Earlier this week, J.D. Power gave the automotive world a new score to ponder: the Vehicle Launch Index (VLI). This addition to the survey giant’s quality canon aims to measure how well manufacturers launch new or redesigned models. It’s a worthy endeavor; a new model’s success in its first few months often predicts its long-term sales and profitability. But what do these new J.D. Power scores tell us? In the immortal words of Jeff Spicoli: I don’t know.
This first set of VLI scores includes 27 models introduced in the first ten months of 2008. The top scorer: the Hyundai Genesis. The bottom scorer: the Toyota Matrix. Speaking to Automotive News [sub], J.D. Power Senior Vice President Gary Dilts characterized the Hyundai luxury sedan’s launch as “flawless;” a characterization that captured the media’s immediate attention and pleased the Korean automaker to no end. The rear wheel-drive sedan’s score: 689 out of 1,000. Flawless? On a curve?
That’s only the first of many VLI oddities. The VW Routan falls just three points shy of the 582-point average. The slow-selling German-engineered (or not) minivan had a decent launch—who knew? The Audi A4 ranks much lower, tying with the now-defunct Pontiac G8. So should Volkswagen of America aim for more launches like the Routan’s, and fewer like the A4’s? Meanwhile, the Ford Flex, which has had an agonizingly slow start and (according to J.D. Power’s IQS) poor initial quality, ranks near the top of the VLI.
How can models that are selling poorly or that have poor initial quality attain decent, even high scores for their launch/re-launch? Drilling down, J.D. Power’s VLI scores include turn rate, vehicle revenue, dealer gross profit, incentive spend, credit quality, residual value, customer appeal and initial quality.
J.D. Power’s famous Initial Quality Survey (IQS) combines two totally different elements: design quality and mechanical quality. As a result, the line between them is muddled. In fact, it’s still not entirely clear what a particular IQS score represents.
With the VLI, J.D. Power has taken this lack of clarity to a whole new level. Where the IQS has two subscores, the VLI has [at least] eight. So when looking at any particular score, it’s impossible to say what it represents. The Ford Flex and VW Routan did well at . . . some things. The Audi A4 blew it with . . . maybe the same things. Maybe something else. Who knows?
These model subscores aren’t the only part of the VLI that J.D. Power’s keeping from public scrunity. The actual formula has also been kept under wraps. Nothing is said about how the variables and weights in this secret formula were determined—except that the formula followed from “carefully analyzing more than 90 vehicle launches.”
Without knowing this formula, it’s far from clear how much weight each factor receives. Judging from the scores, initial quality doesn’t carry much weight. Nor does how close a model comes to achieving its sales targets. Maybe all eight-plus factors are equally weighted, such that none of them carries much weight by itself?
One thing is clear: the VLI isn’t evaluating launches from a car buyer’s perspective.
Car buyers, who’ve learned to be wary of buying a new model in its first year, are focused on their need for information on a vehicle’s initial reliability. “Appeal” they can judge with their own eyes. They couldn’t care less about dealer profitability and inventory turns (whatever that is). Car buyers’ needs would be much better served by a metric that focuses on initial reliability and that provides this information as quickly as possible. [Fair disclosure: Mr. Karesh’s TrueDelta offers a Car Reliability Survey offers that information.]
Of course, it should come as no surprise that a J.D. Power-crafted metric would focus on the manufacturer’s needs. J.D. Power earns its millions by serving manufacturers, not by serving car buyers. Their M.O.: persuade manufacturers that they need a high score based on a proprietary formula, then sell them the data and consulting services that will help them boost their scores.
In the official press release, J.D. Power offers some tips to its corporate benfactors gratis: set realistic prices, set realistic sales targets, style the vehicle well, and achieve high initial quality. (Notable by their absence: advertising and PR.) Of course, everyone in the biz already knows this much; much as everyone knows that the way to lose weight is to eat less and exercise more. The hard part isn’t knowing what should be done, but doing it.
Meanwhile, J.D. gets another service to sell to the biz, while auto industry execs get more hardware for their ego shelf and another bullet point for their resume. Well good for them. But the VLI’s yet another example of car biz navel gazing, a metric that marginalizes a central, well-established fact: the customer comes first.
Absolutely, it’s an industry-facing measurement, not a customer-facing one. What’s wrong with having such measurements as well?
Having done product planning in the past, I can vouch for the fact that product launches are complex beasts that are difficult to execute well. The VLI provides the manufacturers with an independent scorecard on the success of their launch planning and activities (yes, they’ll pay JD Power to get the detailed data) and gives them external feedback on which areas were done well and which need improvement.
Customer comes first? Sure. But you still need to launch new products, and if you don’t do that well, those products will flop in the marketplace even if they were well-designed and well-engineered, with the customer in mind.
I sure see the new A4s all over the place. But in NJ, you’ll find more German sedans than American.
I don’t doubt that the details behind the scores can be very interesting and informative. But what do the overall scores add to the subscores behind them? Anything? What do these overall scores even mean? I’m at a loss as to how to interpret them. Higher is of course better, but better at what?
I received a call from a journalist the other day. He couldn’t tell me exactly what he was calling about, because this study was embargoed at the time. But he asked me what I thought of the Hyundai Genesis launch success. I asked him if he meant in terms of initial reliability or sales. Even without knowing that we were discussing the VLI, I said that they were two totally different things, and needed to be considered separately. On what basis could you combine the two into a single score?
Even for a manufacturer-focused score, it seems unwise to weight initial quality as low as it seems to be weighted. Though I suppose manufacturers already have enough incentive to improve quality from other sources, including the IQS.
It’s a pity that J.D. Power isn’t using the word “launch” in this context….
Summary: it means SQUAT.
Compare the new car launch of these two vehicles:
1. Miata
2. Last-gen T-bird
Statistically , the survey will be 50% right 100% overall…
Can’t manufacturers evaluate their own launches?
Set goals. Plan. Execute. Evaluate. Repeat.
Duh.
Without knowing this formula, it’s far from clear how much weight each factor receives. Judging from the scores, initial quality doesn’t carry much weight. Nor does how close a model comes to achieving its sales targets. Maybe all eight-plus factors are equally weighted, such that none of them carries much weight by itself?
You can get a basic answer here: http://www.jdpower.com/corporate/automotive/vehiclelaunch/figure2.htm
The point, though, is that you are attacking this as a strawman. I’m reasonably sure that those who pay to get the reports get all the detail and then some. It’s not that the detail isn’t available, it’s just not available to you. Pay gobs of money and your perspective will change.
Car buyers’ needs would be much better served by a metric that focuses on initial reliability and that provides this information as quickly as possible.
Sure. But this isn’t particularly geared for car buyers. You’re complaining because some of the data produced for the industry is geared toward the industry instead of the consumer?
The only reason that JD Power pushes any of its data, such as the IQS, to the public is so that the manufacturers will want to pay for it. The VLI takes two existing indexes/ product lines (IQS and APEAL) and combines them with a third, so it sounds to me to be a bit of marginal revenue booster.
I’d suggest that any auto company that can’t evaluate the success (or astounding lack thereof) of their own vehicle launch, shouldn’t be in business in the first place.
You mean, like, I have to pay millions of dollars to a third party company to know whether I’m making money on new vehicle X? Are my tax dollars being spent on this B.S.??
So they are offering a launch rating based on criteria and formulas they will not disclose? How is that supposed to help anyone – even a manufacturer? If they don’t break out particular aspects of the launch project how can anything be learned or improved?
I give this a big fat zero on my personal Survey Launch Index (SLI).
See how easy it is? Check please!
Michael Karesh
I am soooo glad this is being discussed.
When it came out, I felt the same as I did after listening to the Obama speech on healthcare the other night.
What was just said????
Instead of making things clearer, I was even more confused.
First, Model Lot Movement.
There is another sight that shows “the most successful sales each month”, based upon how long a car stays on the lot.
Wouldn’t this depend totally upon how many of a model there is in inventory?
I mean, if a dealer keeps low inventory of a model, the less the model sits…right?
So, what good is this information at all?
And, doesn’t the Genesis ONLY sell 1K to 2,000 cars a month?
Is this a great introduction?
The Flex does 4K and the MKS 1 to 2K, perhaps the same as the Genesis.
So…why the scores?
“Can’t manufacturers evaluate their own launches?
Set goals. Plan. Execute. Evaluate. Repeat.”
Yes, of course we do. Quality, in the various ways we measure it, is plotted continuously. When a new car is launched it will have various quality improvement actions in it, the success, or otherwise, of these is closely monitored.
But what we can’t easily do is compare our launches success with other manufacturers, or other countries. So, this VLI thing may be a cheap way of getting that data.
Or not.
You guys missed the answer and it is right in front of your face. The score’s primary driving factor is the vehicle’s “Gotta Have it Factor”. Can’t believe the B&B missed it…
paulie: And, doesn’t the Genesis ONLY sell 1K to 2,000 cars a month?
Is this a great introduction?
The Flex does 4K and the MKS 1 to 2K, perhaps the same as the Genesis.
The launch of a new 7-series BMW can be a rocking success even if it only sells 1/50th of the Camry sales. It’s not the total numbers, it’s meeting the objectives and making people want the car.
And part of that, too, is managing the pipeline and dealer inventory — if the dealer lots are overflowing with Genesis it will be perceived as less desirable than if the dealers sell them quickly as they arrive on the lots.
As Greg said, having some data on launches outside your own sandbox can be highly valuable.
In my city, downtown (Orlando, Fl), I see a ton of VW sedans (Golf, Jedi, and Passat) along with two tons of A4. I’ve never seen a Routan, and hope I never do. On the other hand, for my own perverse and depraved purposes, I would not mind owning the life-sized cardboard cutout of Ms Shields that the local VW dealer has hidden in an obscure corner of his showroom.
@paulie: And, doesn’t the Genesis ONLY sell 1K to 2,000 cars a month?
Is this a great introduction?
Consider it’s a brand new, near-luxury sedan from Hyundai (Hyundai!), with an introduction during the worst recession in decades. It seems rather remarkable that Hyundai is able to sell 1K-2K units/month. That speaks volumes about how much the public perception of Hyundai has changed for the better. Acura and Buick (and perhaps BMW and MB) are green with envy.