A) If this is a project developed on request:
1) Hourly rates (development, test phase) X developers + server cost + adm. time
2) Fixed cost. Stipulated estimate of 1). Add some slack for unseen costs/time.
In addition segment the total cost by development, operation and maintenance.
Hourly rate is based on market price + adjustments for how good you are/experience and your market value in general (a well known company will be more expensive than a new established one).
B) If this a volume project for the market:
1) Your net cost (time, marketing, rent etc.) / estimated sold units + provision.
2) Competitive: What do your competitors price their similar product at?
In addition you could choose sub-models such as subscription, upgrades etc.
As always, there are different risks involved, especially with model B.
I suppose that depends upon the type of app, the customer base etc.
Our company develops a app for a small group of customers. They yearly pay a fixed perc of the initial prize to get support. If they have additional wishes, they have to pay additional per hour or fixedprice.
A new (custom)module is prized the same way: they buy it, normally fixed prize, and the support prize per year rises with the same perc.
The prize for the module is based upon investment in resources plus a risk margin and sometimes devided by the number of clients we know that will buy it.
I don't know how TotalCommander does it, but they charge only once, and give updates for free.
They probably have a very big userbase.
You have to determine the pricing model first that you want to do, for example a one time fee or a subscription service. It is a great question but not one anyone else can answer. You have to have a feel for what the market is willing to pay. You have to decide.
There are only 10 types of people in the world, those who understand binary and those who don't.
1.Efforts for implementation
2.If you are implementing as project then price should be more depend on point 1
3.IF you are implementing as product then it depends on your market penetrating strategy which covers point 1 as well as marketing ,support . it also requires consideration in making enhancement in product as per its market type
I'm working in the automation world, programming robots, CNC, PLC... to get machines work.
In my case the time it takes to make an app is one of the most important factors.
If one cpian asks that here I understand he/she is at least a small company, I don't know how many small companies become as big as Microsoft, but probably billing depending the amount of hours won't be a bad idea.
In the other hand, if you are in the mobile app industry then you have to play with the number of apps you plan to sell and find another way to count the money the app should cost...
So it is not easy to give a proper answer without clarifying the sector the OP works for.