An oil change shop video audit is a scheduled review of bay camera footage to confirm each service was done the way your process says: hood up, drain plug back in, the right oil in, level checked, nothing left loose, and a ticket that matches the car. Pull a small random sample of visits per store every week, score each one against the same checklist, and review every claim, comeback and voided ticket on top of that.
This guide gives you the 12 checkpoints, a scoring sheet you can copy, how many clips to pull, and what cameras can and can't prove.
Quick answer
- What to check: 12 steps from car-in to car-out, scored pass, miss or can't see.
- How many: a random sample per store each week, plus every claim, comeback, void and drive-off.
- Who reviews: someone outside the store, so the review holds up and protects good techs.
- What it needs: a camera on the top side of the bay, one in the pit if you have one, clocks synced to the POS, and footage kept long enough to pull.
What do the top results on this search get wrong?
When we searched "oil change shop video audit" in November 2026, the top result was a sponsored white paper on NOLN, and the next two were the same paper on the vendor's own site (Solink's oil change guide). It lists six problems (drive-offs, bay time, OSHA risks, unlogged services, service claims and false alarms), but the actual guide sits behind a form. After that, the page fills up with forum threads from drivers whose oil was never changed or whose filter went on with no oil added.
So owners searching this get a gated vendor guide or a horror story. Nobody lays out what to actually look for on the video, how many visits to review, or what to do with the result. That's what this page is for.
Why audit oil changes on video at all?
Because the ticket says what was sold, not what was done. The two biggest risks in a quick lube both happen under the hood and in the pit, where a manager standing at the counter can't see:
- Service quality misses. A drain plug left finger-tight, an old filter gasket left on, the wrong oil grade, no oil level check. These are the stories that end up in the news, like a 2014 CBS Minnesota report on a student's car that died after an oil change.
- Billing misses. Work done with no ticket, or a ticket with a service that wasn't done. We cover the billing side in the off-book oil changes guide and the void and discount audit. This page is about the process side.
A video audit catches both early, while the car is still in your system and the tech still remembers the visit.
What should you check on each visit? The 12 checkpoints
Score every visit against the same list. Keep it short enough that a reviewer can do one visit in a few minutes at double speed.
| # | Checkpoint | Camera that shows it | What a miss looks like |
|---|---|---|---|
| 1 | Car greeted and checked in before work starts | Top side or lot | Hood up before anyone talked to the driver |
| 2 | Hood up and fender cover on | Top side | No cover, or tools resting on paint |
| 3 | Drain plug removed and old oil drained | Pit | Plug out, then car lowered or moved before draining finishes |
| 4 | Old filter off, gasket checked | Pit or top side | Old filter goes straight in the bin with no look at the mount |
| 5 | Drain plug back in and tightened with a tool | Pit | Plug in by hand only, or no tool seen |
| 6 | New filter on | Pit or top side | Filter step skipped on a ticket that lists one |
| 7 | Oil added from the bulk line or bottle | Top side | Short fill, or wrong product line selected |
| 8 | Engine run and leak check below | Pit and top side | No one looking under the car after start-up |
| 9 | Oil level checked on the dipstick | Top side | Hood closed without a dipstick pull |
| 10 | Extra services on the ticket actually done | Top side | Ticket lists air filter or wipers; nothing changed |
| 11 | Hood closed, cap and dipstick secured, tools out | Top side | Cap left on the engine cover, rag left in the bay |
| 12 | Car leaves with a closed ticket | Top side, lot, POS | Car out, ticket still open or cancelled |
Two things cameras can't tell you. They can't read a torque value, only that a tool went on the plug. And they can't measure litres or quarts, only that oil went in. For fill quantity, compare the bulk oil meter or the POS line to the vehicle's spec. Our bulk oil variance guide covers that check.
How do you run the audit each week?
1. Sync the clocks first
Your recorder, your bay cameras and your POS need to agree on the time, within a minute. If they drift, you'll waste the review searching for the right car. Check this once a month.
2. Pick the visits
Use two lists:
- Random sample. A set number of closed tickets per store per week, picked at random from different days and shifts, so every tech and every shift shows up over a month.
- Always review. Every damage or service claim, every comeback within 30 days, every voided or cancelled ticket, and every drive-off.
Random visits tell you how the process runs on a normal day. The always-review list catches the expensive ones.
3. Pull the clips the same week
Most recorders overwrite old footage once the drive fills up. How many days you keep depends on your camera count, resolution and storage, so check it on each store's recorder. Pull claim clips the same day the claim comes in, and save them outside the recorder.
4. Score each visit
Mark each of the 12 checkpoints as pass, miss or can't see. "Can't see" isn't a pass. A lot of can't-sees on one bay means a camera needs to move, not that the tech did something wrong.
5. Talk to the tech within a week
Show the clip. Most misses are habits, not intent, and a short clip fixes a habit faster than a memo. Our service advisor coaching guide uses the same approach for the counter.
6. Track it by store and checkpoint
One miss is a conversation. The same checkpoint missed across a store is a training gap. The same checkpoint missed across every store is a process problem, and that's on the process, not the techs.
Scoring sheet you can copy
One row per visit. The columns are what make the monthly roll-up possible.
| Store | Date / time | Bay | Ticket # | Tech | Checks passed | Misses | Can't see | Miss details | Coached on |
|---|---|---|---|---|---|---|---|---|---|
| Store A | Tue 10:42 | 2 | 104883 | T1 | 11 | 1 (#9) | 0 | Hood closed, no dipstick pull | Fri |
| Store A | Wed 15:10 | 1 | 104951 | T2 | 10 | 0 | 2 (#3, #5) | Pit camera blocked by a cart | Camera moved |
The second row is the important one. Two can't-sees on the pit steps meant the camera was blocked, so the fix was the camera, not the tech.
How much review time does it take?
An example calculation (made-up numbers, not real store data):
- 4 stores
- 10 random visits per store per week, plus about 3 always-review visits per store
- About 4 minutes per visit at double speed with the checklist open
That's 4 × 13 × 4 = 208 minutes, or about 3.5 hours a week for one reviewer. Swap in your own store count and sample size. If that's more than anyone has, cut the random sample before you cut the always-review list.
Where does AI help, and where do you still need a person?
AI on your existing cameras is good at the parts that are about time and presence: when a car pulls in, when the hood goes up and down, whether someone is in the pit, when the car leaves, and whether a ticket exists for that visit. That lets it flag the visits worth watching (a three-minute service, a car with no ticket, a hood that never opened on an engine service) instead of picking at random. Our quick lube camera analytics guide covers placement, accuracy and cost in detail.
A person is still better at the judgment calls: did the tech actually look at the gasket, was the right product line selected, was the customer shown the old filter. Use AI to choose the clips and a person to score them.
What about staff privacy and trust?
Tell your techs what the cameras are for and who sees the footage, put it in writing, and post signs. Use video only. Recording audio brings in consent rules that vary by state and province, so check before you turn it on. Most important, share good clips too. Techs trust the audit when it's used to back them up on a bogus claim as often as it's used to correct a miss.
How Trilix runs this for multi-store operators
We build this for quick lube and auto service groups. AI on the bay cameras you already have logs every visit, matches it to the POS ticket, and picks the visits worth reviewing. Misses and can't-sees roll into an owner dashboard by store, bay and checkpoint, so you see the pattern without opening a single clip unless you want to. You own the footage and the data.
Want to see what a video audit would flag at your stores?
We'll walk through your bay camera setup and POS export with you and show what a weekly review would look like.