...and I get less than 2,500. That's less than 20% of the expected traffic. Does this mean that it's not that accurate with all keywords? Or should the top spot only expect to get 20%, and the results underneath get the rest?
If you haven't already, make sure that the match type is set to Exact. Don't forget that users have a lot of options when they do a search. Even at the top spot you still have the sponsored ads above you. Also, check your description meta tag. If people don't find it interesting they won't click. As for the overall accuracy of the data, I'm not too sure about
Yeh, it is set to exact. but yeh there are sometimes 3 ads above the search results. If anyone else is first for anything that has reasonable searches, what percentage do you get of that?
One of my clients ranks #1 for a keyword with close to 10,000 monthly volume and gets ~6,000 monthly. They also rank #3 for a keyword with close to 30,000 monthly volume, right under two wikipedia articles, and gets ~1,500 monthly. Here are a few things to keep in mind. Is it location based? Does the 7-pack or 3-pack local listings pop up? Where is your traffic coming from? You may rank differently in different countries/regions/etc. Is your meta tag description enticing/accurate? Eye tracking studies prove that people actually read those. Are there any other verticals in the SERP? Eye tracking again shows that videos, images, etc. are more likely to be clicked than regular results.
I actually have a similar concern as I am currently ranked No. 2 or 3 in most large search engines (yahoo, Google, bing) and according to the google keyword tool, the word gets searched 110,000 times yet I am averaging only 7000 hits per month. According to Wordtracker's free keyword tool, the keyword gets searched about 6800 times a month. This leads me to believe one of two things: Either my hits are low because I am not No 1 in search engines, or google's keyword tool is not accurately representing this keyword.
I guess if your keyword are too competitive like "make money online", some of those counts tracked are from people wanted to rank for that set of keywords too. They might be curious to check out how their own site fare in rankings - just my two cents