GMAT Strategy

probably the most effective single bit of advice i can give you would be a redux of advice that i've previously given about how to study problems.
most students simply study the problem at hand, with little or no thought as to how the lessons from that problem might be extended to other problems. that's a very bad idea: after all, the problem at hand really doesn't matter in the big picture (since you're not going to see that specific problem on test day). all that matters is lessons that you can extend to other problems.

with that in mind, here's a redux of some advice that i gave in another post:

no matter what you're studying, make sure that you study with the appropriate DEPTH, not just BREADTH. this is going to be difficult at first - there won't be that many connections to make when you're just digging back into the material after 'retirement' from academic study - but you should try to RELATE problems to each other in order to gain the maximum benefit from them.

for EVERY problem:
* don't concentrate on the solution to that actual problem, since you can be sure you aren't going to see that actual problem on the exam
* instead, try to find TAKEAWAYS from the problem, which you can then APPLY TO OTHER PROBLEMS. this is key - DO NOT LEAVE A PROBLEM until you have extracted at least one piece of information, whether a formula, a strategy, a trick/trap, etc., that you can apply to OTHER problems.
do not leave a problem until you can fill in the following sentence, meaningfully and nontrivially:
"if i see _____ ON ANOTHER PROBLEM, i should _____"
* notice the SIGNALS in the problem that dictate which strategy to use. if you miss the problem, then notice the strategy that's used in the book's solution (not always the best solution, in the case of the o.g., but better than nothing), and go back to see if there are any signals 'telling' you to use that strategy.

(the post is here:
https://www.beatthegmat.com/need-tips-fr ... tml#107595)

--

some specific points related to the areas you've mentioned:

* DIVISIBILITY & PRIMES is is traditionally one of the most difficult areas for gmat students, especially students whose backgrounds are in engineering, accounting, or other rigorous and rules-based fields.
essentially, the reasoning required for divisibility and primes problems is close to 0% "engineering thinking" (memorized rules, formulas, and protocols) and close to 100% "mathematician thinking" (creative thinking, formal logical deduction, and proof-style reasoning).
if you are a student whose usual mode of studying is to memorize lots and lots and lots and lots of rules, then you are going to find these problems extremely frustrating at higher levels.

* a lot of people on this board make PROBABILITY AND COMBINATORICS more difficult than it needs to be, by insisting on applying the combinatorial formulas ("c" and "p") to every single problem, whether they are really appropriate to the problem or not.
for instance, see this problem:
https://www.beatthegmat.com/deck-of-card ... tml#122220
noted very simple solution in the post to which i've linked (a solution that DOESN'T use the formulas -- it just multiplies together some consecutive probabilities), compared to the obnoxious difficulty of the solution that uses C and P formulas.
you'll find that this is the case with most probability problems: they are generally much harder with the C and P formulas than without those formulas. (this is not always the case with combinatorics problems, however. you will occasionally see a combinatorics problem that can be directly solved by the application of a C or P formula.)

(continued below)

(continued)

* for OVERLAPPING SETS, you should be okay if you can get into the rhythm of using a diagramming device (such as the "double set matrix" in our manhattan gmat guides).
if you find that problems involving three overlapping sets are unusually difficult, then you are probably safe paying less attention to those problems. while not totally absent from the test, they are extremely rare; i find that they are vastly overrepresented in practice materials.

* GEOMETRY is a vast field, but one tip you would do well to keep in mind is to remember that ESTIMATION works on a lot of geometry problems (problem-solving, not data sufficiency).
in other words, if you have a decent diagram and can guess the approximate magnitude of the answer to a geometry problem, you will often be able to reduce the answers to one or two choices even if you have absolutely no idea how to solve the problem!

Received a PM asking me to respond. (Sorry I'm just doing so now! I've been on vacation since the day you sent your request.)

Ron gives some great advice above. I'm also wondering whether part of your problem might be due to timing. Are you spending your time appropriately during the quant section? (All questions below apply specifically to QUANT questions.)

How often do you go over 2.5m on a question? How about over 3m? What's your performance on questions on which you spend less than 1.5m?

A time of 2.5 to 3m on a single question should be fairly rare, and a time over 3m should be nonexistent. Any question that would take you longer than 3m to answer is (a) likely to be wrong anyway, and (b) likely to cost you another question later in the test.

Any question that's answered in under 1.5m should be in one of two categories:
(1) something you knew you weren't going to get, so you pulled the plug and made a guess, or
(2) correct

If it's wrong but you thought you should have gotten it right, then you made some kind of mistake that you might not have made if you hadn't been rushing ("rushing" = spending less than 75% of the time you're allowed to spend). In other words, this is what your 2.5m+ problems cost you.

So, what's the differential? For a lot of people, the 2.5m+ questions aren't significantly upping the "percentage correct" levels, and that's before you even add in the "too fast and wrong" questions. If this sounds like you (and go check your last couple of practice tests to see), then you're misallocating your time and that's preventing you from raising your score. If so, let us know and we can talk about what to do to fix the problem.

Also, see this article for more about how to analyze your practice tests:
https://www.beatthegmat.com/a/2009/09/23 ... tice-tests