An explicit price, mileage or tariff supplied by the user takes priority over a default.
CzymPojade methodology
We do not guess.
We show how we calculate.
The advertised price is only the beginning. RealTCO combines depreciation, energy, maintenance, insurance, tyres, tax and financing into one result for a specific driver and car.
BEVHEVPHEVpetroldieselLPG
01Reference scenario
One comparison baseline. Then your circumstances.
When the user does not change the inputs, every car is compared over the same time and driving profile. This keeps the calculation repeatable and the comparison fair.
5 yearsanalysis horizon
15,000 kmannual mileage
40 / 35 / 25%city / road / motorway
75,000 kmtotal mileage
This is a starting point, not a verdict. Mileage, roads, speed, driving style, home charging, tariff, solar, financing and current-car data override the reference scenario.
02Four rules
Comparability first, personalisation second.
The methodology separates observed inputs from forecasts. Every value has a defined place in the source priority.
Variant data and model-specific schedules take priority over segment averages.
A current market feed takes priority over the last valid value and the documented fallback.
Brand, powertrain and commercial relationships cannot change cost categories or buy a ranking position.
03The RealTCO formula
We count the money that actually leaves the budget.
The headline result is the nominal economic cost over the selected period. We subtract the resale value and benefits available in the chosen scenario from total spending.
purchase + financing cost
energy + service + repairs + insurance + tyres + fees
residual value
tax and energy benefits
net TCO
Items such as inspections, parking zones, low-emission zones, parking, cleaning, fluids, assistance, downtime or carbon cost enter the formula only when they apply to the selected profile.
04Seven calculation layers
From purchase price to the risk of future bills.
Each layer is calculated separately and then reconciled in one ledger. The result can be shown by month, kilometre, year and cost category.
Value and depreciation
Final value is calculated on an age x mileage surface, not with one annual percentage.
- interpolation for age and total mileage
- brand, powertrain, battery and used-car adjustments
- a separate residual-value uncertainty band
Fuel and energy
Consumption responds to road mix, season, temperature, motorway speed and driving style.
- petrol, diesel, LPG, HEV, PHEV and BEV
- charging losses and home, work and public charging shares
- home tariffs, dynamic pricing, solar and battery storage
Scheduled maintenance
A cost appears when a time or mileage interval is crossed, not as a flat rate per kilometre.
- minor and major services by brand tier and powertrain
- model schedules and wear components
- dealer, specialist, independent or DIY service channel
Repairs and ageing
Unplanned repairs are an expected value based on age, mileage and warranty coverage.
- calibration using TÜV and ADAC statistics
- separate major-failure and traction-battery risks
- manufacturer warranties and current-car risk control
Insurance
Liability and comprehensive cover change with vehicle value and supported driver and market factors.
- brand, segment, engine size, driver age and region
- comprehensive cover based on declining vehicle value
- a market estimate, not an insurer quote
Tyres, tax and fees
Tyres are separate, while tax and fees follow the country and type of use.
- segment, rim, season, mileage and driving style
- inspections, parking zones, vehicle tax and registration
- VAT, deductibility, income tax, leasing, private or business use
Uncertainty and explanation
The deterministic result is complemented by 500 simulations of future prices and events.
- fuel, electricity, mileage, maintenance and insurance volatility
- powertrain-specific residual-value risk
- base result plus P10, P50 and P90 scenarios
05Uncertainty instead of false precision
One number cannot describe the future.
For the analysed car, the Monte Carlo layer runs 500 reproducible simulations. Fuel or electricity prices, mileage, maintenance, insurance and final value vary in a controlled way. We show a distribution, not a promise.
P10lower-cost scenario
P50median simulated outcome
P90higher-cost scenario
The P10-P90 range contains the middle 80% of model results under the stated assumptions. It is not a guarantee of resale price, repair bill or insurance premium.
06Sources and freshness
Observed data, calibrations and fallbacks play different roles.
We do not call every number live data. Technical records, price feeds, used-market calibration and reference values are stored separately.
| Area | Source | Cadence | Use |
|---|---|---|---|
| Fuel | e-petrol.pl | daily | petrol, diesel and LPG; last valid record and then a documented fallback |
| Home electricity | URE + PSE day-ahead | tariffs / daily | retail tariffs and the hourly-price anchor for dynamic charging |
| Public charging | operator price lists | daily / reference | GreenWay feed; maintained reference rates for the other networks |
| Reliability | TÜV Report + ADAC | annual | fault probabilities, ageing and repair-risk calibration |
| Residual value | vehicle listings + industry forecasts | periodic | age x mileage surfaces, brand adjustments and forecast provenance |
| Solar | JRC PVGIS | per location | hourly solar profile with caching and a fallback model |
| Tax and fees | legislation + country configuration | after legal changes | Poland, Germany and the UK; not a substitute for tax advice |
| Vehicles | CzymPojade database + manufacturer data | continuous | prices, variants, powertrains, batteries, consumption and model schedules |
Resolution order: user input -> specific variant -> current feed -> last valid record -> documented reference value. A missing feed must never silently reduce a cost to zero.
07Model boundaries
What the result means, and what it does not promise.
TCO is a decision tool. It makes comparisons consistent, but it cannot know the future or the history of one particular vehicle better than its owner.
Estimate, not quote
Premiums, finance payments, repair bills and resale prices can differ from modelled values.
Inputs matter
Incorrect mileage, tariff or consumption produces a precisely calculated but unsuitable scenario.
Fallbacks are explicit
If a live source fails, the engine continues with the last valid data or a reference value.
ML does not change TCO
Machine learning helps classify and recommend cars, but never overwrites the cost formula.
Law and markets change
The engine version and methodology date matter. An older result may need to be recalculated.
Independent ranking
Affiliate revenue and commercial work do not change the formula or recommendation order. Sponsored material is labelled.
Inspired by ADAC
Transparency matters more than a magic number.
ADAC publishes a common five-year, 75,000 km scenario and separates depreciation, running costs, fixed costs, workshop visits and tyres. It is a strong standard for comparability and disclosed assumptions.
CzymPojade keeps the same reference horizon, then adds the driver's profile, used cars, home and public charging, local tax, energy optimisation and a P10-P90 range. This is our methodology.
Test it with your data
A methodology matters only when you can challenge it.
Choose a profile, enter your mileage and compare cars under identical conditions. The result shows monthly cost, cost per kilometre, the TCO structure and uncertainty range.