Ads are often the lowest hanging fruit for ROI. Is it often far easier (less time consuming, less resource intensive, less risky) to manipulate human perception at large into buying your product than providing a functionally better product. Techniques are well established and broad reaching across all domains since it focuses on the consumer's perception. There's a reason businesses pour so much money into marketing and advertising.
Even if you do provide a functionally better product/service than competitors, it's often easier for a competitor to convince consumers their product/service is better than your functionally better product/service, if nothing else, by hiding in ambiguity and complexity or well crafted claims that deceive consumers.
There's also the obvious case that if you have an overall better product/service, if people don't know about it, it's not likely to succeed by word-of-mouth alone (there are exceptions) so advertising to aide discovery provides a necessary function but not to the level of manipulative practices we have today (marketing) that are employing sophisticated psychological techniques and are now even driven by targeted behavior data.
It's one thing to inform people "Hey we made this, it exists, it does this, and here's the cost, you can get it here..." but it's another thing entirely to manipulate consumers perception or play on behavioral faults over--I don't know--delivering a product or service of genuine value?
Given modern business environments, it seems to me that it's often financially best to deliver the minimal functional product/service a consumer will accept and convince them its better than it is (manipulate perceived value). That, to me also, doesn't create a strong economy, it creates a system that helps disproportionately redistribute wealth to those who can (and will) play that game.