for the battery: a percentage of the cost, an amount per kWp or a fixed amount, each with an optional upper limit, plus an optional limit for both parts together. A battery rule can require solar, as Sweden’s deduction does. Every rule works on the cost before subsidies, and together they never pay more than a part costs. Defaults follow your country (see the table below) and can be changed or removed. The calculator lists the cost before subsidies, each rule and what you pay; payback, the verdict, “Is it worth it?” and the size finder all use what you pay.

Regional default costs are a fixed part (inverter, scaffolding, electrician, paperwork) plus a part per kWp or kWh, fitted to typical 2026 prices including VAT; the table shows them with each country’s default subsidies and a worked example. These are rough model assumptions, not installer quotes or a statement of eligibility: replace them with your own quote and rules in Your system. Currency changes do not convert historical prices.

• Sweden: Green technology deduction (grön teknik), taken off the invoice: 15% for solar, 50% for a battery connected to solar or installed with it (a battery on its own gets none). On a fixed-price turnkey job Skatteverket counts 97% of the price, so in practice it is 14.55% and 48.5%. At most 50,000 SEK per person and year for both together; if two of you own the home and share the invoice, the limit is 100,000 SEK.
• Norway: Enova grant for solar: 2,500 NOK per kWp, up to 37,500 NOK (15 kWp) and at most 25% of the cost. Applications are open with no deadline (Enova’s terms of January 2026). No grant for the battery itself; Enova’s grant for smart control covers the steering system, not storage.
• Finland: No national grant for solar or batteries (the energy grant ended after 2023). The household deduction (kotitalousvähennys) covers part of the installation labour: 35% up to €1,600 per person in 2025, proposed 40% up to €2,100 for 2026. Add it as a fixed amount if it applies to you.
• Denmark: No national grant for solar or batteries. The green part of the craftsman deduction (håndværkerfradrag) covers installation labour for solar, and for a battery installed with solar (not a battery on its own): up to 9,000 DKK per person in 2026, worth about a quarter of that in tax. Add it as a fixed amount if it applies to you.

10 · Dates, gaps & saved results

Dates, gaps and saved results.

Dates stay real throughout: results keep your home’s timezone, so daylight-saving days have 23 or 25 hours, and leap years keep their extra day. A 12-month period that contains 29 February expects 8,784 hours.

Missing hours are gaps, not zero consumption. Results flag missing prices (counted as zero), estimated spot prices and other coverage warnings. Gaps in weather data can lower modelled generation. If historical weather can’t be loaded for the public example, it uses a clearly labelled illustrative sunlight profile; your own simulations tell you when they recover or fall back.

Calculate saves your settings and runs the first simulation. After that, Edit system opens a draft: Save keeps your changes, Run saves them and recalculates, and Cancel discards them. Recalculate reruns with your saved settings. Saved results are marked as outdated when inputs change, and hidden when you switch to a different home or data source. A reload restores completed results without rerunning them; results from an older calculation engine need recalculating.

Every chart has a text or table alternative, and tables can be downloaded as CSV. Downloads match the level, period and columns on screen (only the parts of your system that were simulated); explorer downloads also include units, currency, settings and timezone.