Attribution: Do you give the last touchpoint all of the credit for a sale?

[Alexander Lau]

https://www.coxautoinc.com/solutions/nvision
Hmmm.... free Sales Attribution for their dealer "partners"....?
I'm unsure what constitutes "partner", but it sounds like if your website is being hosted by Dealer.com or you're paying for a number of different Cox-based marketing tools you get this for free. Smart, it supports their marketing arm.

 

[Alexander Lau]

Dealers aren't going to want to keep paying for continuous use of a stand-alone attribution tool, once they figure out what works (vendors / services that convert) for them. IMO, that's a big issue and smart dealers know this.

IOW, once a dealer knows who "sold it" for them, why keep paying a monthly fee for attribution, as most offerings are expensive? Multi-touch journey measurement only means so much and chances are dealers will see the same pattern and trends monthly.

The monthly subscription model is flawed. If I'm a dealer, I'd license an attribution tool for 90 days, enough of a sample size to tell me what's working and then dump it. Maybe bring it back a year later, maybe...

Plus, are attribution groups providing actionable insights? Which direction is a dealer supposed to head after analyzing their data? Are dealers responsible for making marketing decisions after they see their reports? What should be their expectation?



I believe this would be a good subject for its own thread.

 

[Alexander Lau]

Automotive solutions are going to be dead in the water. It's being morphed.

https://www.google.com/analytics/attribution/

Even TVA.

 

[Alex Snyder]

Automotive solutions are going to be dead in the water. It's being morphed.

https://www.google.com/analytics/attribution/

Even TVA.

We're going to give this a try on DealerRefresh. I'm adding this to our July todo list.

Thanks @Alexander Lau :thumbup:

 

[charles cokeley]

I read, with a great interest, a post on DMNews.com Marketers' Twisted Cross-Channel Attribution [Infographic].

At one point it was stated "...almost one quarter of marketers (23%) still give the first or last touchpoint all of the credit." I added the emphasis on the word 'still' because the article was published in October 2014. Many verticals have moved beyond this sort of attribution, but my sense is that isn't the case at most dealerships.

View attachment 2964
​

QUESTION ONE: Do you give the last touchpoint (Advertising Source) all of the credit for a sale?

Why is this? Perhaps it is because car dealers are one of the only B2C verticals that have embraced CRM systems. I am a huge fan of CRMs and I can't imagine running modern dealership without one. But there is one glaring shortcoming; In a world where consumers average 24 sources of influence (touchpoints), most CRMs allow for one "source" to be assigned credit.

Knowing this shortcoming exists we can start looking at other methods of evaluating attribution and influence. As I wrote in "We are being beaten by socks. SOCKS!" ...

Let’s start with the fact that they are employing advanced attribution analytics. They understand that many of their customers are ‘multichannel’, using multiple sources to do their research. Because of this fact, the sock guys aren’t just relying on Google Analytics. Google Analytics does a great job on one very specific task – looking the performance of one particular channel out of many used; your website. Now remember the sock guys don’t sell the majority of their socks using ecommerce from their website (just like most car dealers don’t sell most of their cars using ecommerce from their website), the website is just one channel out of many used. They want, no, they need to see the big picture.

QUESTION TWO: Is this a problem?

QUESTION THREE: Is there is a solution?

There is a problem, but there is also a solution. Clarivoys' Multi Touch Sales Attribution with anonymous matching capabilities. We provide a bottom up tool that matches a sold customer to the event trail (click trail) left behind to include those who choose not self identify. Our ability to uncover additional and more accurate Pii data on sold customers provides 100% certainty on our matchbacks.

Our dealers have the ability to view multiple models so they can assign credit based on their business strategies. These models include any touch model, and fractionalized credit models where our Algorithms provide a parabolic and binomial view which assigns credit based on the impact (influence) of said ad channels.

I would love to provide a demo for any dealer or agency who wants to start using an industry leading attribution tool.

 

[Alexander Lau]

There is a problem, but there is also a solution. Clarivoys' Multi Touch Sales Attribution with anonymous matching capabilities. We provide a bottom up tool that matches a sold customer to the event trail (click trail) left behind to include those who choose not self identify. Our ability to uncover additional and more accurate Pii data on sold customers provides 100% certainty on our matchbacks.

Our dealers have the ability to view multiple models so they can assign credit based on their business strategies. These models include any touch model, and fractionalized credit models where our Algorithms provide a parabolic and binomial view which assigns credit based on the impact (influence) of said ad channels.

I would love to provide a demo for any dealer or agency who wants to start using an industry leading attribution tool.

I would argue dealers need an attribution solution, but not past the 90-day mark. A stand-alone attribution tool is a huge waste of dealer dollars past that point, because you're going to see the same trends. If a dealer is unable to surmise which vendors, products or services are selling cars within 90 days, something is wrong. A three month analysis with clear actionable recommendations (should come from the attribution company = beware dealers set your expectations correctly), then decide what to do.

 

[ed.brooks]

I would argue dealers need an attribution solution, but not past the 90-day mark. A stand-alone attribution tool is a huge waste of dealer dollars past that point, because you're going to see the same trends. If a dealer is unable to surmise which vendors, products or services are selling cars within 90 days, something is wrong. A three month analysis with clear actionable recommendations (should come from the attribution company = beware dealers set your expectations correctly), then dump it.

Didn't you work for Clarivoy for a while @Alexander Lau ? If so, does this fit into the 'disgruntled ex-employee' category?
On a related note, 3 months is not enough time to really be confident in any trend. It's more of a snapshot of one moment in time. Being able to observe year-over-year activity might be important. And lastly, in marketing wouldn't you say that what used to work for an advertiser won't always work as well if conditions or competitors change? And in the car business aren't things always changing?

 

[Alexander Lau]

Didn't you work for Clarivoy for a while @Alexander Lau ? If so, does this fit into the 'disgruntled ex-employee' category?
On a related note, 3 months is not enough time to really be confident in any trend. It's more of a snapshot of one moment in time. Being able to observe year-over-year activity might be important. And lastly, in marketing wouldn't you say that what used to work for an advertiser won't always work as well if conditions or competitors change? And in the car business aren't things always changing?

Don't put words into my mouth Ed. I didn't mention a company.

It's not directed at any specific group. It's more of a "don't waste your money on any tool that fails to provide actionable recommendations, especially when the trends are similar each month." Dealers are left clueless in looking at attribution reports when they don't provide true direction and that's what I see as lacking. I've seen it hundreds of times in a number of environments (and typically the attrition / churn rate is astronomical, partly due to the latency in received data from 3rd party vendors = massive delays in delivering reports / IOW, dealers get pissed. WTF am I paying for...?). I can say it doesn't make sense past the 90-day period because I've been knee-deep using a number of attribution tools and there are plenty of out there for dealers to use (you know that) and I see the same issues. The marketing is not THAT dynamic, I've heard the same out of others. For the most part, dealers are using similar products across the board. Once you figure out "who sold it" and it's consistent, keep riding that strategy. Ohhhh and then there's the lack of benchmark data loaded into these types of systems.

Let's go to bat for the dealer, isn't that the point? Expose it for what it's worth. If you don't like that, we can agree to disagree.

 

[Tallcool1]

Didn't you work for Clarivoy for a while @Alexander Lau ? If so, does this fit into the 'disgruntled ex-employee' category?
On a related note, 3 months is not enough time to really be confident in any trend. It's more of a snapshot of one moment in time. Being able to observe year-over-year activity might be important. And lastly, in marketing wouldn't you say that what used to work for an advertiser won't always work as well if conditions or competitors change? And in the car business aren't things always changing?

@Alexander Lau , I agree with @ed.brooks on this one.

I agree with what you are saying with regard to seeing the same trend repeat itself once a 3 month data pool is gathered.

The problem is not what the tool will observe, the problem is that the data WILL change. Competitors will change their pricing model on a weekly basis. The market will shift especially on used. This will need to be an ongoing measurement.

 

[Alexander Lau]

@Alexander Lau , I agree with @ed.brooks on this one.

I agree with what you are saying with regard to seeing the same trend repeat itself once a 3 month data pool is gathered.

The problem is not what the tool will observe, the problem is that the data WILL change. Competitors will change their pricing model on a weekly basis. The market will shift especially on used. This will need to be an ongoing measurement.

That's not what I've seen, but again we can agree to disagree. Typically, dealers are going to see extremely similar trends, it's not that dynamic. So, if we want to get picky, OK. Use a tool for 3 months, dump it, sign back up a year later for three months and compare. Stop dumping vendors (services), look at their results over that period and hold them accountable. The idea is to solidify a digital strategy, not keep hopping around like a frog, every month.

Lack of benchmark data is clearly a massive issue with these types of tools. How does a dealer know what their performance (KPIs) should be? What's the comparison?

Granted, if this is going to be eventually FREE or cheaply priced: https://www.google.com/analytics/attribution. Hey.... why not use it...? You'll be able to add in all the channels that exist, with automotive sources.