TrueDelta

The latest major change: the bottom half of the homepage now includes links to the brand pages, from which you can get to a page where all the site’s informaton for a specific model is either displayed or directly linked to. This should make it easier for people seeking information on a single model.

As part of this change, “reviews” and “specs” now each have their own button in the nav.

Those model pages can also be accessed by clicking on the model’s name elsewhere on the site, including an added “breadcrumb” on the Fuel Economy, Repair History, and “Why (Not) This Car?” pages.

Is any of this helping? As always, I’ve been receiving virtually no actionable feedback.

It took TrueDelta nearly six months to attract the first 1,000 members. After all, people were signing up for a survey that hadn’t even started happening.

Well, as of yesterday we have 1,000 of a single model, the Honda Accord. No real surprise that the Accord reached this mark first. Hondas have always been the most-researched cars on the site. Clearly Honda buyers love data more than the average car buyer. And the Accord is usually the top selling Honda model (though the Civic has outsold it some months this year).

This said, a forum for the Honda Accord, DriveAccord.net, has really helped get the word out. Anyone interested in owners’ insights–that’s the place to visit.

The BMW 3-Series will probably be next to 1,000. Currently 785 are enrolled. Quite amazing, since the BMW doesn’t sell nearly as many of these as Honda does Accords.

Various flavors of compact VW (Jetta, Rabbit, Golf, GTI) are in third, with 676 enrolled. After that a number of models are around 600, including two more Hondas–the Odyssey and Civic.

I’ve written before about ”hidden price increases,” where the sticker price stays the same but the invoice price goes up. Though such increases, dealers’ margins on car have gotten smaller and smaller over the past 20 years. And when the dealer pays more, you tend to pay more, even if the sticker price remains the same.

The latest development: after cutting the margins on base prices to five percent or so a few years back, this year Volvo is cutting the margins on options from 14 percent to six percent.

For example, invoice on the V70’s $2,995 Premium Package was $2,575 last year. For the 2009 model year, it’s $2,815, essentially a hidden price increase of $240.

Now, the factory makes more money, sometimes far more money, on options than on the base car. But when these profits aren’t shared with the dealer, what incentive does the dealer have to sell you stuff you don’t want? So in a way, hidden price increases might actually be bad for the manufacturer and good for the consumer.

Even as it has made advances in other areas, Hyundai has lagged its more experienced competitors in fuel economy. Even the tiny Accent, with a 1.6-liter engine and automatic transmission, only managed 24 miles-per-gallon in the city and 33 on the highway in the EPA’s tests last year. A Toyota Yaris: 29 in the city, and 35 on the highway.

But, among other things, the new Genesis luxury sedan signalled that Hyundai was making great strides in this area. Even with a 375-horsepower V8 this large luxury sedan earned ratings of 17 city and 25 highway, besting Toyota’s equivalent, the Lexus LS 460, by one MPG in each test. The new Borrego SUV from sister company Kia similarly earned best-in-class ratings.

And the Accent? Both numbers are up by two this year. How’d they do it? Hyundai credits “smart engineering enhancements to the Accent’s engine, transmission, tire pressure, powertrain management system and power steering pump.” In other words, a little bit here and a little bit there. No breakthroughs, just intensive refinement.

TrueDelta started out on your regular basic $4/month shared hosting plan. Then, in the spring of 2006, we moved to a VPS (Virtual Private Server) plan at Apollo Hosting. Apollo was selected based on a strong recommendation, and for the first year or so they were excellent. Support was 24/7, via toll-free phone or live chat. And little support was needed.

Then the support was outsourced and no longer 24/7. And the server started going down more often. The last straw came a few weeks ago, when the entire Apollo system was down for two hours–their own site was down, so no way to contact them online. A call to the 877 number posted on their homepage got a “this number is no longer in service” response. Had they simply closed up shop? That was not a good morning.

Their site came back, and so did the 877 number. Also, supposedly support has been brough back in-house. But for months we’ve been looking for a new host, and finally learned of one that seems very good this month.

So, as of this morning, we’re hosted by WiredTree, a small hosting company in the Chicago area that, so far, seems outstanding. Some of the people running WiredTree used to work for a larger host, LiquidWeb. The usual  story, LiquidWeb started out strong, but went downhill as it grew. Supposedly the people at WiredTree learned from that experience, and won’t repeat it. We’re hopful…

A downside of adding pieces one by one to the TrueDelta site is that, as these accumulate, the whole can become disorganized. More than a few people have found the resulting site organization confusing and even frustrating.

I’ve been working on the site organization. Among the changes: clicking on “Reliability” on the homepage now goes to the latest survey results rather than to the page about joining the panel. To compensate, other changes were made to make it clearer where to go to join, and why.

These changes are minor compared to another. For various reasons, when the Vehicle Repair History and Fuel Economy Surveys were launched they used a second, entirely separate vehicle list from the main reliability survey. In hindsight, this was a bad decision. Aside from the extra work of enrolling each vehicle twice, too often members thought they’d enrolled a new vehicle in the main survey when they’d only added it to the list for the other surveys.

So for the past few weeks I’ve been working on merging the two lists. Nearly every major script used to operate the site was affected, and much new code was required.

But as of late last night this project is complete. There is now a single vehicle list. It is now only necessary to enroll each vehicle once to participate in all the surveys. And hopefully it will be clearer when the Vehicle Reliability Survey should be used to report repairs (whenever possible) and when the Vehicle Repair History Survey should be (only when a repair cannot be reported on the main survey). Probably still some work to do there…

One result of this integration: it is now possible to enter specific bodystyle, trim, and powertrain information for nearly all of the 2002 or newer vehicles in the panel, just like it has been possible with the second list. (It will be possible to do this for earlier model years in the future.)

Now on to the next project.

GM plans to launch the Chevrolet Volt in 2010. The Volt runs on battery power for the first 40 miles or so, and after that a small gas engine will cycle on as needed to recharge the batteries. Plug-in versions of parallel hybrids, where the engine can directly power the wheel, are also coming.

Earlier today I learned of an interesting discussion at Volt forum gm-volt.com. At issue: how should the Volt’s EPA ratings be calculated. After all, some owners will rarely if every use any gas.

GM wants to simply run the Volt through the standard EPA cycle, which would yield a number over 100. The EPA wants to assume that the gas engine must be used to recharge the battery at the end of the trip, and have it do so. Reportedly, this would yield a 48 MPG figure.

For obvious reasons, GM is fighting for the method that yields the higher number.

Neither party is necessarily correct. If in the real world someone could plug the car in at the end of the trip, they would, and no gas would be used to recharge the battery. If they had to keep driving after the battery was depleted, then the gas engine would have to recharge the battery.

So, it’s really a question of trip length, and how often and for how long the gas engine comes on after the battery is depleted to its low mark (about a 30 percent charge, to preserve battery life).

To me, it seems a bit silly to force the Volt into the regular EPA system. What is really needed are three sets of city/highway numbers.

The first would represent efficiency and range prior to the initial depletion of the battery. The test would have to include two short cycles of just a few miles, one for city and one for highway, to be repeated until the battery reaches its low mark. The standard EPA cycles might or might not work for this purpose, depending on how gameable they are by tweaking battery size and how often and for how long the gas engine runs. The calculation would them divide the number of miles driven by the number of kilowatt-hours used, to get a miles-per-kwh figure.

This is the first pair of numbers. The second pair would report the distance the vehicle could be driven in electric-only mode.

The second test would start with the battery at its low mark and run over the regular EPA cycles. This would yield a third set of EPA numbers.

With plug-in parallel hybrids, the first set of numbers would also have to include MPG figures, for the amount of gas used prior to battery depletion. So, essentially a fourth set of numbers.

Complicated, yes, but it’s not possible to simplify these figures further without oversimplifying them.

The Volt and other plug-ins also pose a challenge for TrueDelta’s Fuel Economy Survey.  Clearly a different survey will be required for these vehicles to usefully report their energy efficiency.

No one has better vehicle reliability data on fairly new vehicles than the manufacturers, since they keep detailed records of warranty claims. Problem is, they’re not about to publicly release this data.

But publicly traded American corporations do have to report how much they pay out in warranty claims each year at the corporate level. I recently learned that these figures are posted at WarrantyWeek.com.

For GM, we see about $4.5 billion in annual warranty expense, which according to the site represents about 2.5 percent of sales. For Ford, the figure is about $4.0 billion, and 2.6 percent. You’ll also find warranty payments for many automotive suppliers.

Interesting information, but a few obvious shortcomings:

1. We only have figures at the corporate level, not for individual models.

2. Unless sales are steady, and they’ve been in sharp decline for these companies, then the “percent of sales” figure is off, since it’s based on sales in the latest year, while most of the warranty claims will be for vehicles from past years.

3. Some warranty costs are paid by the suppliers–and we don’t know who made the vehicles they’re paying claims on.

Last month we posted the lastest results for TrueDelta’s Vehicle Reliability Survey.

This set of results included a couple of early 2009 models, the Nissan Murano and Jaguar XF, well head of any other public source of vehicle reliability information.

The Murano’s launch has been cleaner than most recent Nissan new product launches. Owners of the all-new SUV reported a repair rate of 16 successful repair trips per 100 cars per year, which is much better than the typical new car.

The Jaguar XF has not been faring nearly as well. The sample size for the XF is very small, so TrueDelta won’t be issuing an official result for this model. But the reported repair rate is so high–well over three times the average–that even this limited amount of data clearly indicates a higher than normal number of “teething issues”–chiefly minor electrical problems–for the new Jag. A more solid result will be included with the next update, in November.

The new results also include the 2008 SMART ForTwo and Saturn ASTRA. The SMART has attracted a strong fan base owing to its unique character and exceptional city fuel economy. Averaging a little under a single repair trip per year, the ForTwo requires about 50 percent more repairs than the typical 2008 model.

The Saturn ASTRA is assembled by GM in Europe. The last car GM imported from Europe, the Cadillac Catera, could be troublesome. Not so with the ASTRA, whose owners reported a repair rate of 29 successful repair trips per 100 cars per year–about half that of the typical 2008 model.

We’ve previously reported much worse than average repair rates for the 2008 Cadillac CTS and 2008 Chrysler minivans. Both had similarly poor scores in the latest update.

One possible exception to GM’s typical first-year problems: the 2008 Chevrolet Malibu. Only 15 owners participated, but none of them reported a repair. This suggests a better than average repair rate.

There is hope for the Jaguar XF, SMART ForTwo, Cadillac CTS, and new Chrysler minivans. Many new models improve in their second model year, or even later in their first year, as the manufacturer identifies and corrects “bugs.” With its prompt quarterly updates, TrueDelta will report any such improvements soon after they occur.