As stated in the Official Guide, it is true that the first ten or so questions do give the computer adaptive system an "initial estimate of your ability." However the test continues to adapt throughout.

As I tell my students, there is no way to fool the test. You can make it think that you are deserving of a higher score for a while - by using additional time and effort on early questions - but there is plenty of time for the test to figure out where you actually belong!

Yet, I would always prefer to put a good impression out there and to be starting from a higher estimated score, so don't throw away the first ten either. Instead treat all questions on the test the same.

For quantitative a good way to sort questions is not when they appear on the test but whether you are able to address them. Give yourself about 1 minute to 1:15 on each question to develop a strategy to deal effectively with the problem. If within that time you are not able to see how you might solve the problem then it is unlikely that you would be able to get the correct answer in less than 3 minutes, so just select from the possible answers and move on.

I don't think anyone would recommend a "first ten questions at any cost strategy." It would be foolish to use 5 minutes on each of the questions within the first ten and then end up blindly guessing at questions later in the test.

Thanks David for the piece of info.
In a nut shell, I must give equal importance to all the questions that appear on GMAT-Correct me if I am wrong.