But also marketing seems like a money black hole to me. I could sink tons of money into marketing and have little to show for it.
More like - if you have an idea for a commercial website or app or something, it's best to start by finding potential customers, building relationships with them, and getting to know what they actually need and are willing to pay for. This is much more important than what language and framework to use, how you'll host it, and other technical decisions. Do the research first to figure out if people are willing to pay for your idea in a form that's in your "runway" budget of what you can afford to build and run before you get money coming in.
It's also perfectly fine to build something without that to try out a new tech you think is interesting, or to scratch a personal itch, etc. Just don't expect to bring in any money with it without doing that research.
It’s like waiting to build the backend until the frontend is 100% complete. While building assumptions change based on things you learn. Marketing for your product works the same way.
By putting the effort into marketing (meaning any form of reaching out: blog posts, direct sales, social media or even paid) you might even avoiding building features that nobody wanted anyway.
> It’s like waiting to build the backend until the frontend is 100% complete
It's more like building a backend, then realising that the data structures could be improved when you're building the front. Marketing at best provides enhancements, not core changes (unless your product itself was incorrect to begin with).
Honestly, if your marketing affects your product as much as you're suggesting, you have a pretty unstable product.
Marketing is about exposing your product to your target customers. It's relatively orthogonal to product features.
By overly internally inconsistent I mean that at some places they will be impossible to fulfill. The code will always breaking at edge cases. The ui will do three things at the same time, will be incoherent and will be complicated for user to understand and learn.
By complicated I mean that you will have to code super complicated hard to maintain structures to fulfill everything at the same time.
Marketing can not be requirements gathering, because marketing is motivated to promise everything to everybody and to exaggerate. I dont use exaggerate here as euphemism for "lie" through that happens too. Exaggerating itself is a problem. Marketing is not motivated to build coherent analysis.
I know because I worked on projects where marketing was driving requirements . It did not led to happy customers, because customers were unhappy over instability and our inability to fill their requirements in a reasonable way.
I feel what you are talking about is a siloed approach where developers build what the marketing team promises to customers. If these teams are the same people you are the developer and marketing team, and thus over-promising things to customers is only shooting yourself in the foot.
Also if you want to hit the ground running marketing can be developing plans for some time ahead of launch.
As for a black hole of money, it can be. Any area is if spent poorly. A failed product is a development money pit.
Marketing should be looked at like an investment, not a cost. Track your activities and if a dollar spent brings back more than a dollar, spend more and see where it plateaus. And vice versa, stop spending or find another channel/marketer. Ive seen developer lead companies looking at marketing as a cost line reject budget increases even with marketing showing a dollar spent was directly returning significantly more. I've also seen money shoveled into the furnace. Marketing is not just good/bad, it's recognising good marketing and bad marketing.
The marketing is critical but I’m just saying there is a cost also