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

"Everything Is Fake": Internet Traffic Metrics Are Bullshit

Alexander Lau

Sr. Refresher
Feb 11, 2015
2,189
606
113
First Name
Alex
But, but, but... attribution...!?

"Everything Is Fake": Ex-Reddit CEO Confirms Internet Traffic Metrics Are Bullshit

https://www.zerohedge.com/news/2018-12-27/everything-fake-ex-reddit-ceo-confirms-internet-traffic-metrics-are-bullshit

"It's all true: Everything is fake," tweeted Former Reddit CEO Ellen Pao regarding a Wednesday New York Magazine article which reveals that internet traffic metrics from some of the largest tech companies are overstated or fabricated. In other words; they're bullshit.

Credit to @joe.pistell via LinkedIn share.
 
Last edited:
Reactions: Alex Snyder

Tarry Shebesta

Getting Refreshed
Mar 17, 2011
70
37
18
First Name
Tarry
But, but, but... attribution...!?

"Everything Is Fake": Ex-Reddit CEO Confirms Internet Traffic Metrics Are Bullshit

https://www.zerohedge.com/news/2018-12-27/everything-fake-ex-reddit-ceo-confirms-internet-traffic-metrics-are-bullshit

"It's all true: Everything is fake," tweeted Former Reddit CEO Ellen Pao regarding a Wednesday New York Magazine article which reveals that internet traffic metrics from some of the largest tech companies are overstated or fabricated. In other words; they're bullshit.

Credit to @joe.pistell via LinkedIn share.
Just like Fake News and Fake Products in the auto industry :sssh:
 
Reactions: Alexander Lau

John.H

Getting Refreshed
Nov 12, 2018
79
70
18
First Name
John
But, but, but... attribution...!?

"Everything Is Fake": Ex-Reddit CEO Confirms Internet Traffic Metrics Are Bullshit

https://www.zerohedge.com/news/2018-12-27/everything-fake-ex-reddit-ceo-confirms-internet-traffic-metrics-are-bullshit

"It's all true: Everything is fake," tweeted Former Reddit CEO Ellen Pao regarding a Wednesday New York Magazine article which reveals that internet traffic metrics from some of the largest tech companies are overstated or fabricated. In other words; they're bullshit.

Credit to @joe.pistell via LinkedIn share.
I agree to an extent.

The only metric people should be relying on is SALES. Which can easily be justified by the quantity of acquisitions taken place, and thereon calculated to predict averages on returns.
 

C Dorman

Getting Refreshed
Aug 7, 2018
63
50
18
First Name
Chip
I agree to an extent.

The only metric people should be relying on is SALES. Which can easily be justified by the quantity of acquisitions taken place, and thereon calculated to predict averages on returns.
Yes and No. Sales is the ultimate goal, but the question is how to generate them.

To use a car analogy, let's say you build a car and your goal is top speed. You take it out to the track and find out it's top speed. Now what? How do you make it go faster? There's a lot of variables that go into your car's top speed, which one do you focus on first? What's your budget? What're your time constraints?

Marketing is no different. And just like experienced automotive engineers, experienced automotive marketers will start with a well established model. Engineers work with data and analytics to improve their performance. So do good marketers. For marketing, this takes the form of a regression analysis equation. But we have to use the term model in place of equation because math is scary. With Sales as the dependent variable and all of our marketing and advertising components becoming the independent variables. Looking at the R-Squared measure will then tell us how accurate our model is in explaining variance in Sales outcomes from the mean.

But, to accomplish this takes longer than 30 days... and the odds of anybody at the dealership but you ever looking at, or understanding, your model are slim to none.

This is where experience really comes in handy. You can drill down into the performance metrics of individual components to look for known issues, like overly broad search terms that are too expensive.

An experienced automotive marketer also knows they are dealing with a perishable product. If a car has been sitting on a lot for a long time there's usually a reason. People for the most part aren't stupid, an aging overpriced lot queen is not going to sell until something changes to increase it's value to buyers. Spending more marketing money on it is just going to waste that spend. This is where marketing, sales and inventory managers have to work together. I can tell you from personal experience that adding inventory demand and quality metrics to my analysis greatly improved its accuracy.

So is everything fake and all Internet traffic metrics bullshit?

No.

But you do need to identify the metrics that matter!
 

Alexander Lau

Sr. Refresher
Feb 11, 2015
2,189
606
113
First Name
Alex
^^^^
"Also mobile user counts are fake. No one has figured out how to count logged-out mobile users, as I learned at reddit. Every time someone switches cell towers, it looks like another user and inflates company user metrics."
 
Last edited: