Hacker News new | ask | show | jobs
by ChuckMcM 2691 days ago
I like your description but it has challenges in the implementation when hardware is involved, especially manufactured hardware.

To reason about pricing in a hardware world, I find it helpful to think of dollars as a stream of units coming in for each sale, and a stream of units going out when I am buying parts and assembling things. Then one can start from time 0 (the initial start point), run the math over a period of time, and then at the end compare how much money you have. If we think of the money supply at time zero to be M0, and the money supply at the end to be M1, we can make some initial statements like this:

If M0 > M1 then we have lost money over the time period.

If M0 == M1 then we have neither gained nor lost money (but we may have acquired things that have some value so is isn't necessarily bad)

If M0 < M1 then we have gained money in addition to anything else we may have acquired over that time period.

When building new hardware, there are four kinds of costs we might consider. One is salaries, benefits, and office space for people who are designing that hardware. The next is anything that we have to pay for once but can reuse during the manufacturing process (we will call that our non-recurring expense or NRE). The third is the cost to by the parts that go into one of our widgets. The final cost is the wear and tear on our equipment used to build the widgets, we know that once those things wear out we will have to replace them if we want to keep making widgets!

Now we can analyze the "life cycle" of our product. In phase 1, we look at are first the money we pay out in salaries and NRE to get ready to make widgets. Since there is no money coming in at that time, it all comes out of savings (or the M0 starting value). We will call that the startup cost or $S.

In phase 1, each widget we sell nets us some dollars (revenue yay!). We have to pay for parts (boo hoo), and we incur wear and tear on our equipment and we will label those $C and $D. The time it takes for your equipment to wear out is a guess, it could be 5 years it could be 20 years, but we need to guess what it is. This will be our first guess where we probably don't know the exact value. Remember it is the money we have to replace all the gear (we know that cost) but we have to guess how much the time it will take to wear out. Given our time guess we will divide the cost by the time and that will give us cost per unit time. The other thing we have flexibility on in phase 1 is the price we sell our widgets for, we want to do all the pricing exercises that you mention, but before we do those, we ll will take the cost of parts that go into a widget (call it $C), the time it takes to make a widget, and the price we want to sell a widget for $P. This slightly more complicated math lets us write out a formula for money gained, or lost like this:

   $M = ($P - $C) / time to make - $D.
That is actually in units of dollars per unit time, but we can set the variables and solve for $P to find out at what price we don't lose money while in Phase 1, if we can sell enough units per unit time.

Now we have enough things we can write some interesting reasoning about this. If we sell our units so that they only cover the cost of production. Then at the end we will end up with (M0 - $S) we'll have less money overall because our startup costs are not recovered, but we will have a design that of a widget. If the time period is less than the life time of our factory tools, we will still have a factory that can make widgets.

We can also solve for a price and number of widgets, that is equal to $D (assumming $M is zero you can move $D to the left hand side and see that $P - $C / time has to equal $D. That tells us the minimum rate, of widget manufacture and sales we need to keep up with the cost of wearing out our factory.

Now lets throw a wrench into the mix. Lets say you didn't do any of this, instead you just stuck your M0 money in a fund somewhere making 5% guaranteed compounded annual interest. If you do that, you can come up with a number for M1 (the ending balance of money) based on time.

Now after all of that, you can compare making widgets at a variable price $P and reason about if you make more or less money over a time period $T than you would make just sitting there with the money in an investment fund doing "nothing"[1].

What I like to keep in mind when doing this sort of exercise, either on an idea I have or when someone is telling me about theirs, is to evaluate if there is a price $P where, at the end of a period of time, you have more money than if you just sat on it. That extra money will fund additional ideas, having less than I started with means if I do it long enough I have to go back to work for someone else to trade my time and expertise for more money.

The amount someone is willing to pay for your widget that is greater than the number where you end up with exactly as much money as you would have if you did nothing, that is the "value proposition" for your idea. That is how much "value" it has, over and above just investing in a fund. But, as you recognize, you can just charge anything for a product, you want to charge just a bit less, or have way more perceived value than the competitor's product. If the number you get when you plug that price in to the above equations doesn't leave you with the same or more money than the 'do nothing' choice, then you need to look at the cost equation and figure out if there is some way to lower costs so that it does. On the other hand, if the prices you can charge are easily competitive and you still make more money than the do nothing option? Then you need to figure out how to protect the advantage you discovered so that you can make that money for a period of time before competitors copy you.

And of course it is all different for software only products. But that is a much longer note.

[1] Yes the funds are invested in companies that are doing things but "you" are not doing anything with your money.