I created a site that enables quick, thorough price comparisons. Acuras have a large number of standard features; this site adjusts for them.

A new feature I recently added instantly converts all prices to monthly payments.

Page for the TSX:

www.truedelta.com/models/tsx.php

 

Why does it say the navi is $1200? It also makes it sound like you can order voice controls for $150....this is not possible as far as I know. Am I reading this wrong?

Also, in the last section it would be helpful if someone can add a trade-in value and how much you owe on the trade, instead of just how much you want to put down. In addition, punching in your own interest rate would be nice, as sometimes you want 4.9% or something along those lines, not just in increments of .25%.

I don't know, I think Acura's own web site is better. At least you know the info is correct, as opposed to sites like this where prices and options can be off. With so many sites like this out there (Edmunds, Vehix, cars.com, KBB, etc.), do we really need another one? What does this site offer that the others don't?

 

great site! i just played around with it and found it really QUICK and easy to use compared ot edmunds and carpoint

 

The site lists features, not options. Options can contain multiple features, especially if they are packages. The advantages of this system:

1. Makes it possible to create one list of items for two cars

2. User doesn't need to know the contents of options, just which features they want

The dollar figures are standard values I assign, not prices. They serve as a rough guide. Again, this enables one list for two cars. I also use these standard values in some calculations.

The main advantages of this system:

1. Only site that provides price comparisons with user-selected features.

2. Only site that automatically selects trim lines and option packages.

The second advantage is least apparent with a Honda/Acura, because there are few if any options.

I went with increments of 1/4% to speed processing, because this way I only need a table of factors. You'll find that differences under 1/4% have a very small impact on payments. 5.9% vs. 6.0% on a $30,000, 60-month loan is about $1.20/month.

Amount you get for a trade is the same as a down payment as far as the calculations are concerned. The only difference can be with taxes, and I don't include taxes.

Finally, you cannot ever be sure the info is correct. I see errors on manufacturer sites all the time. I've even called them or emailed them when there has been conflicting or incomplete information and received an incorrect answer. The only way to be sure information is correct is to see it printed on a window sticker, and even then there might have been a price change after that car was made. For features I sometimes visit dealers to examine cars when the available information is conflicting or unclear.

Every site out has errors, mine included. If there was an error free data source managing my pricing database would take half as much time.

 

3. Only site that doesn't list correct prices.

4. Only site that allows you to pick completely bogus option combinations.

5. Only site with piles upon piles of blatantly wrong and obviously unchecked information.

Nice try but no thanks.

 

Totally unusable site for me.

 

Care to substantiate? To date price comparisons have been run about 65,000 times, yet you're the first person to make these claims, and I honestly don't have a clue what you're talking about.

3. As I've made clear, nowhere do I claim that the values in the features list are prices. At the top of the page they are described as "standard values." All prices on the final results page are correct, at least as much as they are on other sites. The only very rare exception to this is I sometimes split or combine option packages to facilitate processing, but this does not affect the totals.

4. The program automatically eliminates illegal combinations of options. In contrast, at Edmunds you can select options that cannot be ordered together, and it will simply total them up.

5. I don't know what to say about this one. I review prices at least once a month for accuracy, often comparing multiple websites because they all contain errors.

So let's have some examples. Since there are "piles upon piles," let's have five pieces of "blantantly wrong and obviously unchecked information."

If you cannot produce evidence, please explain why you made these claims. I've been pouring a great amount of time and a not insignificant amount of money into a service that I give away for free. I'm very interested in figuring out what sort of person would feel the need to trash it without reason.

So I hope you can produce some evidence. Because the alternative is pretty pathetic.

 

Here's a good one to try out on Edmunds, the Sienna minivan.

http://www.edmunds.com/new/2006/toyo...on=1&x=63&y=20

Check off all the packages and see what happens.

Next, price a Sienna at my site.

http://www.truedelta.com/prices3.php...;0]=275.2006.0

On which site was it easier to get a valid result?

 

The "standard values" part is what is confusing. List prices are much easier to understand, at least for me. And I also prefer the format of listing everything offered in an option (Acura/Honda don't have a lot of these anyway though). For me, it's easier to see what I'm adding to the car when I can see what features are in certain options, not just picking individual features that get lumped into a package anyway if that's the way they come.

In regards to the interest rates, I like to see the exact figure for the finance rate I put in so I can go to the dealer and know what I'm talking about and not be "off" by a few dollars. Maybe a small nitpick, but something I prefer anyway.

Ease of use, at least for ME, is not on your website. It might be useful for others.

 

Fair enough on your points--I hope I was clearly reacting to jpt's malicious post.

If you want exact payments, there are many payment calculators on the Internet. Try bankrate.com. Even using my site, if you enter the two closest interest rates then extrpolate you'll get the right answer--it's a linear function within such a narrow span. But this is more work than I expect people to do. I've considered baking this method into the programming, but have higher priorities at the moment.

Unfortunately, there is no simple way to handle option packages. With other sites, you have to read through the contents of all of them and figure out which makes the most sense. I personally find consumerguide.com most useful for this, though they are perhaps the most error-prone of all of the sites I've tried.

With mine, you select the features and it finds the best package. In some cases this package will not be a good value for you. When pricing a single model it lists the selected packages, their contents, and the percentage of those contents you actually asked for. It's then possible to return to the features page and uncheck those features to drop that package. Pretty cumbersome, so I doubt this is done often.

What the site was designed for, and what the great majority of people use it for, is price comparisons. Selecting features rather than options is the most feasible way to handle comparisons, as option packages and even options like "power seat" differ from car to car.

Say you want to compare the TSX and BMW 325i and get differences adjusted and unadjusted for features:

http://www.truedelta.com/prices3.php...1;1]=13.2006.0

You can't do this anywhere else.

 

I see, so you just toss a useless number on that page for no reason (or is it there to explicitly distract and confuse?) -- thanks!

In response to 4 and 5, I must say that I stand corrected after some further perusal of the site. Of the several errors I thought I had spotted only one turned out to be wrong (the Audi S4 is listed as having an optional engine that is identical to the standard engine). The "standard value" listing that you have is quite bad and makes the rest of the application seem much worse than it is, but the "minimize/maximize shared features" features are in fact extremely nice and your final prices seem accurate.

I therefore make a full reversal of my earlier opinion -- this is a cute site and I would have benefitted from using it when I was shopping. You should add the ability to compare more than two prices at a time though (six would not be too many, and an unlimited number via an "add more" button would be best). And lose the "standard values" for options -- only I know how much these things are worth to me, and actual prices are totally uncorrelated with the ones you list on that page.

 

Oh, and for fuck's sake reorder the transmission options so manual comes before automatic

 

Make that two: the TSX is listed as having a body kit standard.

Also the speed-sensitive windshield wipers are not listed.

 

Very nice website!

lots of time i see was spent on it. very very very nice!

 

No feature gives me more trouble than "body kit." Sometimes this feature is obvious, especially when it's optional as on the Mazda3, Mazda6, and some others. But in many cases its hard to tell when a rocker panel molding becomes a side skirt.

I've thought of changing the term to "side skirts," but fear this would confuse people even more. I've even considered dropping the feature altogether, or at least reducing its standard value to 0.

The TSX is very borderline, but I decided that especially with the tweaks for 2006 the car in standard form appears to have about as much of a body kit as a Chevrolet Cobalt SS or Mazda6 Sport.

Some features I track but do not present because I don't think I have complete information on all cars. Speed sensitive wipers is one of these.

Another that I am more concerned about is the MP3 input that is standard on the TSX. This is an increasingly common feature, but information on whether a specific car has it continues to be spotty. In a few years it'll probably be standard on everything anyway.

 

Forgot about the S4 issue.

The database includes the engine with HP, RPM, TQ, RPM, even though all of these specs are not presented. In the S4 the engine is slightly different depending on transmission, such that the Tiptronic engine peaks at 6800 rather than 7000.

As for allowing more than two cars, because of option packages it's nasty enough to make two fit together. Fitting even a third would get very messy in many cases.

One way this has been done by other sites available only to the industry (jato v4, autosite pro) is that you first configure one car, and then have the option to automatically configure additional cars as closely as possible to the car you first configured. The advantage of this approach is you can compare 3+ cars. The disadvantage is that you're still only really comparing pairs, each car to the focal, and if you select a feature on the focal that isn't available on one or more of the others, tough luck.

When you compare cars on my site, the feature list generated is composed of features available on both cars but not standard on both cars. So that last problem generally won't occur. Sometimes trim lines cause problems here, but then the program sorts the conflict out and removes an ineligible feature before pricing.

Finally, packages are handled differently with min and max than when you select a feature. When you select a feature, it'll buy the lowest cost package no matter what. But when using min and max the program assumes you just want a tight comparison, so it will only "buy" a package if at least half of its standard value is sought. It's a bit more complicated than this, but that's the gist of it.

 

I'd like to. Of all the cars I've owned, only one has been an automatic. But I set the powertrain list up so that the typical user would have to make as few selections as possible. So the auto is always listed first.

I'm very much looking forward to the restyled 3-Series coupe and convertible. This year the lists for that model are very confusing. Luckily it's an extreme case.

 

Interesting. The reason I specifically called the TSX body kit is that there is a different optional (usually but not always DPI) body kit. Honestly I would probably drop the feature altogether from the comparisons, as cars look different regardless.

I see your difficulty in multi-way comparisons, but I still suspect a computer could do a better job than a human at finding "comparably equipped" vehicles in a multi-way contest (particularly with minimize/maximize shared features turned on). It is actually one of the most difficult problems I had when shopping and my seriously-considered set only had four cars in it.

I suppose you are right about the ordering of the powertrains. I just wish the typical car buyer would have better taste :-P

 

Thanks for taking this discussion in such a constructive direction.

The computer can only do what I program it to do. I'm tempted to attempt a 3-way, but it gets messy. There are some other issues, but they're relatively minor.

I've come very close to redoing the body kit for the exact reason you describe. The main thing that keeps me from doing this is it is sometimes an extra-cost feature.

If I do make a change, what I'm most likely to do is change the value to 0 and change the name to "body kit (optional)." I'd then only list it when it's not a feature on the base trim.

Credit is also currently given for a rear spoiler. This isn't quite as messy a feature, but its close. I give the same value for a tiny lip spoiler as a huge wing.

A final similar feature is leather, because there are different grades and extents of coverage. I currently give a single value, but obviously it's far from exact. Luckily, in most cases the cars being compared are in the same segment, so any leather they offer is likely to be similar in quality and coverage. But compare a Mazda3 and S600...

 

I finally understand what you're doing with the "standard values" for features and I still don't like it. Accounting for a feature that is standard on one car and not on another is great; putting a price tag (which, even if it is meticulously calculated based on the average cost of that feature in cars in which it's optional, will have a high standard error, as you noted in the extreme case of leather) on something that isn't available is just confusing, I think. Then again, since you give what I would call the correct price comparison as the first (well, second) result on your page, I suppose it doesn't hurt much.

At the very least, from a UI point of view I would refrain from listing the standard values in the Step 5 checklist.

 

I'll think about it.

I didn't include the "standard values" on that page until recently. I thought they might help people decide which features they truly wanted. Otherwise you get a "just check 'em all, they're free" mentality. For this reason any decent market research on demand for features will always include prices.

But I can also see your point of view. No one else has complained about these values before, but then most people who have a problem with something probably just leave the site and I never hear from them.

I'd much rather hear criticism, as it helps me improve the site. I probably don't seem terribly responsive in this case, but I've made many improvements in the past based on feedback I've received.

 

I'm always happy to provide constructive criticism (As you've seen, I'm occasionally a little to quick to write off an entire product as worthless, too, but if I haven't done that, I'm happy to provide constructive criticism ). Your application is very nice and I'm glad to provide any input that may make it better.

I develop (as part of a very small team) a software product with a very complicated interface so I spend most of my days debating user-interface minutiae; I have therefore become much more picky than average about minor UI features. You would not believe how heated some discussions about a matter of a few pixels one way or another can get

This is true, but you aren't doing demand research; you're providing comparisons. The "prices" (which have a high probability of being very far from the actual price -- especially given that packages will likely have to be pulled in) on the step-5 page seem to be counterproductive for this purpose.

 

would changing "standard price" to "average price" be better, or is that still not the accurate word to use?

 

I don't use "price" in this context, just "value." I base it on the lowest amount charged, not the average, because I figure everyone pays roughly the same for the feature but mark it up to varying amounts.

I have added some explanation to the top of the page. I think this will actually help many people make sense of the step in general. So the feedback here has helped.