When I was finishing high school, I went through a bad patch. Nearly dropped out. I fell out with the math teacher at week one & never went to a class.
The school decided to let me have a chance at taking a few versions of the year's final. I took about 5 hrs 1-on-1 tuition with approximately 1.5 hrs preparation each + about 10 hrs of self study + about the normal amount of pre test practice (maybe 20-30) hrs. That adds up to approximately 45 hrs all up compared to maybe 5-6 times that much pursuing the normal study path. I had similar examples at university (though not as sharp).
Assuming that 5 hrs 1-on-1 is probably cheaper the 1/30 X 1 year of teacher time, that's an efficiency. But you could probably have 7-8 hours of 3-on-1 with similar results. 1 teacher could potentially "teach" 10-15 students how to pass that test in a week.
Of course there what you have is a totally different experience. I was taught how to pass the mathematics test, not mathematics. My classmates were taught mathematics, but they still learned how to pass the test.
Universities have terrible inefficiencies. They basically subsidise research with tuition. The students receive some of the prestige gained via research in the form of their degree. But if you look at it from a purely educational perspective, it's inefficient.
In many subject areas attendance in lectures was about 40%. My own attendance was in no way linked to my results. What I did find important were the tutorials. But that was only about 100 hrs per year (10 - 15 students). Everything else, from lectures to course administration is very scalable.
I could have received a very similar education with 10 hrs per student tutorial time if the structure was behind it.
But if I had to place a bet on what the future disruptive innovation would be I'd say it won't be the same thing just more efficient. I think there is a lot of potential in separating education from certification.
The knowledge is accessible online for the most part. It still needs to be arranged. But that will almost certainly occur if certification exists separately to education.
I also think/hope that it will occur in the developing world. Employment is becoming accessible via the web already so the motivation is there. Knowledge is already there. I don't think that a situation where $1000 in hardware & internet access can take someone from a year 9 education to paraprofessional or professional level of education.
There is a huge amount for the world to gain. The biggest economic inefficiencies are barriers to trade, barriers to investment & barriers to immigration. Trade barriers & investment barriers can (& have) been lowered to allow the flow of capital, technology, products & services but not labour. Education is part of that.