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

Who Loves Them Some Data

Dec 19, 2018
447
407
Awards
6
First Name
Bill
I have an INSANE amount of data going back as far as 2019. This data is all MANUALLY entered, meaning that it's probably about the cleanest, most accurate dealer data out there because I clear up things like leads that were logged as walk-ins that actually started as a phone-up or sales that were repeat customers for the store even if they weren't repeats for the salesperson. Examples would be that I can tell you how many used cars I had in stock that were 61-90 days old on April 19th, 2019, how many sales we had from pure phone ups (excluding repeat, referral, etc.) in June of 2020, and how many active used car deals I had working on November 1st in 2021. I'm wondering if: 1. AI could do some crazy stuff with this data and 2. If there is someone who would be interested in experimenting with this data with me. (a.k.a. someone who is much smarter than me).
 
Last edited:
I'm trying to stay away from shiny objects.
Cuz I am interested in seeing what can be done.
I'm just buried under a huge push for my next release and need to get my site finally published ...

I also need to spend a some time learning how to train the AI to do something like this.

yeah ... I'm interested
 
I'm trying to stay away from shiny objects.
Cuz I am interested in seeing what can be done.
I'm just buried under a huge push for my next release and need to get my site finally published ...

I also need to spend a some time learning how to train the AI to do something like this.

yeah ... I'm interested
Prioritizing your big project is the right call!
 
  • Like
Reactions: Carsten
I might be able to help. I'm an old school retired government computer programmer/developer and OG computer software geek who is just getting back into learning some of the latest coding technologies and AI to keep busy. If you can get me a smaller sampling of your insanely large dataset , I'll see what I can do with it, also let me know what you are trying to accomplish with it. I'm currently a part-time BDC manager from my recliner 1,000 miles away from a used vehicle dealership.
 

Attachments

  • _c36f62ec-5b47-4550-8b08-88c0c3fcd400.jpg
    _c36f62ec-5b47-4550-8b08-88c0c3fcd400.jpg
    210.9 KB · Views: 3
I have an INSANE amount of data going back as far as 2019. This data is all MANUALLY entered, meaning that it's probably about the cleanest, most accurate dealer data out there because I clear up things like leads that were logged as walk-ins that actually started as a phone-up or sales that were repeat customers for the store even if they weren't repeats for the salesperson. Examples would be that I can tell you how many used cars I had in stock that were 61-90 days old on April 19th, 2019, how many sales we had from pure phone ups (excluding repeat, referral, etc.) in June of 2020, and how many active used car deals I had working on November 1st in 2021. I'm wondering if: 1. AI could do some crazy stuff with this data and 2. If there is someone who would be interested in experimenting with this data with me. (a.k.a. someone who is much smarter than me).
This sounds like a field day for me. I haven't messed around with AI data modeling at all but am a whiz with a pivot table and graphs. What sort of insights are you looking to gain from this data?
 
I don't have any specific goals as yet, but I can certainly think of some ideas.

For instance:

I have inventory, lead counts, and sales numbers. How much inventory on average + how many leads = what volume on average for my store?

An algorithm using salespeople's active deals, leads taken, and how many sales to decipher whether they are scrapping deals or selling them.

Demonstrating how aging combined with sales volume affects expected gross.

Correlations between showroom floor closing percentages, leads counts, and inventory on hand.

YOY gross compared to YOY inventory on hand.
 
I don't have any specific goals as yet, but I can certainly think of some ideas.

For instance:

I have inventory, lead counts, and sales numbers. How much inventory on average + how many leads = what volume on average for my store?

An algorithm using salespeople's active deals, leads taken, and how many sales to decipher whether they are scrapping deals or selling them.

Demonstrating how aging combined with sales volume affects expected gross.

Correlations between showroom floor closing percentages, leads counts, and inventory on hand.

YOY gross compared to YOY inventory on hand.
At a minimum I'd love to follow and see where this goes. Most of my focus in-store is on merchandising data, model profitability, individual buyer performance, recon per model, initial margin vs realized margin, etc. Our data isn't clean so I'd love to see what I could do with well groomed inventory data.