So, the subject of the OP is doing a successful startup.

On that, three of my thoughts are:

First, nearly all first, wild guesses for a successful startup are doomed to fail.

Why? (A) Nearly no one wants the product/service, say, the results, even for free.

(B) Lots of people like the results, but nearly all of the those people regard the price as too high.

We can call getting past both (A) and (B) as finding product-market fit.

(C) Okay, but, still, say, for a battery with 10 times the currently best in energy in KWh per kg of weight, we don't know how to do that. Similarly for some software that is as intelligent as a human in all respects.

Second, to have some idea that a project won't fail due to (A)-(C) or anything else serious can easily think of, one should have some good plans and, of course, check the plans as carefully as possible.

Then we can formulate a

Saying: If you fail to have a good plan, then you have a good plan to fail.

Lesson One: So, to avoid problems such as (A)-(C) and to improve chances of success, IMHO good planning is important.

Third, startups that are successful enough to make money for venture capitalists (VCs) and their limited partners (LPs -- the people the VCs get the money from) are rare.

How do we know this? A VC firm may look at 1000 unique proposals from entrepreneurs for each proposal they fund, and fund 20 proposals for each one that is very successful and actually makes money for the VCs and their LPs. So, for such a success, that VC firm has looked at 20,000 unique proposals.

Yes, if in some year there are a total of 20,000 unique proposals, there are 200 VCs, each proposal is sent to all the 200 VCs, each VC funds one proposal out of each 1000 they see, then each VC funds 20 proposals and in total there are (20)(200) = 4000 proposals funded of the 20,000. But, again, only one in 20 is successful for only 200 successes out of the 20,000 or 1 in 100.

So, depending on assumptions about the data, the chance of success from a proposal is 1 in 20,000 to 1 in 200.

So, in a word, the desired success is exceptional.

Lesson Two: To have one of the 1 in 20,000 proposals that is successful, instead of just luck, about the best approach is to have some good planning.

Is it possible to have some effective plans? Yes, e.g., there was the first Xerox photocopying machine, the first daisy wheel printer, the first good dot matrix printer, the first good inkjet printer, the first good laser printer, the first good program to drive such printers, the first good spreadsheet program.

Now, let's look at the arguments of the OP:

"But we have done something in the ecosystem to encourage this type of outlandish promotion ... where you feel like you need to use words like trillion."

Why is the planning for a trillion necessarily "outlandish"? We know that we are planning for something exceptional, and maybe good and careful planning, which is the kind we want, says that, really, if we do well and take all the market, then we do get a company worth $1 T. An investor would prefer the planning to be for $1 B or $1 M instead? Okay, if the company really looks like it could be worth $1 T, then it is easy enough to cut down the estimate to something much lower.

"Reality is, that for every thoughtfully articulated and executed world domination master plan, most of the biggest and impactful companies started out with much more humble ambitions. Some just wanted to give students an alternative to a summer job. Others just wanted make their friends feel like pimps."

This situation is likely true but doesn't say much:

Why? The situation says that most successes are from luck. Then the suggestion is to forget about planning and count on luck?

Here is an analogy that explains the situation: Go to a famous golf course and to a par 3 hole. Get the data for the past 10 years on who made a hole in one.

See, first, what fraction of the holes in one were made by (A) professional golfers and (B) everyone else. Will likely observe that nearly all the holes in one were made by (B), not the professional golfers but by everyone else.

How can this be true? Sure, there were only a few pro golfers but many more of everyone else. So, in the end, the luck of the many got more holes in one than the skill of the few.

See, second, what the probability of a hole in one was for (A) the professional golfers and (B) everyone else. Will likely find that the chances for the pros was at least 10 times higher than for everyone else.

So, we have that (i) nearly all the hole in one shots were from luck but (ii) the chances of a hole in one were much better for players with real skills.

So, if you were betting on a hole in one shot, then you should put your money on the pros with real skills.

Similarly for picking startup projects: Go with solid planning and not just with luck.

"Most wouldn’t have cleared the hurdle of the billion dollar idea."

Fine: Discovering that fact is part of evaluating projects. But when do find a project that looks like it should be worth $1 T, don't automatically throw it away as "outlandish".

Sure, maybe on average the $1 T projects take more risk capital than the $1 B or $1 M projects, but guessing here is foolish and not necessary. Instead, the amount of risk capital needed should be part of the planning, the good planning that is believable. As we know from many projects -- long bridges, tall buildings, deep tunnels, big dams -- it is possible to plan big projects with accurate time and cost estimates.

Yes, time and cost estimates can be especially difficult for software projects, but just multiply both by a factor of about 20 to account for work not directly for the project but, say, for getting around bugs in infrastructure software, bad documentation, time to learn new APIs, computer system management Excedrin headache #228,884,454, etc. and might be closer to reality.

"Over the years I’ve watch as that little company has grown from a couple thousand dollars a month to a couple million dollars a month. Next year, that unfunded family run business will do over $100M in revenue."

Terrific. But without good planning, that example represents some astounding good luck, close to winning a lottery ticket. Betting on lottery tickets is foolish for nearly everyone.

"You may be surprised with how little ambition it really takes to eventually change the world."

Luck and lottery tickets are still poor bets.

Instead, we should have good planning. And if the plans point to a company worth $1 T, then check the plans numerous times and ways and then cheer.

Once a father quite successful in business told his son:

"You have a lot of good ideas to invest $1 million and make $1 billion. Why not have an idea to invest $1000 and make $1 million."

Yes, but better still, have an idea to invest $1000 and make $1 B or $1 T.

It appears that here is the main reason for regarding $1 T plans as "outlandish" and discarding them: There are so far no $1 T companies.

So, in effect, this says that we can't plan and practice to make a hole in one and instead should look like the non-pro golfers who made a hole in one with luck So, we should not plan but copy the pattern of the non-pro golfers, copy their brand of cubs, shoes, hat, shirt, etc.

Instead, it really is possible to have an idea for something really new and terrific, to have good plans to achieve it, and to achieve it essentially on time and on budget, e.g., as a low risk project.

Many of the best examples are from the all-time, unique, world-class grand champion of advanced information technology projects, the US DoD. For an example? Sure, GPS.

Betting on luck instead of planning? I remain surprised that people would suggest such a thing.

Or, when the given point does not make good sense, maybe there is a hidden point that does. Or there is the advice "Always look for the hidden agenda". For VCs, one guess at a hidden agenda is publicity for more "deal flow".