The most useful slide in this NADA Show 2026 session is titled “Quality Doesn’t Matter w/out Volume”, and it is a funnel of four bars. Start with 100 repair orders. A dealership doing this well records video on 90 of them, sends 88 of those, and gets 67 watched. The handout carries the version for a dealership doing it badly: 80 recorded, 60 sent, 36 watched.
Same 100 cars. Sixty seven customers who watched their technician point at a worn brake pad, against thirty six. Warner Jones calls the number at the end an Absolute View Rate, and the point of leading with it is blunt: the finest inspection video in the world is worth nothing if it never gets recorded, never gets sent, or never gets opened.
This is the first session in our NADA series to sit in the workshop rather than on the sales floor, alongside our earlier piece on driving service revenue with customer data, and it is the most operationally useful of the twelve. It is also, and these two things sit together uncomfortably, the one whose numbers need the most careful handling.
Who is talking, and why that matters here
Warner Jones is Senior Vice President at TruVideo, which sells the video inspection platform the session is implicitly about. Every figure quoted comes from TruVideo’s own installed base, described on the opening slide as eleven years and fifty million videos.
That is not a reason to dismiss the session. Fifty million videos is a serious dataset, and nobody outside the vendors has one. But it does mean the numbers are a supplier describing the performance of its own product to an audience it would like to sell to, with no published methodology, and we will come back to what that does and does not support.
The good news is that the operational content stands entirely on its own. You could disbelieve every statistic in the deck and the rubric below would still be worth adopting.
It is also worth noting what this deck does not do. Another session at the same show made the case for video using the familiar circuit of generic statistics, the 95 per cent brand recall claim and its relatives, none of which survived our checking in Wired to Win. Jones uses none of them. He uses his own operational data instead, which is a more honest choice even though it carries the problems described further down.
The rubric worth stealing
Key eight is “Good, Better, Best”, and it grades inspection videos on a scale a service manager can actually apply while watching one. The handout adds a fourth tier the on-stage slide leaves off, which is Fail.
| Tier | What it looks like |
|---|---|
| Fail | Under 30 seconds. No narration, no vehicle, just the repair order sheet. Car not on the lift, bonnet closed. |
| Good | 30 to 60 seconds. Car on the lift, bonnet open. Problems only. Measurements shown. Inform and recommend. |
| Better | 60 seconds or more. Car up and down. Good news as well as bad. Filters and battery results. Measurements taken with gauges. Inform, educate and recommend. |
| Best | 120 seconds or more. Car up and down. All major components, full walkaround. Filters and battery results. Measurements with gauges, explained in context. Inform, educate, recommend and implicate. |
What makes this good is that every tier is observable. There is no judgement call about enthusiasm or rapport. Either the car is on the lift or it is not. Either the bonnet is open or it is not. A workshop foreman can grade ten videos on a Friday afternoon without needing to agree on anything subjective first, and that is precisely why it will survive contact with a real service department.
Note also what separates Fail from Good: the failing video is one where the technician filmed the paperwork instead of the car. That is not laziness so much as a technician who has not been told what the video is for.
Inform, educate, recommend, implicate
Key five is “Word Tracks”, and it is the four-part narration structure that the Good, Better, Best tiers are built out of:
- Inform. Let the customer know what you observed.
- Educate. Tell them how the item works on their vehicle.
- Recommend. Provide your recommendation.
- Implicate. Let the customer know what happens if they do not address it.
Most technicians stop after the first one, which is why most inspection videos do not convert. “Your brake pads are at 4 millimetres” informs and does nothing else. The educate step is the one that turns a number into a fact the customer owns, and the tagging convention on key four supports it: green is good, yellow means future attention, red requires immediate attention, with the honest context that tyres normally start around 10/32nds and should be replaced at 2/32nds. Told that way, six is not a scary number, it is half your tread, and the customer can decide.
Implicate is the step to handle carefully, and we will return to it under the Consumer Protection Act below.
Staging, which is where most of it is won
Key one splits into inspection flow and staging, and the staging checklist is the part that costs nothing and changes everything. Before the camera comes out:
- Bonnet open, car on the lift
- Gauges ready for tyres and brakes
- Measurements already taken
- Filters pulled
- Washer fluid reservoir open
- Battery printout or digital screen to hand
- Alignment report ready
The flow advice is the same idea: run the multipoint inspection on the ground then in the air, and run the video in the reverse order, in the air then on the ground, so that half the inspection is done before filming starts. The technique key adds three words that are worth putting on a workshop wall: slow, smooth, light on. And the “Tell a Story” structure gives the running order: introduction, licence plate, primary concern, major components, good and bad, recommendations, thank you.
The telephone game, which is the real argument for video
One slide under key seven does more to justify video inspection than any of the revenue figures. It shows the same six findings three times, under the headings Technician, Advisor and Customer Confidant, and the list gets shorter each time.
The technician finds all six: air filter, tie rod, brakes, battery, CV boot, tyres. By the time the service advisor relays it, two have gone. By the time the customer repeats it to whoever they phone for a second opinion, only brakes and battery remain.
That is the case for video stated without a single statistic: not that video is persuasive, but that the current process loses two thirds of the findings somewhere between the workshop and the person paying. Video removes the relay. It is the most honest argument in the session, and it is the one that needs no vendor data to support it.
The money, converted
The deck carries several monetary figures. Converted at 16.21 rand to the dollar, the mid-market rate on 17 August 2026:
- Parts and labour per repair order rising from about R1,799 to R2,205 between the second and third quarters of 2025, a delta of roughly R405, which the slide labels a 23 per cent increase.
- Hours per repair order of 2.43 where the customer viewed the video against 1.5 where they did not, a difference of 0.93 hours, labelled plus 62 per cent.
- A delta rate comparison of about R551 against R1,508, roughly R956 more, or 2.7 times, for technicians scoring above 80 on TruVideo’s own Video Quality Score.
- A technician incentive model built on a bonus of roughly R49 per video against an assumed R973 improvement per repair order at a 50 per cent margin.
We checked the internal arithmetic and it holds: the percentages, deltas and multiples on those slides are consistent with the figures they sit next to. That is a point in the deck’s favour and worth saying, because it is not always true of conference slides in this series.
Where the evidence does not reach
Internal consistency is not the same as proof, and the central claim has a problem that no amount of sample size fixes.
The headline comparison is correlational, and the causation plausibly runs the other way. Repair orders where the customer watched the video show 2.43 hours against 1.5 where they did not. That is presented as video driving revenue. But customers who open and watch an inspection video are not a random half of your customers. They are the ones already worried about their car, already expecting bad news, and quite often the ones whose vehicle had more wrong with it, which is exactly why the technician had more to film. A bigger job produces a longer, more interesting video and a more motivated viewer. The arrow between “watched” and “spent more” points in both directions, and the slide only draws it one way.
The quarter on quarter uplift has no control group. Parts and labour per repair order rose about R405 between two quarters of 2025. Nothing on the slide compares that to stores which changed nothing over the same period, and service revenue per order moves on parts pricing, labour rates, vehicle age mix and season regardless of what anybody films.
The Video Quality Score is the vendor’s own instrument. The claim that technicians scoring above 80 produce 2.7 times the delta is a supplier grading dealerships on a proprietary scale of its own design and then reporting that the high scorers do better. That may well be true and useful. It is not independently checkable, and the score’s construction is not published.
The eleven years and fifty million videos figure describes the platform, not the finding. A large installed base tells you the company has been operating a while. It says nothing about how the specific comparisons above were sampled.
None of this makes the advice wrong. It means you should adopt the rubric because the rubric is sound, measure your own before and after, and treat 62 per cent as the vendor’s number rather than your forecast.
What changes in a South African workshop
This section is ours. The session assumes an American service drive, and four things do not carry across.
The delivery channel is WhatsApp, not SMS or email. The whole funnel above turns on the “sent to viewed” step, and in South Africa a link that arrives by email will lose to one that arrives on WhatsApp by a wide margin. If your platform cannot send on WhatsApp, your absolute view rate will sit closer to the deck’s bad example than its good one, regardless of video quality.
The customer pays for the data, and that argues against the Best tier. This is the single biggest difference. The rubric rewards length: 120 seconds or more scores Best. A customer on a prepaid package is spending their own airtime to watch you walk around their car, and a longer video is a more expensive one to open. We would not assume the American tier ranking holds here. Test it: if your 60 second videos are watched to the end and your 150 second videos are abandoned at 30 seconds, then Better is your Best, and your rubric should say so. Measure completion rate, not just view rate.
Language. A narration recorded in English only will not do the educate step for a large share of customers in most South African markets. If a technician is more fluent in isiZulu or Afrikaans or Sesotho and the customer is too, the video should be in that language. This is a quality gain, not an accommodation.
Power and connectivity. The technique key says light on, and the technology key is about footage quality. Both assume a lit workshop and a connection good enough to upload. Where load shedding is still a factor, the practical fix is a torch discipline and letting videos queue and upload when the connection returns rather than losing them.
POPIA, and the line under “implicate”
Also ours, and worth resolving before you roll this out rather than after.
The “Tell a Story” running order puts the licence plate second, right after the introduction, as proof the video is of the right car. That is sound practice and it also means every video you make contains a vehicle registration, tied to a named customer and a repair order, stored on a third party platform. Under POPIA that is personal information and it needs the ordinary treatment: a lawful basis, a retention period somebody has actually decided on, a processor agreement with the platform, and an answer to where the footage is hosted if it leaves the country. Technicians should also know not to film other customers’ cars, plates or people in the background.
The second issue is the implicate step combined with the technician incentive on key seven. Paying a technician per video is fine. But the moment somebody is paid in relation to what they recommend, and is also the person telling a customer on camera what will happen if the work is not done, you have created a pressure that the Consumer Protection Act takes an interest in. The Act prohibits false, misleading or deceptive representations, and a consequence stated on video is a representation you have recorded and handed to the customer in evidence. Implicate honestly, or not at all. In practice that means describing what the wear means and letting the customer draw the conclusion, rather than predicting a failure you cannot support.
None of this is legal advice, and it is worth a conversation with whoever handles your compliance before the first video goes out.
Where to start
- Measure your own funnel first: of 100 repair orders last month, how many were recorded, sent, opened and watched to the end. That single number tells you whether you have a quality problem or a volume problem, and they need different fixes. Our piece on building your own cost and conversion reporting covers the mechanics if you do not have them.
- Put the staging checklist on the workshop wall. It costs nothing and it moves videos out of the Fail tier immediately.
- Grade ten videos against Fail, Good, Better and Best this week. Do it with the technicians in the room, not about them.
- Teach the four word tracks in order, and check for the educate step specifically, because it is the one that goes missing.
- Get it onto WhatsApp, then test length against completion rather than assuming longer is better.
Of the twelve sessions we have covered from this show, this one has the widest gap between the quality of its practical advice, which is the best of the set, and the strength of its evidence, which is a supplier’s own. Take the rubric. Run your own numbers.
Source
- Ten Keys to Video Quality That Drive Revenue: Next Level Inspection Videos, presented by Warner Jones, Senior Vice President at TruVideo, at NADA Show 2026, Las Vegas, 3 to 6 February 2026. Reviewed from the thirty page handout and the full forty two minute recording.
- The recording carries no audio track, so the slides and the inspection footage played on stage were read in full but the spoken commentary was not available to us. Everything attributed to the presenter is text shown on a slide. Slide titles are taken from the recording, which labels them more fully than the handout’s contents list does.
- Every statistic in the deck originates with TruVideo, the presenter’s employer and the vendor of the platform described, drawn from its own installed base and published with no methodology, sampling frame or control group. The internal arithmetic on those slides is self-consistent, which we checked. The criticisms of causation, of the missing control group, and of the proprietary Video Quality Score are ours.
- Dollar figures were converted at 16.21 rand to the dollar, the mid-market rate on 17 August 2026, and rounded. They are the presenter’s demonstration figures, not benchmarks for a South African workshop.
- Ours rather than the presenter’s: the WhatsApp delivery point, the argument that customer-paid data undermines the Best tier and that completion rate matters more than view rate here, the language point, the load shedding and upload queue note, and the entire POPIA and Consumer Protection Act section including the interaction between technician incentives and the implicate step. Nothing here is legal or compliance advice.