• Stop being a LURKER - join our dealer community and get involved. Sign up and start a conversation.

Are you Measuring your online Digital Marketing Assist???

Rick Buffkin

Sausage King of Chicago
Oct 29, 2009
730
833
Awards
8
First Name
Rick
Guys and Girls,

I've been thinking about different ways to measure the 3rd party online vendors and classified sites I'm associated with outside of direct referrals, leads from their websites and call tracking phone numbers on their sites. I would like to measure the Online Marketing Assist that these vendors provide me. For those of you who don't understand what I'm talking about, here's an example.

Customer comes into the market for a vehicle (Great). Customer does some research and goes to AT, Cars, ect.. and see's one of my vehicles in the SRP's or VDP's on one or both of these sites. 2,3 or 5 days later or could be the same day, the customer turns to google (or any other search engine or directly types in the url in their address bar) and types in my dealership name or website address. See's my PPC campaign and clicks it or clicks in the organic result. Either way the customer gets to my website. The 3rd party vendor is the online marketing assist here but currently we don't measure the marketing assist the 3rd vendor provided us. We can see reports for SRP's, VDP's, phone and email leads but not the online marketing assist that was provided way up stream. There was a conversion that happened on the 3rd party site. The customer did make their way to the dealership site. It wasn't a direct referral or a phone / email lead. But, there was a conversion. That's what I want to see and measure. If you check any of your analytics / reporting. Chances are your number one search term is a branded term (Dealership Name or Dealership name and make). Part of the reason it is branded terms is because the customer saw your listing on a 3rd party site and searched your branded term to get to your website.

We all think / know these vendors do provide some sort of online marketing assist but we are not currently measuring it (at least I don't know of anyone who is). We are measuring SRP's / VDP's, Direct Referrals, etc... If you talk to any of the big classified vendors, they're going to say alot of the traffic they provide your dealership is walk-in traffic. I would be willing to bet (BTW I'm not a betting man) alot the visitors on those sites do visit the dealers websites that have the vehicles they're interested in prior to visiting that specific dealership.

I want to figure out a way to measure this marketing assist that was provided by the 3rd party. I have some ideas on how to do this.

1) Build a script to check for the 3rd party sites cookies that was placed on the visitors PC.
If the 3rd party site cooperates with this experiment, maybe place vehicle info in the cookie. Look for that data and try to match it to vehicles that were viewed on my website. That data then gets pumped into my GA account.

2)If the customer submits a lead on my website, Place a conversion tracking tag on the /thank-you.htm page that will fire if the cookie is found on the visitors PC from that specific 3rd party site and pump that into GA as well. That way I can measure if the Marketing Assist provided me a lead.

3)Add a piece of JS to change out the call tracking phone numbers if the cookie is found on the visitors PC to a call tracking number we can use to include in the reporting for the marketing assist report in GA and know if the Marketing Assist provided a phone call for me.

I was thinking about including a time frame as well. If the cookie is less than 2 weeks old then do the above process else skip it.


These are just a few ideas that I was thinking about implementing. I would love to hear some of your thoughts on this.

Are any of you guys currently tracking your online marketing assist from your online vendors outside of direct referrals?
 
Last edited:
  • Like
Reactions: 1 person
I lived and breathed "there has to be an introduction before there can be a consummation" as a Field Rep for Cars.com in a past life. I think if you can figure this out with certainty you'll have more than a few interested parties wanting to talk to you. The big two are right that most of their traffic walks in rather than submitting an email or lead form. I'm sure they'd rather stake claim to the sale and are working hard to come up with a tracking methodology that augments the study data we've all seen on walk-ins.

Here are a couple thoughts in no particular order that I hope will help you to work through this:

  • VDP's have meaning. Sure, some people are perpetual shoppers, but I think you have to assume that a unique VDP view represents a real opportunity.
  • This may be dated, it has been a while, but "map prints" were an interesting metric to track when I represented a classified site. That web action definitely precipitated a specific consumer action. Again, it may be dated with the rise of mobile adoption.
  • Cookies on a single device isn't the best answer. As I type this post I have 3 "second screens" running. I often use my desktop and laptop simultaneously, and sometimes pull out my smartphone too. I think you are going to have a huge hole in your analytical data if you can't reconcile multiple device users that could take you to the wrong conclusion. Cars.com's "send to mobile" feature for a VDP is a pretty compelling argument to identify a means to unify the data by user not device. I know that makes it a lot more difficult, but easy data that gets you to the wrong conclusion isn't helpful.
  • Regressive technology might be a good starting point. Have you tried a pen and paper post sale survey with logos as opposed to names? I think your best data source may be the customer. It isn't technically flashy like tracking cookies with scripts, but it might have a better margin of error. I think it is important to include a few controls in the survey like ad sources you aren't using to eliminate bad data.

Good luck wrestling through this and again, I hope something here spurred a new thought for you or encouraged you to stay the current course.
 
Thanks for the replys guys.

@Bill
I've done the DAA with AT for our stores along with the sourcing studys they offer for each of the stores as well. It's good data to see. It was still somewhat vague to me though. You may have diff / better results. For my team, it basically confirmed what we already know. Everyone's an internet shopper these days and you need to be marketing online (to break it down in a nutshell).

@Ryan
I like your suggestion to use logo's instead of names in the post sale survey. I will probably get that implemented into ours.

I'm not sure if I can overcome the issue with multiple devices. That will take a little time to figure out if at all possible. Even with Cars.com's sending to mobile (I haven't checked this yet), I would think they would send the link to the VDP on the cars.com website to pull the customer back to cars.com. The device would still be tagged once the customer opened the link. It could be that I need to look for the tracking pixel and fire events based on that. I don't know yet.
 
Hummmm.... Looks like the content url is encrypted

Name:SSM_UTC_LS
Content:aXA6OjIwNy4xMDYuMjI1LjIyNXx8fGd1aWQ6OjUzZjM4YTEzMTMxMzZkZmQyODkyMWJmNXx8fHJldGFyZ2V0X3RhZ19pZDo6QzBBODAxQ0YwNTkxZDI3RDgzSkdIMUEwOENBMHx8fHByb2R1Y3RfaWQ6OjM3NzI5NTkwNg%3D%3D
Domain:www.autotrader.com
Path:/cars-for-sale
Send for:Any kind of connection
Accessible to script:Yes
Created:Thursday, August 28, 2014 4:12:40 PM
Expires:Friday, August 28, 2015 4:12:40 PM


 
This one stores the makes and models the visitor looked at and if it was a VDP or a SRP. BTW my browser is on the VDP of a Toyota Tacoma.

Name:bn_inventory
Content:%7Btimestamp%3A1409260364165%2Ctype%3A%22VDP%22%2Ctaste%3A%7Bbc%3A%22TRUCKS%22%2Cma%3A%22TOYOTA%22%2Cmo%3A%22TACOMA%22%2Cpr%3A33000%7D%7D%2C%7Btimestamp%3A1408469524096%2Ctype%3A%22VDP%22%2Ctaste%3A%7Bbc%3A%22SEDAN%22%2Cma%3A%22TOYOTA%22%2Cmo%3A%22CAMRY%22%2Cpr%3A33000%7D%7D%2C%7Btimestamp%3A1408469233647%2Ctype%3A%22SRP%22%2Ctaste%3A%7Bbc%3A%22any%22%2Cma%3A%22TOYOTA%22%2Cmi%3A%22any%22%2Cmo%3A%22CAMRY%22%7D%7D
Domain:.autotrader.com
Path:/
Send for:Any kind of connection
Accessible to script:Yes
Created:Thursday, August 28, 2014 4:12:44 PM
Expires:Sunday, August 25, 2024 4:12:44 PM
 
This one stores the makes and models the visitor looked at and if it was a VDP or a SRP. BTW my browser is on the VDP of a Toyota Tacoma.

Name:bn_inventory
Content:%7Btimestamp%3A1409260364165%2Ctype%3A%22VDP%22%2Ctaste%3A%7Bbc%3A%22TRUCKS%22%2Cma%3A%22TOYOTA%22%2Cmo%3A%22TACOMA%22%2Cpr%3A33000%7D%7D%2C%7Btimestamp%3A1408469524096%2Ctype%3A%22VDP%22%2Ctaste%3A%7Bbc%3A%22SEDAN%22%2Cma%3A%22TOYOTA%22%2Cmo%3A%22CAMRY%22%2Cpr%3A33000%7D%7D%2C%7Btimestamp%3A1408469233647%2Ctype%3A%22SRP%22%2Ctaste%3A%7Bbc%3A%22any%22%2Cma%3A%22TOYOTA%22%2Cmi%3A%22any%22%2Cmo%3A%22CAMRY%22%7D%7D
Domain:.autotrader.com
Path:/
Send for:Any kind of connection
Accessible to script:Yes
Created:Thursday, August 28, 2014 4:12:44 PM
Expires:Sunday, August 25, 2024 4:12:44 PM

The problem is you can't read a cookie from any computer but your own. If somehow, they offered a third party tracking service and you could install the tracking code on your site as well, you would be in great shape :) I'd imagine this is something the big classified sites would not want to be a part of, though.
 
Great topic everybody! Obviously ensuring there is ROI for an electronic marketing is key to continued success and from a vendor perspective, ensuring the client keeps coming back. It allows you to see who went where, what they showed interest in, how much they might spend, how hot a lead they are and so on. It gives you an abundance of information before they ever contact the dealership - allowing you to make conversions faster and easier. We build the function through our platform that allows for site tracking and then engagement based on how they interact with emails and site visits.
 
You could always kick it old school with Joe's post sale survey. I tried a few searches to find it, but came up short. With a sale survey you could find out what % of people use which websites, which combination of websites was most commonly used and how long they had been shopping.