The repair data behind our writing

We read each repair order from our shop-management system, classify its work into a vehicle system, and add it to a bucket for that vehicle, system, and period. The bucket is all we keep. The order itself is never stored here.

Only electric and hybrid system work is counted. Oil changes, tires, and general maintenance pass through unclassified, so these totals are smaller than the shops’ full workload and should not be read as business volume.

When an article is published, its number is frozen and stamped with the date it was true. Later growth in the corpus does not silently rewrite an older article.

No customer names, no vehicle identification numbers, no license plates, no individual visits. The aggregate is the smallest unit that exists in this dataset, so there is nothing personal to expose, on this page or behind it.

Recent periods are partial. A quarter still in progress shows fewer orders because it is not finished, not because the work slowed down.

8 records carry a date before the period above, which we read as a mis-keyed year rather than a genuine repair. They are counted in the total and excluded from the coverage dates.