I realize I didn't really answer "why aren't they widely used"? It somewhat depends on the definition of "widely used"; the worldwide count of Esperanto speakers is, I'd estimate, probably in the tens to hundreds of thousands depending on the level of proficiency being measured. And it was somewhat more than that in its heydey in the early 1900s. But it's true that you almost never find Esperanto on the radio.
Why not? Most people prioritize what appears to be most practical; a chicken-and-egg effect. If 500 million people spoke Esperanto, there would be a lot more incentive to learn it over Spanish, French, Arabic, or Chinese than there is today, when there are perhaps 500,000 who speak it at an A2 or B1 level, and they're spread out all over the planet. Most people would prefer to try to learn a widely spoken language with a home country, even if the chances of success are lower. But as importantly, most school curricula offer those existing national languages; if you want to study Esperanto or another conlang, you pretty much have to to it outside of formal educational channels.
Another factor is that for much of the world, English has become the lingua franca that is being taught everywhere. Even in multilingual nations, you'll learn your native tongue, and English, and a third language of your choice, in many places.
In an alternative timeline where the British Empire was not quite so dominant, and where there was more balance in the relative benefits from learning various national languages, perhaps Esperanto or another constructed language could have been endorsed as a language of trade, and from there become a truly international language.
Finally, there's the aspect of smaller native corpuses. While some may want to study Dothraki just because they love Game of Thrones, national languages have much larger literary vocabularies. Perhaps you love Italian opera, or Japanese manga, or want to read Hugo in the original French. If your have an interest in European history, then even today there are practical benefits to learning Latin, not so much for speaking but for reading historical monuments and documents. No constructed language can compete with these corpuses. If enough momentum were built, that could in theory change. People continued to publish books in Latin for centuries after it was no longer a national language, because that's what could garner the largest audience of book-reading customers. If two billion people spoke Esperanto, I'm sure there would be a healthy stable of Esperanto publishers, including some that only published in Esperanto. But as it is today, it isn't economic to publish your works in Esperanto, and especially only in Esperanto.
Many of these reasons are also applicable for the question of, "why aren't people learning Sorbian?" or other less-mainstream organic languages. In some cases, such as Irish, there's been a resurgence of interest from a community that had largely moved to another language, but for every such case there are many more where the number of speakers is at best stagnant, and often precipitously declining as the younger generation only learns Dutch/German/Chinese instead of Frisian/Sorbian/[insert minority language from China here].