Having knowledge of your subject matter at the granular level is necessary. Very detailed, fine-tuned distinctions and perceptions. Almost everyone can achieve this, too, given enough time. There's not one of my colleagues who doesn't have this kind of micro-expertise in their chosen subspecialty.
The trick, then, is to come up with significant claims arising from micro-expertise. This is what I've sometimes called "vertical integration," the integration of small details and larger interpretative frameworks. We've all known people who seem to know a lot but don't have much to say. Or people whose larger claims are grandiose, but who lack knowledge at the level of the capillaries. The most common difficulty is with linking up larger claims and the evidence for these claims, especially in an interpretive discipline such as mine, where the relationship between claims and their evidence is not transparently clear or given in advance.
Another concept related to vertical integration is the hermeneutic circle, with its oscillation between a global interpretation and the supporting details. For example, if my initial interpretation of a novel leans in a particular direction, then I will interpret a detail in the novel in a way supportive of that interpretation, but with perhaps a slight modification. Then I will return to other details and interpret those based on my revised concept, and so on. This is not a vicious circle of self-confirmation, since any detail can alter the original conception. As Gadamer argues, the hermeneutic circle is productive and self-correcting rather than merely "vicious."
Applying theory to the text is problematic when the text is not allowed to talk back, to modify the theory.