Trifurcate Twitter into three distinct parts:
* The social--what users actually use it for. * The ad platform. * The realtime gestalt/search platform.
The latter is what is inadequately funded and considered.
Twitter has a latency advantage on the pulse of thought across the user base. The user base is already high-value for this relative to the morass of Facebook users. This is the key differentiator. This is what you base the platform and the value of search upon. You don't bury it by restricting the API, or trying to shift it into the realm of the social platform with "Moments"--you build dependency upon it, and monetize that unique value by those who need to keep the pulse. This creates value for the other two parts. It is also woefully underfunded.
Someone mentioned an AWS-like model. That is thinking along the right track. Make the gestalt a utility that people will pay for.
Yes, you still have all the spam/hate issues on the social platform, and those need to be solved. But those only become a corporate imperative if Twitter is resigned to merely being a social network rather than a platform.
Maybe it's actually 4 parts--I forgot about the live event deals (e.g., NFL games). There's a raw, traditional media part, but I'd align that with the advertising part.