Cargurus, Please tell me how this benefits your customer (aka... The Dealer)????

[Dan Sayer]

Roses are Red
Violets are Blue
Classifieds are Brokers
Why we so mad, bros?

TrueCar, Edmunds, CarGurus, and a plethora of others offer information to shoppers in order to generate traffic in order to sell the supplier a spot on the crazy train that we know as Classifieds. I've spent a lot of time in the past complaining and trying to beat up Classifieds because I felt like they were basically charging me to look like an idiot (I mean I'll be an idiot for free). Like @Tallcool1 stated (but with fewer F-bombs), if your model doesn't fit with the information wars that the Classifieds are arming shoppers with, don't ride the train. If it does, play the game and play it well with merchandising/pricing.

Pricing transparency has been around a looooong time and we still sell cars. We still get leads and walk-ins that have seen lower prices but at least we get to have the conversation of why you can't get that price on this car. Simon Sinek once said, "Transparency doesn't mean sharing every detail. Transparency means always providing the context for our decisions." He was speaking about leadership decisions but I think this also applies to how we speak with shoppers once they have information that may or may not be accurate in order to leverage against us. "Mr. Customer, that price you see on CarGurus is an average sale price range within that model and trim. This unit has the Thinga Majig 2.0 which adds $600 to the base model." IF that fails you can always just squash the data by showing the shopper the https://www.cargurus.com/Cars/TermsOfUse.html section under the "Use of CarGurus Listings Service" which provides price data disclaimers up the wazoo (G-rated is how I roll) amongst other things like they are NOT a BROKER in ALL CAPS . You can find that similar disclaimer on most every Classified and even Trade appraisal sites...

I used to get so mad at the Classifieds and so irritated at the shopper for arming themselves but all they are really doing is reacting to a stigma that we started. Yes, the Classifieds may be helping to perpetuate it but that is their Click Bait for the unforeseeable future and I wouldn't be paying them a dime if they didn't get the traffic they do.

The way I'm approaching 2019 is making sure I have a balanced placement for all stores on the Classifieds that make sense for their region. More importantly, making sure the sales managers and sales people are able to have a calm conversation with the opportunities from any source that bring up price whether as a question or objection.

As for scraping, looking forward to watching @Alex Snyder take on NADA and the world ! I mean CarGurus has language on their own site preventing use of any content in the "Ownership And Intellectual Property" of the aforementioned link. So does KBB at https://www.kbb.com/company/terms-of-service/ in section #3.

 

[ryan.leslie]

"Mr. Customer, that price you see on CarGurus is an average sale price range within that model and trim. This unit has the Thinga Majig 2.0 which adds $600 to the base model."

The way I'm approaching 2019 is making sure the sales managers and sales people are able to have a calm conversation with the opportunities from any source that bring up price whether as a question or objection.

Best post I've read on the forum in a long time @Dan Sayer. You smoked it over the fence.

Averaging anything with a dissimilar data set leads to a false conclusion. I've been talking about this a lot with dealers lately. Many sacrifice gross by relying on the same wide datasets and simple averages to determine a listing price and potential days on lot for used too.

You quoted Sinek, so I thought I'd share a few quotes that I have loosely paraphrased in those discussions with dealers to help point out the problem with a simple average. The third is funny, but admittedly more @Alex Snyder's speed. Hope they help you or your team.

I abhor averages. I like the individual case. A man may have six meals one day and none the next, making an average of three meals per day, but that is not a good way to live. ~Louis D. Brandeis

Say you were standing with one foot in the oven and one foot in an ice bucket. According to the percentage people, you should be perfectly comfortable. ~Bobby Bragan, 1963

The average human has one breast and one testicle. ~Des McHale

 

[Alexander Lau]

Those are dangerous words coming out of Bradley's mouth. Perception is reality and he's got the reach to make an impact. I'll be interested to hear CG's retort.

 

[Dan Sayer]

So, in keeping you up to date on my conversation with CarGurus, here is the string thus far. The first response (very bottom) was from my question to him of "How is the range (New car) determined? Is that range only within that model and trim? If the high end of the range is with all the features and the low end is the base then why would a fully loaded unit get a Poor(red) rating?". This is a Dealer Account Executive replying (I removed his name) and not high level management so please consider that the views may not be completely on par with corp CG. I do appreciate this guy's speed and candor but also realize he is speaking with general caution. I did reply to the last one asking about scraping data from a dealer site. I'll let you know what they say.

Yes completely. I appreciate your feedback and I’ll make sure I forward this to my product team because they are always looking for new ways to improve it. Our new car product is far from perfect and a working progress right now. 95% of your CarGurus leads are going to come on used vehicles. I understand that this can be aggravating if you do work a CarGurus new car lead but this will change. We move very quickly as a company so when something is flawed and recognized it will be changed.

Dealer Account Executive
CarGurus

From: Dan Sayer <[email protected]>
Sent: Tuesday, November 27, 2018 9:08 AM
To:
Subject: Re: Average Price Paid

(EXTERNAL)

There are price variation on New by region though. There are differences in both incentives and volume in markets. I don't think it is safe to say "similarly equipped". Shoppers aren't looking at that term thinking and interpreting it in any other way than "same". Also, you are deciding what a customer should pay by adding red, yellow, green color bands and labeling the range as Good, Fair, etc. If you literally just provided an "average range paid" along with the number range and then clearly stated that "This is a national average of what customers paid for this model and trim level. Your price may vary based on options and equipment, supply, and regional incentives."

I understand the spot a Classified vendor is in and wouldn't pay you a dime if you didn't have the shopper traffic you attract BUT there has to be some agreement on not confusing the shopper or creating anxiety unnecessarily. We do not make money on New cars (and the really really strong dealers are maxing out at a 3% margin). We price our New using tools from vAuto and Cox, who I would assume have more sold data than you. So when we price New to be competitive in the market and tell the shopper we "price to market", having a 3rd party show a national average range on what may or may not be an accurate sample size of data for our region, it erodes at our trust.

Does that make sense?

Dan Sayer | VP of Digital
Anderson Auto Group
Nebraska - Missouri - Kansas

402.464.0661 Office
402.730.7632 Cell
[email protected]
www.andersonautogroup.com

From:
To: "Dan Sayer" <[email protected]>
Sent: Monday, November 26, 2018 6:08:50 PM
Subject: Re: Average Price Paid

Great question. For new cars this is determined by nationwide sales data which is different than are used car pricing analysis which compares pricing on similar vehicles within 50-100 miles.

We get thousands of sales data records per month and compare what price a similarly equipped (make/model/trim) vehicles has been sold. It’s good to note that we are excluding outliers such under or oversold vehicles. We are not deciding this by what we think a customer should pay. This analysis has come from what real customers have already paid.

Another thing to note is we don’t actually give a good or poor rating on the vehicle we just provide this range. I think it’s safe to say that “similarly equipped” doesn’t always account for a fully loaded car with all the bells and whistles. If you come across a well-educated customer that wants a better price on a fully loaded vehicle then you can explain that the range is only taking general make/model/trim into account and not all major options.

Dealer Account Executive
CarGurus

 

[Alexander Lau]

So, in keeping you up to date on my conversation with CarGurus, here is the string thus far. The first response (very bottom) was from my question to him of "How is the range (New car) determined? Is that range only within that model and trim? If the high end of the range is with all the features and the low end is the base then why would a fully loaded unit get a Poor(red) rating?". This is a Dealer Account Executive replying (I removed his name) and not high level management so please consider that the views may not be completely on par with corp CG. I do appreciate this guy's speed and candor but also realize he is speaking with general caution. I did reply to the last one asking about scraping data from a dealer site. I'll let you know what they say.

Yes completely. I appreciate your feedback and I’ll make sure I forward this to my product team because they are always looking for new ways to improve it. Our new car product is far from perfect and a working progress right now. 95% of your CarGurus leads are going to come on used vehicles. I understand that this can be aggravating if you do work a CarGurus new car lead but this will change. We move very quickly as a company so when something is flawed and recognized it will be changed.

Dealer Account Executive
CarGurus

From: Dan Sayer <[email protected]>
Sent: Tuesday, November 27, 2018 9:08 AM
To:
Subject: Re: Average Price Paid

(EXTERNAL)

There are price variation on New by region though. There are differences in both incentives and volume in markets. I don't think it is safe to say "similarly equipped". Shoppers aren't looking at that term thinking and interpreting it in any other way than "same". Also, you are deciding what a customer should pay by adding red, yellow, green color bands and labeling the range as Good, Fair, etc. If you literally just provided an "average range paid" along with the number range and then clearly stated that "This is a national average of what customers paid for this model and trim level. Your price may vary based on options and equipment, supply, and regional incentives."

I understand the spot a Classified vendor is in and wouldn't pay you a dime if you didn't have the shopper traffic you attract BUT there has to be some agreement on not confusing the shopper or creating anxiety unnecessarily. We do not make money on New cars (and the really really strong dealers are maxing out at a 3% margin). We price our New using tools from vAuto and Cox, who I would assume have more sold data than you. So when we price New to be competitive in the market and tell the shopper we "price to market", having a 3rd party show a national average range on what may or may not be an accurate sample size of data for our region, it erodes at our trust.

Does that make sense?

Dan Sayer | VP of Digital
Anderson Auto Group
Nebraska - Missouri - Kansas

402.464.0661 Office
402.730.7632 Cell
[email protected]
www.andersonautogroup.com

From:
To: "Dan Sayer" <[email protected]>
Sent: Monday, November 26, 2018 6:08:50 PM
Subject: Re: Average Price Paid

Great question. For new cars this is determined by nationwide sales data which is different than are used car pricing analysis which compares pricing on similar vehicles within 50-100 miles.

We get thousands of sales data records per month and compare what price a similarly equipped (make/model/trim) vehicles has been sold. It’s good to note that we are excluding outliers such under or oversold vehicles. We are not deciding this by what we think a customer should pay. This analysis has come from what real customers have already paid.

Another thing to note is we don’t actually give a good or poor rating on the vehicle we just provide this range. I think it’s safe to say that “similarly equipped” doesn’t always account for a fully loaded car with all the bells and whistles. If you come across a well-educated customer that wants a better price on a fully loaded vehicle then you can explain that the range is only taking general make/model/trim into account and not all major options.

Dealer Account Executive
CarGurus

Ask them what they'll do if a dealer decides to use a 3rd party to disallow the scraping. Is the deal off...?

 

[Dan Sayer]

So, below is the answer from CG, @Alexander Lau. Just in case we are going to look into something like this anyway - https://blog.jscrambler.com/protect-your-site-against-web-scraping/

It is important to kinda weigh the effort required here and to whether or not this is a hill to die on. I recall plenty of 3rd party vehicle value vendors with claims that they scrape the classifieds multiple times during the day, which is interesting since we have already established that they have the legal language preventing that very thing. There must not be anyone pursuing this type of litigation because ultimately it really may not be hurting anyone in the way it may within other industries?

Yes, the website would need an “Ownership and Intellectual Property” statement in a Terms Of Use section. This would prevent a 3rd party from gathering data from their website.

--
Dealer Account Executive
CarGurus


From: Dan Sayer <[email protected]>
Date: Tuesday, November 27, 2018 at 10:02 AM
To:
Subject: Re: Average Price Paid

(EXTERNAL)

There is a big hubbub in my dealer forums about the legality of 3rd parties scraping dealer sites, that don't have an agreement with the party, for their listing and inventory data. If a dealer has a legal Terms of Use disclaimer on their site, like CarGurus does, would that prevent a 3rd party from gathering data from a site?

Dan Sayer | VP of Digital
Anderson Auto Group
Nebraska - Missouri - Kansas

402.464.0661 Office
402.730.7632 Cell
[email protected]
www.andersonautogroup.com

 

[joe.pistell]

wwaaaa????

"Another thing to note is we don’t actually give a good or poor rating on the vehicle we just provide this range. I think it’s safe to say that “similarly equipped” doesn’t always account for a fully loaded car with all the bells and whistles. If you come across a well-educated customer that wants a better price on a fully loaded vehicle then you can explain that the range is only taking general make/model/trim into account and not all major options.

Dealer Account Executive
CarGurus"

[face palm]
Someone shoot me. No wonder margins are shrinking!! Dealer's pay extra $$ for stuffed up units, only to be FORCED TO REDUCE THE PRICE because CG won't fix the noise from its dealer website VDP scraping.

Unless you log in and fix CG's problem.

What I have learned is that you need to go into your CG dealer dashboard every time new inventory is fed and navigate to the "dealer pricing tool" tab, filter by "newest listings first" and then edit the vehicle options as they are often incorrect or omit important features. Then save and viola, the pricing indicator now refreshes and changes from "fair deal" to "good deal" or "high price" to "fair deal" etc...

Tedious, and a PITA but it works.

@Langley Steinhart, please address this, I'm getting the sense that CG doesn't care about dealers...

 

[joe.pistell]

@Mike Robbins, how often does CG miss? IOW, how often do you find and fix CG listings?

 

[Mike Robbins]

@Mike Robbins, how often does CG miss? IOW, how often do you find and fix CG listings?

Roughly 3/4 of the time there is inaccuracies or ommissions in the options and sometimes the transmissions are wrong as well as incorrect or missing trim levels so I'm in the backend at least once a day updating things.

 

[Tallcool1]

@Mike Robbins, how often does CG miss? IOW, how often do you find and fix CG listings?

I asked my Internet guy, and he agrees with @Mike Robbins . 75% is in his opinion a really good estimate. Some of them aren't even close.