Years ago I purchased a white 2011 Golf TDi with a manual transmission (the TDi moniker means “turbo-diesel” for those unfamiliar with Volkswagen naming). I bought it for myself also looking for my son to learn to drive a stick (a lost skill among millennials).
My son and I both loved the car — he nicknamed it “the Stormtrooper.” It had a shocking amount of room in the back seat and was tremendous fun to drive — despite low horsepower the torque was incredible.
Unlike my experience with BMWs since 2000, the Stormtrooper really WAS a car for those who love to drive. (BMWs used to offer that experience — like those I drove from time to time in the 1980s and 1990s.)
The Stormtrooper was my son’s primary car for over a year during which time he learned to love European design. (Why, after half a century, can’t the Asian car makers execute smart designs like those on good European cars?) Then came the scandal and the VW Diesel buy-back.
We offered final salutes to Stormtrooper turning it in during the spring of 2017 — a sad day despite being well compensated by Volkswagen. (Yes, that’s my son in the photo.)
BUT…Now the Golf TDi Diesel is Back!
My son had been looking for a new car — a hatchback for his film/photography work. About a week ago we noticed VW dealers selling 2010-2014 TDi’s modified to meet environmental requirements. The prices were quite good. And since these vehicles had been off the road since 2016 or 2017, they also generally had low miles and were in great condition.
We test drove several on Friday morning, put a deposit on a black 2013 with 22,000 miles and a 2-year warranty, and he bought the car on Saturday.
He. Is. Thrilled.
YET…After All of VW’s Sturm und Drang (und Modification)…his Modified TDi has the Same Performance and Gas Mileage as Our 2011
Wait. What? It’s was a shock to drive a small bucketload of these modified cars and find them to be essentially what they were before the fix. His new car is as fully satisfying to drive as our 2011 Golf was and he’s been getting 50 MPG on the highway.
Why, then, did VW cheat the tests? We’ll never know. So far VW’s public explanations have all the hallmarks of “blamestorming.” We’re told 2 engineers (or a small group) decided to implement the software crime. They, and a few executives, are held responsible.
Right. The problem is far more likely a systemic problem which put undue pressure on the engineers. And my suspicion (from a career in marketing and engineering) is that specs played an outsized role.
If I’m right, that puts a healthy share of responsibility with marketing. And it gives me opportunity to write about the tortuous paths marketers follow when they and the company become obsessed with specs.
Some cautions about specs. Understand, you can’t live without them. But when you live with them, they quite often lead you astray. As Bruce Clark observed after he read this post: “Making the numbers” may have a host of unintended consequences.
Caution 1: Bureaucracies Put Too Much Faith in Specs
“Market requirements” are far too often reduced to merely to specifications. Bureaucratically, this makes the company happy but those same specs may create marketing mayhem.
Here’s a few screwy metric situations I’ve seen in my career:
- Vacuum cleaners are often sold on the highest possible Watt, Amp, Air Watt, or Volt ratings. Except these ratings have almost nothing to do with actual vacuum performance. Vacuum performance comes down to a design which ensures powerful airflow, attachments which put that airflow to use, a good bag system for capturing and getting rid of dust, and great usability features.
- Supercomputers always had to exceed competitor Mflop results on “industry standard benchmarks” (Mflops = millions of floating point operations per second). Except those benchmarks reflect performance only under very narrow conditions — not the true conditions within which a real computer runs. While I worked in the era of vector processors and minimal parallelism, this issue has become far more serious in a world of massively parallel supercomputers.
- Automobiles have to hit safety ratings even though only certain crashes are tested. There is important debate to be had looking at whether those safety ratings help ensure your safety in the dangerous crash you actually might have.
- Marketing departments in retailers set arbitrary numbers of “pieces” which need to be in hand tools sets — as if purchasing consumers paid any attention to those numbers. (A few might, the vast majority don’t. My conclusion is these “# pieces” specs are all about making the executives happy.)
I suspect VW engineers were given specs without market guidance. The scandal may have happened because no one knew that compromise could have been made on the specs and still ended up with tremendous sales and vast numbers of very satisfied customers.
Caution 2: Specs Do Not Ensure Customer Satisfaction
My son test drove the Mazda 3 as well as some Subaru’s in his search. The Golf TDi has a measly 140 horsepower while those other cars have 170-200 HP. Except, under most driving conditions the Golf is much more interesting to drive. Why? Torque. Oversimplifying it, horsepower starts you up from a stoplight while torque rules everywhere else. And where these other cars have 170 lb-ft of torque the Golf TDi has 236 lb-ft. (The VW salesman repeated an old adage that “we brag about horsepower but we feel torque.”)
Specs cannot ever tell a whole story. That also means, beyond meeting some basic minimums, they typically have little to do with customer satisfaction. Satisfaction depends on whether the product delivers an experience customers value and enjoy enough to come back for more.
Specs sometimes even cause the opposite of what’s intended. Competing to score highly on safety tests has led SUV manufacturers to raise their window sills and lower the roof — since tempered glass isn’t as strong as the other options. My leased 2016 Ford Explorer was like this. Last year we replaced the Explorer with a ten year old (2008) Honda Pilot and were shocked at the superb visibility from the car. We love the car far more than the Ford because of the visibility — visibility is so nice that it is more important than the silly switch choices Honda has made and the pedestrian exterior.
But here’s a question: In helping SUVs score well on crash tests, have manufacturers hurt overall driving safety with reduced visibility? Is it more important to avoid a crash or survive the one you couldn’t see?
Bottom line for VW: The Golf TDi was already a great car — did they need to cheat?
Caution 3: Specs Often Don’t Increase Sales
I’ve worked with a lot of companies who demand engineering create the product so they can make specific advertising claims. In those companies, claims become more important than building a product customers will like. It’s been disappointing in my career in engineering, product marketing, sales, and advertising to see how often companies believe specific claims are the key to market success. It’s not usually true.
This isn’t to say that specific claims are unimportant. But I see are a lot of companies tied in knots in order to make claims that don’t actually matter to customers.
Detailed specifications are critical a small portion of the time. So if you use them too much you set yourself up for problems. The government checks claims. Your competitors can beat your claims.
There are situations where specific claims are important. But 90% of the claims I see aren’t.
It’s hard to envision Golf sales suffering without the cheat.
Caution 4: “Independent” Industry Testing Doesn’t Ensure Quality
The idea of independent organizations testing products to ensure quality seems like a common sense good idea — except it’s not. And here I’m referring to official groups like NSF who “approves” water purification products or the IIHS who tests car safety.
Too often “independent testing” groups become classic examples of Goodhart’s Law:
When a measure becomes a target, it ceases to be a good measure
VW wouldn’t be alone in getting caught between customer satisfaction and a desperate marketing need to ensure they are rated well by independent organizations.
This isn’t only an issue in consumer goods. Just this morning I ran across an article about groups producing “ESG” ratings — ratings relied on by investors as they buy securities and stocks. ESG is supposed to tell us which companies are socially responsible. The article points out that there’s no agreement about what makes a company ‘responsible’. No surprise given that we can’t accurately estimate gas mileage and safety.
Marketers live between a rock and a hard place: If they don’t score well they won’t sell well. But scoring well may require making an inferior product.
Caution 5: Danger from the Press
Another place marketers are caught between a rock and a hard place is with the press covering their kinds of products. (Automotive being incredibly complex here — over-run with a very mature press world filled with highly opinionated writers.)
Many companies pursue specs not to sell to customers but to sell the press. If the press doesn’t recommend your car/vacuum/computer/… then you won’t even have the opportunity to sell to customers. (The same is true within retail — where selling to the retail merchants is sometimes more important than selling good products to customers.)
Hence, a statistic like “horsepower” can get you a headline. The complexity of torque doesn’t often. Oddly, you’ll also get headlines by “beating” your competitor with a 37 MPG vehicle as opposed to the competition’s 36 MPG. It’s a meaningless difference to a customer but may be crucial not in the echo chamber of the press. Safety ratings mights start as important but they become critical when marketers try to get press coverage.
I’ll even suggest that every car is overloaded with tech because the press loves writing about tech — not because all that tech is important to building strong interest among customers.
In looking at the system which led VW to cheat, we need to move outside the walls of their HQ. In part, it’s worth remembering in the late 00’s when this cheat happened the Golf TDi was competing with the fascination and hype around hybrids — and were losing presence in the press.
Caution 6: The Ratings Errors of Consumer Reports and Others
While I started life enthusiastic about consumer reports, I now think there should be a special place reserved in one of Dante’s circles of hell for Consumer Reports.
It’s not that they aren’t well meaning — they are. But CR never clearly acknowledges that their subtle review preferences may actually produce rankings that are misleading — or that this often harms manufacturers who make outstanding products.
CR, for example, starts with a vague qualitative preference for “value” where cost is more important than a good product. This makes them unable to adapt when innovative new things come along — because their structures emphasizes easily comparable “average” quality technology. So if a manufacturer builds something amazing then puts a premium price on it (because it’s worth it), it’s likely CR will rate product lower than the customer value it delivers.
I’ve seen major misunderstanding among CR reviewers in nearly every category — vacuums, power tools, cookware, automobiles, etc…
In one of the “guy” magazines, a client of mine fought a similar rating problem.. Their reviews were based on old categorizations and assumptions about 8V, 12V, 18V and 24V drills. My client had innovated putting out a brushless 12V drill. Brushless tech is incredible — making a 12V brushless drill more powerful than an 18V drill with brushes. Try to sell that to reviewers who have to live in neat little categories. They ended up lumped with all other 12V’s. So they looked low power and extremely expensive when they were well priced for the high power they delivered.
In another case, a magazine tested approaches to joining wood — I believe they looked at 10 joint types. My client ranked 7th on the list. So when a magazine reader looked at the list, it sure looked like their product wasn’t all that good (only 7th out of 10). EXCEPT… The first 5 joint types were minor modifications of the mortise and tenon joint —absolutely no significant variation. So the truth in the testing was my client’s joints were the third best behind only mortise and tenon and one other.
Subtleties like these are lost on magazines and organizations like Consumer Reports. Far too often this system discourages smart innovation.
What about VW and Their Diesel Disaster?
Hopefully, my concern about specs resonates (can I call it “Spec-ticism”?). Were it only the first 3 items on my list, companies could solve the problem, ensure their advertising sells products as well as brand, and they wouldthrive. Unfortunately, societal forces ALSO contribute to the systemic failure which resulted in the VW cheat.
I wasn’t there. So there are also things I don’t know. Perhaps, at the time, the solutions were too costly. Perhaps, at the time, they were impractical. Regardless, VW should look at how specs in the car biz contributed to this massive failure.
THE GAME IS ON!
Specs are a bit of a game. There are those specs which are really important for customer sales. There are those critical to customer satisfaction. There are those specs which are entirely unimportant. And there are those specs which are purely gimmicks to get good ratings and reviews.
This is where it gets fun — because a marketer has to be savvy to all the types and make very smart judgements about the role each type plays in their mix. It’s not straightforward. But playing the game brilliantly (and legally) is the only route to success.
©2019 Doug Garnett — All Rights Reserved
Through my company Protonik LLC based in Portland Oregon, I advise a select group of clients to drive success with better marketing of new and innovative products. I also work with clients attempting to bring new life to Shelf Potatoes or take their existing products to new markets. You can read more about these services and my unusual background (math, aerospace, supercomputers, consumer goods & national TV ads) at www.Protonik.net.