With any experimental program, the question is always “is it replicable and affordable?”
If the program works only on a small scale, it may not be replicable or scalable.
In this discussion of Houston’s Apollo program , we learn that the program started with mass firings and intensive tutoring in math. The designer of the program said it might close the achievement gap in three years, but that seems optimistic. Math scores went up, though not as much in following years, but reading scores did not.
Diane Ravitch's blog 
The flaw in Houston’s Apollo program is assuming that teachers are the problem so all we have to do is fire them.
The challenge doesn’t come from teachers. It comes from the home environment and it often starts before birth while the child is still in the womb. If a child is born in poverty and their environment is dysfunctional, the odds say that child doesn’t arrive at school ready or hunger to learn.
To learn, the child must be willing to cooperate and do the reading, the work, and study. Many who come from poverty and/or severely dysfunctional homes are not ready.
Poverty is a cycle. Without intervention most children born in poverty will continue to live in poverty as adults. “Reign of Error” offers several chapter on how that realistic intervention would work. For sure, iPads and flawed multiple choice tests designed to measure learning will not break the cycle. Schools and teachers with the necessary programs and support may be the only way to break that cycle.
And these poverty breaking programs have a much better chance of working in the public schools because the public schools are held accountable while the private sector isn’t held accountable for anything.
Uh, they spent all of that money to figure out tutoring works? They fired tons of people? When are these foolish programs and gimmicks going to end?????