What to Send Your Tuner: The File, Log and Build Sheet That Save You Money

Remote tuning has a reputation for being slow, and most of the delay is not the tuner. As of August 2026, the single biggest determinant of how many revision rounds a job takes is what arrives in the first message — and that is entirely under the car owner's control.

Here is the package that turns three rounds into one.

The four things that matter

1. Your original stock read. Not a file someone posted as stock for your model. Yours, taken off your own controller. It is the reference point for what the car actually shipped with and the recovery point if a write goes wrong. The reasons this matters more than people expect are covered in what is actually inside a tune file, and the storage discipline in tune file management and backups.

2. The file currently on the car. Whatever it is running right now, even if you did not build it and do not trust it. A tuner cannot reason about a change without knowing the starting point, and "I think it has a tune on it" is not a starting point.

3. A clean datalog of the car doing the thing in question. This is the item that decides everything. More on it below.

4. An honest build sheet. What is fitted, what fuel you run, and what the car is for.

That is the whole package. Everything else is helpful context.

The datalog is the expensive part to get wrong

A calibration is a hypothesis; a log is the evidence. Which means a log that does not contain the relevant moment is not evidence of anything, and a revision built on it is a guess wearing a costume.

Three rules make a log useful:

Capture the actual condition. If the question is about full-load fuelling, the log needs a clean sustained pull in a gear that loads the engine properly — not a snap of throttle in first. If the question is a drivability complaint, the log needs the complaint happening. Our walkthrough of reading your first full-load pull covers what a good pull looks like and how to get one safely and legally.

Include the channels needed to interpret it. A log without intake air temperature cannot explain timing behaviour. A log without a wideband reading cannot say anything about full-load fuelling. A log without engine load cannot place any of it in context. Which datalog channels to record covers the standard set and why each one earns its place.

Keep it short and say where to look. A few purposeful minutes beat forty minutes of commuting. If the event happened around two and a half minutes in, say so. Nobody should have to hunt.

What you send	What it lets the tuner do	What its absence costs
Stock read	Establish the true baseline; recover if needed	Assumptions about what the car started as
Current file	Reason about a change from a known state	An entire round trip to obtain it
Purposeful datalog	See what the engine actually did	A revision built on inference
Honest build sheet	Anticipate what the combination needs	A calibration aimed at a car you do not own
A one-paragraph problem statement	Answer the question you actually asked	A general tune-up instead of your fix

The build sheet, and why honesty beats completeness

A build sheet does not have to be exhaustive. It has to be true.

Include, as best you know:

• Engine and transmission, including anything internal that was changed.
• Induction and exhaust, with part identity where you have it.
• Injectors and fuel system, by part number if at all possible — injector characterisation is one of the few places where a guess is genuinely dangerous, as our injector data article explains.
• Power adder, including pulley or turbine sizing if applicable.
• Fuel you actually run, including which grade and whether you ever use ethanol blends.
• What the car is for — daily driving, occasional track use, towing. This changes what a good calibration looks like, not just how aggressive it is.
• Known faults and recent codes, including ones you consider unrelated.
• Anything you are unsure about, flagged as unsure.

That last point is the one that saves the most time. A tuner can work with declared uncertainty. They cannot work around wrong information delivered confidently — and a wrong injector part number in particular will produce a calibration that is internally consistent and completely wrong for your car.

What slows jobs down

Sending a fault instead of a log. "It hesitates sometimes" is a starting point for a conversation, not a data set. Capture it happening.

Undisclosed prior tunes. Very common on used cars. A car carrying an unknown calibration is a different job from a stock car, and the person working on it needs to know which one they have.

Fixing the car mid-job. Changing hardware between revision rounds invalidates the logs the tuner has been reasoning from. Finish the mechanical work, then start the calibration work.

Untreated mechanical problems. A vacuum leak, a failing sensor, a misfire or a tired fuel pump will not be calibrated away. Time spent tuning around a mechanical fault is time spent building a file that is only correct while the fault persists — and the fault will get worse.

No stated goal. "Make it better" produces a generic result. "It falls flat above 5,500 rpm in third and I want that fixed" produces a fix.

What you should get back

The exchange runs both ways. A well-run job returns:

• A file with a clear name and a note of what changed rather than an unlabelled attachment.
• An explanation you can follow. You do not need a lecture, but you should know what was changed and why.
• Instructions for the next log — what to capture and under what conditions.
• A stated revision count. How many rounds are included, agreed before money moves. This single line explains most price differences between quotes, as our cost guide sets out.

If you are choosing between remote tuning and other routes in the first place, remote online tuning covers where it works well and where it does not, and our comparison of the copilot, e-tuner and dyno-shop paths frames the trade-offs.

The same discipline if you tune your own car

If you are doing the work yourself, you are both parties in this exchange, and the habits matter more rather than less. Archive the stock read permanently. Name every revision by date and change, not by version number. Pair each revision with the log that verified it. Write down what you were trying to fix before you change anything.

Six months later, that record is the difference between understanding your own calibration and re-deriving it. And the round trip you save is your own weekend.

If you want the audit half done for you, the copilot reads your current tables, flags what looks unsafe and gives you an ordered change list — which is the same package a good tuner would send back, without the day of waiting. What that costs is on the pricing page, and if you have a case that does not fit any of the above, get in touch.

Frequently asked questions

What does a tuner actually need from me?

Four things: your original stock read, the file currently on the car, a clean datalog of the car doing the thing you are asking about, and an honest list of what is fitted and what fuel you run. Those four remove nearly all the guesswork. Everything else — photos, dyno sheets, fault-code history — is helpful context but not the core package.

Why do tuners keep asking for another datalog?

Usually because the log they received does not contain the moment they need to see, or is missing a channel they need to interpret it. A log of gentle driving cannot answer a question about full-load fuelling, and a log without intake air temperature cannot explain timing behaviour. Each round trip costs a day or more, so the log is the highest-leverage thing you control.

How long should a datalog be?

Long enough to contain the event and short enough to find it. A few minutes of driving that includes the specific condition in question beats forty minutes of commuting. If you are asked about full load, capture a clean sustained pull in a suitable gear. If you are asked about a drivability complaint, capture the complaint actually happening and note roughly when it occurred.

Should I send my stock file even if the car is already tuned?

Yes, and it is the item people most often leave out. The stock read is the reference point for what the car shipped with and the recovery point if something goes wrong. If a previous owner or shop flashed the car and you have no stock read, say so plainly — it changes what the tuner can safely assume about anything currently in the file.

What if I do not know exactly what is on my car?

Say that rather than guessing. An honest 'I bought it modified and I believe it has a cam and long-tube headers, but I have no paperwork' is far more useful than a confident list that turns out to be wrong. Tuners can work with uncertainty they know about. They cannot work around wrong information they were given as fact.

Does sending more information always help?

Up to a point. The four core items plus a short, specific description of the problem is the useful package. Sending forty logs with no note about which one shows the issue transfers your work to someone billing for their time. Organise before you send — it costs you ten minutes and saves a revision round.

Does this apply if I am tuning the car myself?

Even more so, because you are both parties. The same discipline — archived stock read, named revisions, a purposeful log paired with each change — is what makes your own work reviewable weeks later. The difference is that a bad package only wastes your evening instead of a revision round.