The Surprising Reason Your Used Cars Get Stuck (OSV)

[WillMcG32]

http://www.dealerrefresh.com/why-your-used-cars-get-stuck-on-your-lot/

Bringing this article to the forum.

So I read this article 2-3 times (one of those may have been a skim) and what I failed to get out of it is how one comes to attain "Online Search Volume" easily. So I hope Chad takes a stab at this to help us out.

In light of what transpired in the Twitterverse, Ed and Joe went back and forth on VDPs and some other things like Ed's girlfriend.

Being a vAuto guy, the one metric that can be obtained (even outside of vAuto - just not as good of a number ) is Market Days Supply. The way vAuto calculates it is below:

Market Day’s Supply is calculated by dividing current market quantity of the vehicle at it’s year, make, model, trim level with its exact equipment configuration by the average daily retail sales rate over the past 45 days. In other words, if there’s 10 identically equipped vehicles for sale in the market and over the past 45 days they’ve been selling at 1 a day retail, then you divide 10 by 1 to derive a 10 market day’s supply of that vehicle.

 

[ryan.leslie]

I'm really glad this article has sparked some discussion here on Refresh. I reached out to Chad as well as Dr. Franke for some additional insight into the research and OSV metric. Many of you know Chad already, Dr. Franke is the chief data scientist at Vast. His resume is impressive and includes Trilogy and a company that knows a little something about consumer search that you may have heard of called Google. I'm so excited to be a part of the team at CarStory. I'm never in danger of being the smartest guy in the room.

So I read this article 2-3 times (one of those may have been a skim) and what I failed to get out of it is how one comes to attain "Online Search Volume" easily.

You are absolutely right that it isn't necessarily easy to know Online Search Volume. CarStory has a distinct advantage in that our parent company has been aggregating, de-duping and normalizing features in inventory feeds of 50 million VINS daily for the last decade and powering massive marketplaces as well. I'm still new here but really excited to be a part of the team due to the wealth of analytical data at our disposal and the desire to utilize that data to improve dealer operations. OSV is just one metric we are currently exposing. Over the course of the next few months, dealers with our free Market Reports will be exposed to other similar data points.

Market Day’s Supply is calculated by dividing current market quantity of the vehicle at it’s year, make, model, trim level with its exact equipment configuration by the average daily retail sales rate over the past 45 days. In other words, if there’s 10 identically equipped vehicles for sale in the market and over the past 45 days they’ve been selling at 1 a day retail, then you divide 10 by 1 to derive a 10 market day’s supply of that vehicle.

This article is not a knock on vAuto or any of its employees (present or past Ed), not in any way, shape or form. Mr Pollak has revolutionized the way dealers think about inventory and profitability to mirror a changing consumer's needs and ease of access to information. That is to be commended. You won't see me trying to stuff an "either/or" into a situation that requires "and." Let me speak plainly, no dealer should cancel vAuto because of a free OSV report that CarStory is giving to dealers.

That said, this is a great opportunity for "and." One aspect of our analysis is similar vehicles. In an ideal world analyzing a set of year, make, model and trim with an exact feature set would tell you what you need to know. However, in our experience gaps still exist:

1. This approach assumes that the data is correct on the listings. That is a huge assumption at times. Ever try to shop a classified site for a manual transmission, you know, with a clutch pedal? Give that a try and see how many slapshift autos are listed as a manual.

2. This approach doesn’t always account for mileage or the condition of the vehicle.

3. This approach doesn’t account for the competitive vehicle set. This is the biggy. The only thing changing faster than market day supply is the consumer's consideration set. Consider the recent research from Google on the buyer's journey. Here's a link if you haven't seen it. The highlight reel is 900 interactions over 90 days, 71% of which were mobile and 14 brands in the day 1 consideration set... 14 brands!

A consumer’s decision to purchase a vehicle is based on the competitive landscape for that vehicle, not just a unique piece of inventory. Creating what we call "similarity models" to take into account every aspect of the vehicle including the competitive vehicles results in the best signal for analyzing real-time market demand and therefore pricing. Enter OSV.

Where day supply gives important insight into past performance, OSV provides a key glimpse into the future. We need look no further than the glut of off-lease vehicles about to hit markets to see how important that insight will be to the supply/demand/turn/profit equation.

 

[ed.brooks]

I'm really glad this article has sparked some discussion here on Refresh. I reached out to Chad as well as Dr. Franke for some additional insight into the research and OSV metric. Many of you know Chad already, Dr. Franke is the chief data scientist at Vast. His resume is impressive and includes Trilogy and a company that knows a little something about consumer search that you may have heard of called Google. I'm so excited to be a part of the team at CarStory. I'm never in danger of being the smartest guy in the room.



You are absolutely right that it isn't necessarily easy to know Online Search Volume. CarStory has a distinct advantage in that our parent company has been aggregating, de-duping and normalizing features in inventory feeds of 50 million VINS daily for the last decade and powering massive marketplaces as well. I'm still new here but really excited to be a part of the team due to the wealth of analytical data at our disposal and the desire to utilize that data to improve dealer operations. OSV is just one metric we are currently exposing. Over the course of the next few months, dealers with our free Market Reports will be exposed to other similar data points.



This article is not a knock on vAuto or any of its employees (present or past Ed), not in any way, shape or form. Mr Pollak has revolutionized the way dealers think about inventory and profitability to mirror a changing consumer's needs and ease of access to information. That is to be commended. You won't see me trying to stuff an "either/or" into a situation that requires "and." Let me speak plainly, no dealer should cancel vAuto because of a free OSV report that CarStory is giving to dealers.

That said, this is a great opportunity for "and." One aspect of our analysis is similar vehicles. In an ideal world analyzing a set of year, make, model and trim with an exact feature set would tell you what you need to know. However, in our experience gaps still exist:

1. This approach assumes that the data is correct on the listings. That is a huge assumption at times. Ever try to shop a classified site for a manual transmission, you know, with a clutch pedal? Give that a try and see how many slapshift autos are listed as a manual.

2. This approach doesn’t always account for mileage or the condition of the vehicle.

3. This approach doesn’t account for the competitive vehicle set. This is the biggy. The only thing changing faster than market day supply is the consumer's consideration set. Consider the recent research from Google on the buyer's journey. Here's a link if you haven't seen it. The highlight reel is 900 interactions over 90 days, 71% of which were mobile and 14 brands in the day 1 consideration set... 14 brands!

A consumer’s decision to purchase a vehicle is based on the competitive landscape for that vehicle, not just a unique piece of inventory. Creating what we call "similarity models" to take into account every aspect of the vehicle including the competitive vehicles results in the best signal for analyzing real-time market demand and therefore pricing. Enter OSV.

Where day supply gives important insight into past performance, OSV provides a key glimpse into the future. We need look no further than the glut of off-lease vehicles about to hit markets to see how important that insight will be to the supply/demand/turn/profit equation.

Let me start by saying I like and respect both @Ryan Leslie and Chad Bockius a great deal. I've had the opportunity to speak with Dr. David Franke at some length and found him to be a brilliant guy. I don't think this should be looked at as MDS (Market Days Supply) vs OSV (Online Search Volume). I do think there are some gaps in the OSV metric - the largest is it only looks at demand compared to MDS which measures both supply and demand. Only looking at searches can lead one astray.

This is from one local market on one of the largest marketplace sites. The Ford Mustang that Chad mentions in his post shows up prominently on the left - the Search side of the graphic but is no where to be seen on the removed listings, Sold side of the graphic. Is this because a lot of kids search for Mustangs, but don't have the ability to buy one? That would be a good guess. Now look at the Chevy Equinox. It still shows up in the top 20 for searches AND it's in the number one spot for sales - Supply AND Demand. Most of the major marketplace sites offer a report like this free-of-charge.

Market Days Supply for used vehicles has been around since the inception of vAuto. Dale took his system on a huge leap forward by including additional dimensions when he developed the Provisioning system in 2011.

Seven dimensions to be exact - Demand (analogous to OSV - search volume), Interest (clicks to VDPs, a higher level of focus), Volume, Days Supply, Profitability, Availability, Experience.

Taken together, a very well rounded tool for dealers. Not just a single dimension, a real look at the actuality of the market and real time insight into the consumer search process.

There is no love lost between me and Cox Automotive and I am a EX employee of Dale Pollak and vAuto. But I have tremendous respect for Dale and consider vAuto's Provision suite a Best-In-Class system.

 

[WillMcG32]

@Ryan Leslie - Thank you for the response! Great information to see!

@Ed Brooks! Thank you for your insight and appreciate the well written commentary on Provisioning. 7 criteria is always better than 1.

 

[Chris Leslie]

You gotta love it when someone writes a whole blog post about stuff you should be looking at with no clear way to do so. The only thing I can think of is that Carstory is going to probably release some sort of OSV dashboard that is going to cost us dealers money. Shame on you Chad.

 

[LotPop]

I have seen over the years, the dealerships that do a great job at turning their inventory (less then 15% of inventory aged), the cars that end up aging on their lot are cars that have a bad Interest score "C" or less regardless of their Demand score. If I was a betting man, I would throw money that 80-90% of any "Good" dealers aged cars is due to bad Interest and nothing to do with Demand. So to quote Ed Brooks, Demand is how often the make/model is being searched for (on Autotrader) and Interest is what is the conversion % of searches (no matter how they came up in a search, make/model, type, price range) OF CARS WITH THAT YEAR/MAKE/MODEL AND TRIM to being clicked on VDP.
So you want to eliminate age, then stock cars that have a good Interest and not so much search (demand). I dont care what they are looking for I want to by the cars that people are clicking on. You want to minimize your gross, go chase the cars that everyone else is chasing, pay up for them because everyone wants them, and then you need to price them aggressive because everyone has them (that would be the cars on the top of CarStory's list).
@Ryan Leslie said "The highlight reel is 900 interactions over 90 days, 71% of which were mobile and 14 brands in the day 1 consideration set... 14 brands!" So if people are not brand loyal, I would guess there are a lot of budget shoppers and with these shoppers there are sleeper vehicles that dealers could pursue that could do well just due to price point. As an example you can see this 2014 Mitsubishi Lancer ES that I am sure is not on any of the most popular shopping list but if some was to go to Autotrader looking for a sedan for $12000, this car does very well. There are not too many people running to the internet searching for this vehicle but when they stumble across it, it gets clicked (interested is a B), it gets sold (days supply is 32) and its available at the lane for dealers to buy (availability is A-). But a dealer will need to price it at $12000 (not $11,995 like CarStory likes) to get this vehicle the exposure it needs. If customers are not looking for this specific make and model that probably means there are not too many people at the lane to fight over this car too so you can probably buy it at a good price to make some profit.

Only looking at searches can lead one astray.
Market Days Supply for used vehicles has been around since the inception of vAuto. Dale took his system on a huge leap forward by including additional dimensions when he developed the Provisioning system in 2011.

Seven dimensions to be exact - Demand (analogous to OSV - search volume), Interest (clicks to VDPs, a higher level of focus), Volume, Days Supply, Profitability, Availability, Experience.

Taken together, a very well rounded tool for dealers. Not just a single dimension, a real look at the actuality of the market and real time insight into the consumer search process.

There is no love lost between me and Cox Automotive and I am a EX employee of Dale Pollak and vAuto. But I have tremendous respect for Dale and consider vAuto's Provision suite a Best-In-Class system.

 

[ed.brooks]

But can we all agree that relying on Kelley Blue Book (or any other book, for that matter) as the original article on the blog alludes, is silly and shouldn't really be relied on for buying or pricing inventory (other than for loan values)

 

[LotPop]

But can we all agree that relying on Kelley Blue Book (or any other book, for that matter) as the original article on the blog alludes, is silly and shouldn't really be relied on for buying or pricing inventory (other than for loan values)

I think relying on KBB review score could hurt a car and cause it to age though

 

[Chris Leslie]

 

[ryan.leslie]

I do think there are some gaps in the OSV metric - the largest is it only looks at demand compared to MDS which measures both supply and demand. Only looking at searches can lead one astray.

Dr. Franke suggests that it is important to establish a baseline on both to form an educated trendline. It is very logical that a drop in OSV will likely precede a drop in sales, with the exception of the similar vehicle consideration set, more on that in a minute. I think that the overarching point Chad was making is that looking only at historical data can also lead one astray. He used the example of taking a printed book to the lane to illustrate that assumptions based on what happened in the past and gut instinct are not the best inventory acquisition strategy.

You gotta love it when someone writes a whole blog post about stuff you should be looking at with no clear way to do so. The only thing I can think of is that Carstory is going to probably release some sort of OSV dashboard that is going to cost us dealers money. Shame on you Chad.

Don't hate us until we give you a reason to Chris. I'll ask you to take my word for it, but I wouldn't be here if I thought that CarStory had any intention of fleecing dealers. Case in point, we currently provide best in class market reports that sit inline on the VDP (no chance of lead leakage because there is nothing to click) completely FREE of charge. No contracts, no commitments, no DMS access required, no intention to ever charge for them, simply put, no games. We made a company wide decision to offer something of value before asking for anything of value. That model can't work for every vendor, but it does work for CarStory. That was one of the big reasons I chose to put their name on my business card.

@Ryan Leslie said "The highlight reel is 900 interactions over 90 days, 71% of which were mobile and 14 brands in the day 1 consideration set... 14 brands!" So if people are not brand loyal, I would guess there are a lot of budget shoppers and with these shoppers there are sleeper vehicles that dealers could pursue that could do well just due to price point.

Price is absolutely a factor, but I don't think it is the only one to consider in the shopper journey Jasen. There are a bunch of studies that suggest that price isn't even the most important and even ATC suggests that it lags behind the professionalism of the salesperson in their '15 consumer insight study. There is no brand loyalty because the average consumer can't differentiate the brands. Remember this old joke: "What kind of car did you buy? A red one."

I think you are right on to suggest that there is opportunity in similarly equipped inventory. We've been analyzing consumer behavior across 32 million searches each month and have many of the same observations that Google had in the study referenced earlier. In yesteryear the consumer journey started at the brick and mortar and began with the brand. Today, the shopper journey is one of feature discovery. They know what features are of interest to them and work backwards to identify the make, model, trim that has those features. If the referenced shopper's "must have" feature list were represented by 14 brands, then there are 14 brands in the initial consideration set.

But can we all agree that relying on Kelley Blue Book (or any other book, for that matter) as the original article on the blog alludes, is silly and shouldn't really be relied on for buying or pricing inventory (other than for loan values)

Thanks for bringing this full circle @Ed Brooks. I think we can all agree that the ability to adapt to change, technological and consumer behavior based, is the most important factor for success. The discussion around OSV certainly wasn't intended to suggest that it replaces any of the existing tools on the market like vAuto, it is just one more data point to analyze in an ever-changing marketplace.