Let's imagine - where would the productivity gains come from for fracking?
One can slightly improve the drilling techniques but not by much and any such improvements are as equally applicable to conventional wells as to the shale.
Perhaps the fracking fluid manufacturing can benefit from some economies of scale - but in reality it won't. At least not in the US, where every company seems to like to keep the composition of the fluid a patented commercial secret.
One can aggregate small independents on a large field and save something on improved efficiency of the gathering pipeline network and, maybe, on admin.
There are alternative methods of fracturing being developed - like plasma acoustic resonance but they can also offer only marginal economic improvements.
That's about it, right?
In each well you have to start from scratch - you have to drill down and sideways, then you have to pump the liquid to crack the shale, then you recover the well fluid and separate the oil out of it. If you make 10 wells you multiply the cost x10, if you make 100 wells you multiply the cost x100 - there are no savings with increasing the number of wells.
Also, the nature of fracking is such that its costs per barrel will always be on par with or higher than EOR conventional, which will always be higher than simple conventional. So, if conventional oil were to disappear, the market price can only go up, above the next cheapest alternative.
I think if you want to find a manufacturing analogy in hydrocarbon production, you need to look at synthetic oil. It can be scaled up indefinitely or it can be perfected for an efficient decentralised production. But it depends on the cost of energy - give it a fusion power plant and you can have oil as cheap as water...