Every tool has its use, but sometimes that use is pretty limited. I don't know if it's a "real problem" that Duolingo focuses on this stuff so much, but honestly, it's probably more about Duolingo's engagement numbers than it is about learning.
Is rote memorization essential to language learning? Absolutely. Should it be the focus of your learning after the very early stages? Probably not. It's just something you do, all the time. Memorize all you like, but true acquisition only comes after you've used your words in real conversation / listening, so that's should probably be where you place your effort. But that doesn't help Duolingo.
In the Japanese learning community, the seduction of this kind of rote, mechanistic, "streak-y" learning has been a trap for a long time -- so much so that there's a popular methodology that has people spend a great deal of time "memorizing kanji". It's appealing because Japanese has a big alphabet of complex characters and the methodology suggests that you can turn learning into a simple memorization exercise with discreet, measurable progress. The problem is, it's pretty much orthogonal to the act of speaking (or reading, or listening to) the language. Learn 2000 kanji and you might gain some ability to infer the meanings of words, but you're still essentially illiterate. The same thing applies at the level of words and sentences -- I have lots of words that I "know" on flashcards, but aren't available to me in actual conversation.
People get absolutely addicted to this stuff, spend years on it while gaining no meaningful level of language skill, then get frustrated and quit. The tragedy is that most people could put in a fraction of that time engaged in listening and speaking with a simple vocabulary, and end up in the kind of virtuous cycle that actually leads somewhere.