Help us put a method to the test

Who can take part

All points must apply:

• Tesla Model 3 or Model Y, first registered 2019 or later. Model S and Model X are welcome but run as a separate, exploratory group outside the stratum targets.
• You are the keeper, the holder of the Tesla account, and of legal age. Access to the interface is tied to the account — a family member cannot grant it validly.
• The vehicle is registered in Germany and mostly driven there.
• You keep the connection in place for at least ten weeks.
• You charge at least once a week. Without charging there are no data points.
• You consent to the study processing — separate from any purchase and revocable at any time.

What does not fit the analysis

• A battery replacement or pack repair in the history or during the study: the ageing curve is then no longer comparable. Such vehicles stay interesting as case studies, but we keep them out of the headline figures.
• A planned sale during the measurement phase — a change of keeper breaks both consent and history. If you want to sell, our regular offering suits you better.
• Accident and salvage vehicles, modifications to the battery management, retrofitted packs.

Vehicles from our own circle are not excluded, but they are marked, reported separately and kept out of the headline figures.

What happens

1. Connect

   Connect your Tesla account once, with read-only access. It takes a few minutes.

2. Measure

   After that, nothing. Drive and charge as usual. We read data when the vehicle is awake anyway — we never wake your vehicle.

3. Certificate

   As soon as enough usable data points have accumulated, we issue your signed certificate.

4. Repetition

   About four weeks later, the second certificate. Comparing the two is the real point of the study.

What you get

• Your signed battery certificate free of charge (normally €79).
• A second certificate about four weeks later.
• Free monitoring during the study and for six months afterwards.
• Your raw data as a file — states of charge, energy values, timestamps.

There is no money. Cash for taking part in a study strikes us as the wrong signal.

We do not promise a reference measurement. It belongs to a possible second wave and will only be asked about if that wave actually happens.

What happens to your data

Through the official interface we read: state of charge, remaining energy, odometer, charging sessions, battery temperature and timestamps. Your email address and the details from the application form are added to that. Of your postcode we store only the first two digits — that is enough for the climate region.

We do not request location data. On the Fleet API the position comes from a separate endpoint (drive_state), and that one is not part of our query — we read charge_state, vehicle_state, gui_settings, vehicle_config and climate_state only. We request no command scopes either: the vehicle_device_data scope is read-only, so we cannot wake or control your vehicle even technically.

Your vehicle identification number is stored in encrypted form only. Analysis and publication are exclusively pseudonymous: without names, without vehicle identification numbers, without location details.

We delete your vehicle's raw data 24 months after the end of the study. Application data of people who do not take part is deleted no later than three months after the end of the study. The pseudonymous analysis dataset stays published permanently, because it is the basis of the results.

You can stop at any time, without giving reasons and without disadvantage. We then delete your personal study data. Aggregated results already published cannot be withdrawn; a certificate already issued remains yours and remains valid.

Further reading: Privacy policyStudy protocolParticipant information and consent

Limits

• The Akkubrief is an estimate with a stated uncertainty band, not a laboratory measurement.
• It is not an official document and does not replace an expert appraisal.
• This wave has no independent energy measurement to compare against. What is examined is repeatability and availability — the results of this wave say nothing beyond that.
• People who volunteer are not a random sample. That is why we ask for your own expectation and name the bias in the publication instead of hiding it.
• We run and fund the study of our own method. That is a conflict of interest. It is the reason for the protocol fixed in advance, for the open data and the public participant list — and for publishing whatever the outcome.

Current demand

We do not invite in order of arrival but by demand. This is how the strata look right now:

Enrolled: 40

Public participant list

We are examining our own method. That is why the protocol with all analysis rules is fixed in advance, and why the participant list is public from inclusion onwards.

Within seven days of your inclusion, a line about your vehicle appears in a public list — for example “S-017 · Model 3 LR 2021 · NMC · 90–110 thousand km”. No entry is ever deleted: leaving is added, not removed.

This is the core of the credibility. Anyone on the list from day one cannot later quietly disappear from the analysis — not even when their result is inconvenient.

Visible: identifier, model, variant, model year, battery type, mileage range and status of participation.

Not visible: your name, your vehicle identification number, your place of residence.

The uncomfortable questions