Home | Wiki | Discord | Dev Stream | YouTube | Archived Forums | Contact

Sales calculations explanation


#1

Hey guys,

I played around with LC 4 open beta a bit and I cannot really understand how the sales calculations work. First I thought I should be able to sell a lot more than I actually do (because my cars are awesome of course), but then I started to question that. This is the case:

  • An insane start in Fruinia, difficulty 10.00, competitors at 107%
  • first off, I have 2 trims, 1 for GT P., 1 for Conv. Lux.
  • after a bit of time, I get some awareness and a market size of 7 and 8 in those two categories.
  • Competitiveness is not great, but also not too bad: 98 in GT P. and 97.1 in Conv. Lux.
  • Affordabilities are 75% and 49%
  • I sell 2 and 1 cars only.

I would expect to sell a little less than 75% in GT P., but I sell only 28.6%.
I would expect to sell a little less than 49% in Conv. Lux., but I only sell 12.7%.

So I tried to understand how it works:
Let’s assume a quadratic competitiveness relation with 0 sales at 50 competititveness. That would give me 0.96^2 = 0.9216. This times 0.75 affordability gives 0.6912. There’s still a huge gap to the 28.6%. So I guess it is not that easy. :slight_smile:

After watching this https://www.youtube.com/watch?v=3_TuC_yg6Bk from 2015 (don’t know how up to date this is) I assume that the competitors take away some of the potential buyers. Currently there are 13 competing trims in both categories, with the best one probably scoring 100 competitiveness. Am I right to assume that the sales are so bad because they are all pretty equally matched?

And another question regarding awareness: Let’s say there is a buyer demographic where my competition has exactly the same car as I have (like the 99 trim problem), does that mean I cannot get over 50% of sales of my (small) awareness group? What if my competitors awareness group is the same small size? What if both groups do not overlap? What do the others not aware of any company buy?

As a summary

  • Could you please point me to information about the current sales model? The Youtube search function is not really helpful, the recent updates are just called “Little Dev Update” :wink:
  • I feel that I am not able to estimate how many cars I can sell only by knowing the number of competitors, competitiveness, affordability and awareness. Is the only way to get a feeling the forecasting tool? It’s kind of hard when you don’t have any awareness in the beginning. Maybe an option to manually set a rough awareness number (per country)?
    Edit: Manually setting awareness number guesses is probably a PITA. Would it be possible to auomatically extrapolate the changes in awareness until project completion based on the current marketing and dealership settings? If you are starting from scratch, you could first set your marketing and dealerships and then start to design the first car.

Thanks a lot in advance!






#2

Hi Martin!

When you think there is something wrong here, you are correct. This is an issue we have investigated several times (last time was for LCV3.7) and were not able to figure out why the sales numbers do not end up with what we expect. Seeing that we weren’t able to easily figure it out and considering that we FINALLY are planning to implement the new sales model that I talked about in those videos many years ago, we have chosen to not address it and focus on other things instead. :slight_smile:

So the way it is supposed to work in this current (old) sales model is as follows:
The market size without factoring in awareness is “your” maximum possible piece of the market. Competitors don’t actually compete for your piece of the pie, they just make your piece of the pie look less tasty so that you can’t get as much awareness (depends strongly on competitiveness above 90) and sales (depends on competitiveness between 50 and 100).

If a demographic has 1000 cars in your piece of the market pie, your company has an awareness of 20%, you make a car that has 85 competitiveness and 67% affordability. Then the calculation is supposed to go like this:

1000 x 0.2 x 0.85 x 0.67 = 113.9 = 113 cars that you would sell that month.

Previously it indeed was quadratic between 50 and 100 competitiveness, which I think is reasonable, but not with the bug that you observed hiding there somewhere… that just made it basically impossible to survive with cars below 100 competitiveness.

Now to your question regarding in the context of the 99 trim problem, or “the new sales model”. If a competitor has the exact same car as you do.

In the new sales model there is a true market, you don’t have a predefined maximum slice of the pie. So in an idealized example where both of you make the same car and otherwise everything is equal too, then you would get the same amount of sales, but not necessarily 50% each. The ones that don’t buy any of the two choices will go either of two ways: 1) into the second hand market, 2) they hold off their purchase until the next month where they again go through the calculation.

What the new sales model does however is something pretty tricky, it calculates brand appeal to that demographic by looking at the entire lineup. So let’s say Ferrari makes a city car, and Fiat makes a city car. Let’s say both brands have equal reputation and prestige… … ok! Their city cars are equally good, too, but Fiat has a lot more cars that could appeal to a city car buying demographic. That is why Fiat will get many more eyes on them for that purchase than Ferrari would. There are more relevant potential buyers filtering in through their entire lineup.

So no, unfortunately I cannot help you better understand why the numbers are the way they are apart from saying they are not what we would like them to be :stuck_out_tongue: maybe we should have another look at why it isn’t working the way we expect, third time is the charm?

I’m looking forward to finally getting the new sales model in, it will be so much more realistic in its behavior… unless of course it has bugs in it that make it not do what I’ve planned since many years ago. :wink:


The stuff I built in 2015…

The awareness blindness in the forecaster is a big problem, yes. We’ll have to address that at some point!


#3

Thanks a lot for the detailed answer! This clears things up in my head quite a bit! Knowing that something is wrong really helps.

It makes sense to keep this simple model as long as you have no option to analyse the market and make the new model transparent to the player. I can’t wait for it though :slight_smile:


#4

@Der_Bayer It looks like we found an issue that could be the source of the difference between the ought and is of our sales calcs. We fixed that now for the upcoming patch: your no longer in production trims were all considered COMPETITOR trims! So your own trims were stealing sales you didn’t get because they are no longer produced. :smiley: Let’s hope that was the issue… seems to be a big one as in a complex example save the sales jumped from 700 M$ to 2.1 B$ after the fix.


#5

Awesome, can’t wait to see how it works out for my game!

(Maybe this behaviour just tries to mimic than another company reverse-engineered and copied your car… :smiley: Good game mechanic though: The better you play, the harder it gets!)


#6

Haha yeah, it wouldn’t matter as much in the case your old generation of cars isn’t better than the best competitor. :stuck_out_tongue: But I think it is best to have this one sorted, makes things a bit more predictable!
Cheers