How many of your files are really repeats?

Most tuners feel that they do the same jobs over and over. Few have counted. Counting takes an hour and tells you whether File Match would earn its keep in your shop, or whether the library that comes with packages from Solo up is all you need. We would rather you knew before you pay for it.

What should I count?

Two different numbers, because they mean different things.

Exact repeats are reads that are byte for byte the same as one you have tuned. Only these can go out by themselves.

Near repeats are reads with the same software number, or the same software and hardware number, but different bytes. These need a tuner, but the tuner starts with your earlier work open, instead of a blank file.

For most shops the second number is much larger than the first. Reads differ by tool and by method, and a car carries data of its own. See software and hardware numbers for why.

How do I count exact repeats?

Gather your original reads from the last year in one folder. Then take a fingerprint of each file and count the fingerprints that appear more than once. On Linux or a Mac, sha256sum does this. On Windows, Get-FileHash in PowerShell does the same. Sort the results and count the duplicates. A spreadsheet is enough.

There are two traps. Count originals only, because a tuned copy of a read is not a repeat. And locked files from different tools will look different even if the data is the same, so this count is the lowest your real number could be.

How do I count near repeats?

Your job records or file names usually hold the software number. List the numbers, and count how many came up more than once. Be honest about jobs that were for different work: a Stage 1 and a Stage 2 on the same software are a head start for your tuner, but not a repeat.

How do I turn that into time?

Use your own numbers, not ours.

1. Take the exact repeats in a year, and how long one takes you, from opening the job to sending it.
2. Multiply them. That is the time File Match could save at most.
3. Add the jobs that came in while nobody was there, and how long those customers waited. Out of hours work is often where a repeat is worth most: a customer with a car on the ramp gets the file at once.

Then compare that with the add-on on the price page. If it does not pay for itself, keep the library, which saves your tuners time in any case, and look again in six months.

What does the number depend on?

Mostly on your customers. A shop with a few regular fleets or dealers sees far more exact repeats than one with passing trade. A network sees more too, since dealers send similar cars, and a master's library can answer a dealer's job.

It also depends on your library. Files nobody can find do not count. Importing your old files and naming the work on them are what turn a pile of folders into something that can answer a job.

How do I see the real number later?

Run File Match in Simulate for a couple of weeks, as how File Match decides describes. Its recent decisions show what it would have sent on your real work. That is a better answer than any count of old files.