Manufacturing, trying to better understand salary expense

[bdubbs]

So after my last thread about manufacturing revealed to me that I never considered salary costs in my games and how that affects your overall costs. Determining whether it would be more cost efficient is tricky to figure out for two main reasons, the first is because it's difficult to factor in salary costs and the second is because it's difficult to determine what your true freight cost will be without knowing how much production and sales volume you can expect.

I'm at least going to try and help simplify the salary issue by giving you some nice round numbers you can use to estimate your salary costs in the future. I came up with these numbers by starting a game, skipping forward until march (I always lose the first day of january due to not being fast enough, and I didn't want to use 29 day february) and on March 1 opening two small factories making beds with 5 manufacturing units, 2 purchasing, and 2 sales with no resources connected. I did this in Baltimore with a wage rate of $75 and these were my results.

$78083 salary expense for 1 month. Divide by total # of employees (52) is 1501.59 that each. Divide again to represent what it should be at $10 wage rate (1501.59 / 7.5) is $200.21 paid per employee per month.

Max training also gives us a monthly training expense of 39041. Divide again by 52, then once more by 7.5 suggests that training costs are $100.10 per employee, per month, per $10 of wage.

To test this I start a new game and follow the exact same procedure with the only change being a new city. I picked Moscow for it's nice even wage rate of 55.

57376 Total Salary expense divided by 52 employees divided by 5.5 (adjusts total as if it was a $10 wage) is $200.61 per employee per month per $10 of wage. Checks out

Training of 28689 divided by 52 divided by 5.5 comes out to $100.31 of training expense per employee per $10 of wage, Checks out again.

I'll test this on retail units at some point to see if those are the same but I'd imagine they are. It's actually kind of funny when you think about it. This game literally has capitalism in the name but everyone gets paid the same amount of money as everyone else in every city

As far as determining what your freight costs would be is much more tedious. If you go to a retail store in your city and check the freight costs to bring in products from the seaport you can get a pretty reasonable idea of what your freight costs will be if you were to build your factory next to the seaport. Then you just take the total # of units you expect to sell in a month and multiply it by that freight cost and that would be your number to compare to.

It gets much more complicated with more cities as city A could have modestly cheaper wage rate than City B but City B would offer modestly less freight costs and it becomes unclear which is preferable. Now you're doing all these different calculations for different cities and it can be quite a pain.

Personally if there's 1 city with a significantly cheaper wage rate that's the one I'm going to manufacture in. A large factory should eventually be able to serve multiple cities and even if it's not always the most efficient to import from that city it can help you consolidate your manufacturing. Ideally you'd like to manufacture any inputs you need in the same place you manufacture your products to reduce the overall freight cost of your supply chain. Most of the time I expect the freight saved on inputs plus the salary and training expenses saved to outweigh any freight expense you incur from freight to your retailers.

A situation where you might want to deviate from that plan is if you have a city with a high wage rate that is also very large. A good example is Tokyo which from memory has a population over 4 million and a wage rate over 80. If you're selling something like foods, snacks, or beverages you could possibly manufacture products and inputs locally the overall sales volume may be enough to bring more value than importing, but I don't have any maths to back that up at all.

[eleaza]

Yes, base wage is the same for every employee, regardless their jobs. And the base is $2,000 exactly per worker, and training cost is $1,000 per worker per month.

You don't need to divide it further with $10 again, otherwise the unit of salary would not be the same as the unit of money, (you simply use Real Wage Rate divided with 100 to get the multiplication factor, RWR 80, the factor is 75/100=0.75, hence its $1500 per worker per month. The fractions you see in your figure is due to RWR has two more precision after the integer, like 75.08)

[eleaza]

And indeed freight cost is a little more complicated, but you can find out the exact number in <CapLab game installed folder>/MOD_kit/Data/Product_Types.DBF. The field of FREIGHT (index) range from 1 to 100, means the freight cost per unit compare to the "standard price", defined in the same file as the field of "PRICE".

Then the more complicated part is there are two parts of the freight cost, the base freight cost (regardless of distance), and the second freight charges per mile. And the freight cost is further divided when it's cross cities, with in-city truck freight cost, and long-distance city-to-city shipping freight cost. Also the in-city trucking cost is determined by the source to commercial seaport distance (unless it's CES DLC, where there might not be seaports in a city). Hence to calculate the exact cost is fairly difficult unless the exact setup and map layouts are known.

Generally speaking, cross-city production are only suited for certain low freight cost index products (or product that have a very extreme price concern you can deviate from standard price you can manipulate, like very high brand concern, low price concern, and high branding will give you 3 to 4 times the standard price, and freight wouldn't matter). Also a general simple rule is below 40 RWR cities are worth the long distance and retail strategy should be high profit margin high price, not mass production and cheap price (but the sales volume need to cross certain threshold, which means there's a range where cross-city product is favorable from profit standpoint, and sometimes other concerns needs to be addressed as well).

This is one of my analysis about whether cars should be manufactured in different cities or concentrated.
http://www.capitalismlab.com/forum/view ... 655#p16884

[bdubbs]

@eleaza that wage trick is magical, I had no idea it was that easy to figure out wages.

I checked out the thread about importing / exporting cars and motorcycles and it was an interesting read but I'm still having a difficult time figuring out where I should build my retailers / factories to help improve efficiency.

For example I started a game last night where I had a city with a 27 RWR and tried exporting furniture from there but over time I just wasn't convinced it was optimal. The more I play the game the more I realize I need to learn.

So what you're saying is that the smaller / lighter a product would be and thus realistically easier to ship in large quantities the less of an impact freight has, which makes sense and explains why products from the same seaport can have a different freight burden. It's also now unsurprising that furniture had a significantly larger freight burden.

Trying to understand better with a little math, I just rolled some perfect maps, multiple big cities with Lanzhou at 27 RWR and Lagos at a 31 RWR. Just to test this out I had already given myself substantial starting money, so I opened a large factory in Lanzhou, another in lagos, and built a furniture store in Washington D.C.

To put this to the extreme I maxed the number of employees and am producing beds, 5 manufacturing units, 2 purchasing, 2 sales. Total 172 employees per factory.

Cost to produce in Lanzhou - Default cost to Produce Beds: $95.32 Freight cost to Washington D.C.: $86.35 Monthly Operation Cost: $50,636 Wage + Full Training cost: (516,000 base x .27) thank you so much eleaza = $139,320

Total Monthly Fixed Costs: $189,956

Cost to produce in Lagos - Default Sales price of Beds: $196 Freight cost to Washington D.C.: $63.22 Monthly Operation Cost: $56,038 Wage + Training cost: $159,960

Total Monthly Fixed Costs: $215,998

Cost to Produce in D.C. - Freight cost $5.35 Monthly Operation Cost: $131,685 Wage + Training Cost: 376,680

Total Monthly Fixed Costs: $508,365

Assuming they can all produce beds for the same overall cost (not possible, and something you need to consider but it's a variable I want to omit. In this game it's actually $7 per bed cheaper to produce in D.C. than the cheap wage rate cities because it is closer to the timber seaport, making it even more attractive) Then the only variable left is going to be the quantity to determine which factory is the best deal.

Break Even Point D.C. and Lanzhou - (508,365-189,956) / |(5.35-86.35)| = 3931

So anything below 3931 units per month production is more favorable in Lanzhou and anything above that should be more favorable in the D.C. factory.

Break Even Point of D.C. and Lagos - (508,365-215,998) / |(5.35 - 63.22)| = 5052

D.C. currently buys 13,000 beds / month because it is very large and therefore it would likely make more sense to produce beds locally. However a smaller city like Essen only buys 4000 beds per month, and it would make more sense to produce in a cheap wage city and import.

Now for an important question, is there any way I can figure out what the production volume of a factory? If I want to produce beds locally in D.C. but don't plan to export elsewhere then the one thing I don't want to do is build a large factory when a medium factory will do after it trains up to max.

From what I'm seeing my expectation is that unless you're working with smaller products that more often than not it will make more sense to produce locally if you're in a city of 3 million +

[eleaza]

The production volume is not a fixed number, since each product has different units, so they varied over different products. What's important and decide how many production lines required is the "demand volume vs product volume ratio" per citizen(million) of particular product. And products can be modded, also the overall demand of a city can change depend on the product concen, current average overall rating, brand, economic status, it is nearly impossible to calculate, and there's always a risk where the future may not be as you expected, and this is why warehouses matter. Using warehouse to provide buffers of products, and constantly making the upstream factory working at peak utilization, and depend on the warehouse to determine the downstream retail store supply/demand balance, as well as using the growing or shrinking stockpile in storage unit as a meter to determine whether to add more production line. When you face excessive production, just sell more, increase the downstream by either expand to new market, or rise the overall rating for more demand quantity. Generally speaking it's better stay in a constant slightly overproduction to make the buffer in warehouse storage at full, hence later it's easy to tell if the demand has changed, and you will have time to take actions for the fluctuation.

It's not the exact ratio matters, but whether how quickly and how well can you adapt to the changes and handle the fluctuation. And as I said before there's always a sweet production range where setting up shops locally is better or worse than concentrating in one location (a lot of time concentrating is more about flexibility and management purpose than pure profit concern)

As to furniture product per say, I think they are almost always better stay local production, even local timber, since both timber freight cost and finished furniture freight cost are high. And if you really have to sell them across cities, put the source furniture factories, as well as target retail stores as close to the commercial seaport as possible. Their in-city trucking cost is very very high (automobile actually has very low freight cost index, just 10, but furniture is like 40. The problem of automobile production cost isn't about freight, but rather the long production chain and high fixed running cost. On the contrary furniture is about high quality and good locations). And normally it's very difficult to supply enough timber from upstream source, even if you owned the timber yard. They run out very quickly when you have monopoly in multiple cities.

[Diche_Bach]

I get the impression the developer is a keen follower of the "iterate and test with user base" philosophy of software development. A philosophy which, based on my apprentice's level of experience in the field, makes a helluva lot of sense, especially for game software . . .

Anyway, if that is true, then perhaps we can hope that: in future, the management-worker dynamics are a bit more fleshed out? Treating all workers are as equivalent comrades in a make believe ecology where a single Union dominates EVERYTHING and must also be led by Buddhist Monks (because they never show any greed or demand for more! ), i.e., holding wages constant across: all individuals, all roles, all neighborhoods, all firms, all cities, all everything, is FINE given the game does such a great job of modeling so many other processes.

But a "Labor Dynamics" DLC would sure be neat!

Pie in the Sky dreaming here: imagine if every citizen in a city was represented by an array in an array

Citizens().IDnumber().Lastname[].Firstname[].Age[].Sex[].().Firstname().Age().Sex().EmpStatus(dummy variable with maybe 5 or six values 0= too young, 1=slacker, 2=newseeking, 3=employed, 4=out of work, 5=disabled, 6=retired, 7=deceased).Experience(another ordinal variable that somehow relates to all the job roles in the game), etc.

Lots of interesting things could be done . . .

[eleaza]

I get the impression the developer is a keen follower of the "iterate and test with user base" philosophy of software development. A philosophy which, based on my apprentice's level of experience in the field, makes a helluva lot of sense, especially for game software . . .

Anyway, if that is true, then perhaps we can hope that: in future, the management-worker dynamics are a bit more fleshed out? Treating all workers are as equivalent comrades in a make believe ecology where a single Union dominates EVERYTHING and must also be led by Buddhist Monks (because they never show any greed or demand for more! ), i.e., holding wages constant across: all individuals, all roles, all neighborhoods, all firms, all cities, all everything, is FINE given the game does such a great job of modeling so many other processes.

But a "Labor Dynamics" DLC would sure be neat!

Pie in the Sky dreaming here: imagine if every citizen in a city was represented by an array in an array

Citizens().IDnumber().Lastname[].Firstname[].Age[].Sex[].().Firstname().Age().Sex().EmpStatus(dummy variable with maybe 5 or six values 0= too young, 1=slacker, 2=newseeking, 3=employed, 4=out of work, 5=disabled, 6=retired, 7=deceased).Experience(another ordinal variable that somehow relates to all the job roles in the game), etc.

Lots of interesting things could be done . . .

In a sense, most "successful" economic/business simulation games, are based on "entity level" agents, like a house, a vehicle, etc., instead of individual. It's no different for Capitalism as it's based on firms/buildings agents (right below the intended simulating level of a corporation/city, except AI persons). Detailed simulation usually is just based on agent attributes, rather than "self-acting" entities (like workers in the factory is just one variable).

From a game design point of view, simulating too much details doesn't make much sense, since they probably won't add too much differences compared to averaging models. However we could still make each agent a little more complex, and have more fields to register pseudo "individuals" with some procedurally generated information associated to them and make the illusion of details reaching further down the level. A lot of city-building games I believe use this kind of trick, so don't end up using too much computational resources and too laggy.

Although, I agree it is interesting to simulated at individual level, and see superstructure several levels above emerged naturally. But those simulations are usually less "GUI" friendly since most are academic researches than as gameplays.

[bdubbs]

I agree with the complexity already built into this game that trying to individualize workers in some way, or create a variable salary based on some kind of skill or experience just seems like it would be the kind of thing that would be difficult to implement smoothly. The fact that the RWR changes over time as well as other macroeconomic factors that change the landscape of cities is plenty for me. What I was saying about the one size fits all wage was just something I thought was funny given the ideological contradiction to the name of the game.

I actually have an interesting game going right now. I hit 100 expertise in Beverages but I had AI starting capital, expertise, and aggressiveness maxed out. 7 cities, with 3 or 4 of them being over 3 million population and 50 RWR. Local competitors quality is maxed, seaport quality is minimized, and there is only 1 consumer good seaport per city.

I started the game as I start most games, retailing seaport goods in the best city I can find, in this game London. I went with a Discount Mega to start and realized that because of my in game settings retailing this way was not going to give me the strongest return. I opened a medium farm to sell frozen pork, beef, and lamb which was enough to give me some modest profits but it was clear that this wasn't going to be a big money venture until my farm trained up.

I proceeded to open up some retailers where I could, opened another farm in a big city and managed to get myself to about $40m in profits over a few years. Any seaport good worth selling had demand outpacing supply tremendously so I moved into manufacturing bottled milk by opening a large factory in Lagos which had only modest freight costs to London, D.C., and Pittsburgh. I started to R&D bottled milk and was able to churn some small profits even with the low quality and strong local competition.

I ended up running into trouble over time because of R&D expenses which were ultimately wasted as I couldn't come close to keeping a tech lead in any of my products. I ended up really putting myself against the wall financially and gave up.

If I had to approach it again, which I will since I saved the beginning of that game I would probably retail farm goods again but I'm not planning to R&D my own products since the competition will happily sell to me if there are multiple players in the market. Instead I'm thinking of committing to retailing and farming for several years and focusing my R&D on semi products like glass and corn syrup and products that the AI might happen to ignore. If I can make enough money to support a few idle farms and warehouses I can start producing milk, corn, grapes, cocoa, and other products where quality improvements cannot be researched. This should give me the ability to dominate the food / snacks / beverage arena which are highly dependent on input quality to determine overall quality.

Seem like a plan that could work?

[Diche_Bach]

I get the impression the developer is a keen follower of the "iterate and test with user base" philosophy of software development. A philosophy which, based on my apprentice's level of experience in the field, makes a helluva lot of sense, especially for game software . . .

Anyway, if that is true, then perhaps we can hope that: in future, the management-worker dynamics are a bit more fleshed out? Treating all workers are as equivalent comrades in a make believe ecology where a single Union dominates EVERYTHING and must also be led by Buddhist Monks (because they never show any greed or demand for more! ), i.e., holding wages constant across: all individuals, all roles, all neighborhoods, all firms, all cities, all everything, is FINE given the game does such a great job of modeling so many other processes.

But a "Labor Dynamics" DLC would sure be neat!

Pie in the Sky dreaming here: imagine if every citizen in a city was represented by an array in an array

Citizens().IDnumber().Lastname[].Firstname[].Age[].Sex[].().Firstname().Age().Sex().EmpStatus(dummy variable with maybe 5 or six values 0= too young, 1=slacker, 2=newseeking, 3=employed, 4=out of work, 5=disabled, 6=retired, 7=deceased).Experience(another ordinal variable that somehow relates to all the job roles in the game), etc.

Lots of interesting things could be done . . .

In a sense, most "successful" economic/business simulation games, are based on "entity level" agents, like a house, a vehicle, etc., instead of individual. It's no different for Capitalism as it's based on firms/buildings agents (right below the intended simulating level of a corporation/city, except AI persons). Detailed simulation usually is just based on agent attributes, rather than "self-acting" entities (like workers in the factory is just one variable).

From a game design point of view, simulating too much details doesn't make much sense, since they probably won't add too much differences compared to averaging models. However we could still make each agent a little more complex, and have more fields to register pseudo "individuals" with some procedurally generated information associated to them and make the illusion of details reaching further down the level. A lot of city-building games I believe use this kind of trick, so don't end up using too much computational resources and too laggy.

Although, I agree it is interesting to simulated at individual level, and see superstructure several levels above emerged naturally. But those simulations are usually less "GUI" friendly since most are academic researches than as gameplays.

The game that I am working on as an apprentice developer (details of which, I cannot disclose because of NDA and publisher/developer haven't announced anything publicly about an "update/expansion") is essentially 21-year old C code. Somehow the developers who've continued to expand on the engine and build new games off the engine, have kept it compatible with all the OS that intervene between Win 10 and MS-DOS! Jedi powers that I still merely aspire to . . .

Suffice to say: the game itself is "not unlike" the Capitalism family: lotsa quantitative tables, a look-and-feel that is a bit "old school," no vector graphics or collision (factors that seem to weigh down performance AND development and drive up cost of many contemporary games). One might term such games: interactive dynamic relational-database (well . . . spread sheet anyway) engines with UI designed to be approachable enough that an audience with an interest in the subject matter will 'endure' it in order to learn to play and thus get caught up in it . . . (though I think Cap family has better UI than the one I work on).

What I have been utterly amazed by in getting into the guts of this game is: the ASTONISHINGLY large quantities of data which the game is packaging into each and every save file, the astronomical number of operations ("millions of die rolls per turn" in the words of my mentor) being performed in the mere seconds that it takes for a turn to process and the _literally global_ scale of the war elements being represented, with attributes that go all the way down to the level of individual vessels (ships, planes, land craft) and even smaller in many cases! It is certainly true that most of the "people" (in this case only military personnel) are being abstracted into aggregate units (smallest being platoons I believe, though it might actually be companies), but in some cases ("Leaders") even INDIVIDUAL human beings are being represented in this game.

The game runs like a top. It even ran like a top on older OS with much less RAM and slower chips than typical modern rigs.

I think that proper data structure utilization can go a LONG way toward facilitating massive simulations, though I readily admit that the labor of setting it all up is onerous and it is a question mark as to how much "value" it adds to gameplay for any given game.

Anyway, just some ideas that crossed my mind!

Before I die (hopefully many decades from now) I want to make a simulation/strategy game that takes things nearly down to the molecular level!!

[Diche_Bach]

I agree with the complexity already built into this game that trying to individualize workers in some way, or create a variable salary based on some kind of skill or experience just seems like it would be the kind of thing that would be difficult to implement smoothly. The fact that the RWR changes over time as well as other macroeconomic factors that change the landscape of cities is plenty for me. What I was saying about the one size fits all wage was just something I thought was funny given the ideological contradiction to the name of the game.

I actually have an interesting game going right now. I hit 100 expertise in Beverages but I had AI starting capital, expertise, and aggressiveness maxed out. 7 cities, with 3 or 4 of them being over 3 million population and 50 RWR. Local competitors quality is maxed, seaport quality is minimized, and there is only 1 consumer good seaport per city.

I started the game as I start most games, retailing seaport goods in the best city I can find, in this game London. I went with a Discount Mega to start and realized that because of my in game settings retailing this way was not going to give me the strongest return. I opened a medium farm to sell frozen pork, beef, and lamb which was enough to give me some modest profits but it was clear that this wasn't going to be a big money venture until my farm trained up.

I proceeded to open up some retailers where I could, opened another farm in a big city and managed to get myself to about $40m in profits over a few years. Any seaport good worth selling had demand outpacing supply tremendously so I moved into manufacturing bottled milk by opening a large factory in Lagos which had only modest freight costs to London, D.C., and Pittsburgh. I started to R&D bottled milk and was able to churn some small profits even with the low quality and strong local competition.

I ended up running into trouble over time because of R&D expenses which were ultimately wasted as I couldn't come close to keeping a tech lead in any of my products. I ended up really putting myself against the wall financially and gave up.

If I had to approach it again, which I will since I saved the beginning of that game I would probably retail farm goods again but I'm not planning to R&D my own products since the competition will happily sell to me if there are multiple players in the market. Instead I'm thinking of committing to retailing and farming for several years and focusing my R&D on semi products like glass and corn syrup and products that the AI might happen to ignore. If I can make enough money to support a few idle farms and warehouses I can start producing milk, corn, grapes, cocoa, and other products where quality improvements cannot be researched. This should give me the ability to dominate the food / snacks / beverage arena which are highly dependent on input quality to determine overall quality.

Seem like a plan that could work?

It sounds like you're setting up a fairly harsh playing field for yourself, which is cool . . .

So far, I've mostly enjoyed custom games: main goal $10 billion personal wealth, no industries/products precluded. 15 to 20 computer opponents, all with slightly better startup funds and some competitive/aggressiveness. 7 cities, 2 or 3 each of the seaports . . .

In the little bit that I've played the game, the strategy that I've found to be the most generalizable to get at least 10 years or so down the line and into the "~300 million net worth / ~40m per annum operating profit" ball park (which I've never got past because of me not knowing exactly how the game works and some of it not exactly working 'intutively' and me screwing things up . . . (e.g., killing myself financially with too much R&D, hiring stupidly expensive executive staff whose contribution to bottom line is dubious at best, expanding into endeavors that didn't take advantage of my existing enterprises):

1. Pause the game.

2. Buy up at enough of the companies stock from public shareholders that I am at ~55%, but ideally even more . . . even up to 75% (allows for lots of "stock issuance fund raising" that doesn't endanger my majority sharehold status later).

3. Go to my HQ and make sure my salary is at least low if not zero, maybe setup an HR department or a finance department depending on how I'm feeling.

4. Find a city with at least two preferably three consumer goods that can be sold in a convenience store. Nothing else really seems to matter about the city.

5. Set up a convenience store with at least 1 and up to 3 each of purchase-sell dyad, plus one advertise node which services all three sales nodes (two blank nodes that are not adjacent = no problemo; no point in overstaffing eh?). Set training to a reasonable modicum. Set advertising to be in the ~$25,000 / month ballpark.

Cigarettes, frozen lamb, and cakes (I think) were what I made the most money on in my most recent (now abandoned) playthrough. I figured the cigarettes would be obscenely profitable and lead to even larger profits down the road once I could build farms/factories (which was completely true: customers were thronging to buy them at ~$11/pack and once I got a farm, logging mill, warehouse and factory setup in the HQ city they were costing me like . . . 10 cents per pack . . . obscene yes [my real life Mom died of emphysema and lung cancer about 10 years ago, so yes I do appreciate the . . . paradoxicality of it . . . but hey, I'm an apprentice game developer who works on war games so I don't let it bother me too much . . .].

6. As soon as possible set up a second then third convenience store selling the exact same products . . .

Had I stuck with that theme until I my company was a Gibraltar, "winning" (getting to my $10 B personal wealth goal) probably would've been a foregone conclusion. But given I'm still just playing with all the dials and knobs and seeing how it works, I naturally started diversifying and expanding, and in some cases I think this hurt me more than it helped.

7. As soon as I had the funds, I built additional retail outlets to take advantage of all the other consumer goods in my home HQ city (or at least the ones that didn't face serious local competition from computer-opponent firms that had emerged in the interim).

By this point, I'm up to about 10 years, net worth in the hundreds of millions, company with several hundred employees, etc. At some point along around 6-7 I expanded my convenience store chain to pretty much every other city in the game (7 of them) and that probably was profitable.

8. in HQ city: logging camp, farm warehouse factory to be able to supply my own cigarettes, frozen lamb and (long-term) cakes. Farms are incredibly expensive though, so this phase of diversification really ate up a lot of funds and required considerably borrowing and stock issuance. By this point, the computer-opponents are not even trying to compete with me in my three core products (cigarettes, lamb, and cakes) . . . well in a few cities there are a few outlets that sell some of those products, but they are not really competing. Even companies with brands that are better than mine are not really managing to take much of my market share, simply because my prices are so relatively low.

9. Hired the execs and killed my profitability by doing this all too quickly along with other expenditures of one sort or another . . . at which point I got a bit peeved and decided to bag it.

Having dabbled at a few of the other 'starting themes' that are promoted in other scenarios, I think this is an almost hands down viable strategy unless the scenario design just rules it out categorically:

I. cheap retail goods (at least one ideally 3) if you have a choice, pick products with the following attributes:
1. low mass (e.g., cigarettes, or cakes, NOT washing machines!)
2. small number of raw material inputs (try to only go with products that use two, not three or more) Initially this is largely irrelevant but it will facilitate expanding your endeavours to the production chain later.
3. products which inherently have a 'friendly' cost to price ratio (which generally will mean they use 1st order "raw" inputs that are not particularly "rare:" paper or tobacco = not rare; CPUs or engines = RARE!)
4. obviously if you can get them all from the local seaports that is preferable, but not essential.
5. Consider adding the additional node to "private label" your products. If I understand properly, the real benefits of branding and advertising are not realized unless you do this to your imported goods and I would assume that: down the road when you change suppliers to your own farms (which will presumably provide product of at least as good if not better quality) that the brand you "automatically" apply to your own supplied products will be THE SAME brand as the one you've been selling by private label already . . .
II. keep costs low, generate funds
III. expand using the exact same template
IV. repeat steps II & III until you are a juggernaut
V. keeping focused on your existing product line, and securing control of the production chain for your products all the way back to the dirt (from whence all consumer goods arise!)
VI. DO NOT hire anyone except employees. Executive staff do not seem to add much and they are very costly (I saw the script to half their costs, but still not convinced they are "worth it" in games current form)
VII. Avoid head-to-head competition, go for low hanging fruit, until such time as there IS NO low-hanging fruit.
VIII. Do not worry about your stock price, it seems largely irrelevant. Funds to expand are more important than having a stock value that is "competitive."
IX. Be judicious but not timid about borrowing.