To be good at social media, you have to be social -- but you also have to be good at analytics. Here are some of the main issues with analytics that will effect your campaigns.

Are your followers good ones?
Social media promoters find their new potential followers using keywords. If you use the keyword programming to find followers, you find one set of people. If you use the term networking, you find another set. How can you judge the value of the average follower who you attract using particular keywords? Hiring a psychic may not be the worst idea since its not easy to crunch the numbers in any other way. But, if you contact one audience for three months, and then track how much interaction you get, and then contact another audience for another three months, and measure the interaction once more, you might get some numbers. If the average programmer follower got you one click per month after three months of building a network of them -- while the set of (programmers and networkers) combined got you five clicks per individual per month, then spend your time chasing networkers since its clear that the networkers are the ones who are interacting. There could be other factors why the interaction rate went up too, so keep that in mind.

Which topics did well?
There are infinite topics you can post about in your social media campaigns. Its good to post about anything relevant and then take notes to see how popular it was relative to your other topics in a particular month. It makes sense to create more variations on material that was popular, and to post less frequently about what was less popular. If you have a dead network where nothing is popular, maybe its time to terminate that network. The trick is to experiment and experiment and really think about how to create materials that push people's buttons. If your networks are going to grow virally, they have to be hot. So, find out how to be hot on a regular basis. Your first year will be a lot of trial and error. But, if you do your analytics well, and really meditate on what you can do that is optimal, then your second year could be phenominal. Try it and see how you do.

Assigning a value
I keep a list of all of the topics I post about. I assign them all a value and a pecking order to all of them. For Facebook I see how many "likes" and how many responses a particular topic got in a particular month. I see how similar topics ranked in other months. I have to scale the amount of likes / responses to the size of my network at any particular time. Topics could get a score of anywhere from five to sixty, so I put them in order and scale them to how they "would have performed" on an average month. The easiest way to do this is to have three categories: Hot, average, and slow topics. If a particular topic always gets 3x the average number of likes/responses, then its on the high range of hot, while other topics, generally do well, but might only get twenty likes/responses which is on the border of medium in our network. You can create your own valuation system for topics too assigning them any type of value. You can rank topics on a scale of one to ten which is good for keeping everything relative. Judge these topics on how they performed relative to other topics posted during the same month. Anything 3x the average gets a 10, 2x=9, 1.5x=8, 1.3x=7, 1.1x=6, 1.0x=5, etc., This is a good richter scale for posts. The main thing is to analyze why certain posts did well, and how you can create more posts that are equally incredible. Then, and only then can you grow virally.

Methodologies for tracking.
Google analytics is how I track how many clicks various pages get and which keywords resulted in new search engine traffic to my site. There are other providers that can also help you get graphs and data for what content is doing well or not so well. The purpose is to identify critical areas where you are underperforming, so you can work on those areas. Bitly is a system you can use to track clicks. You need to know how many clicks each link to a various topic is getting so you can know which topics are popular. If you never learn what is popular and whats not, then you can never increase your growth rate. If you have to shoot 100 bullets to find out where the target is, to hit that target a second time, you only need one bullet (unless the target moved). If you don't keep analytics, you will wastefully shoot another 100 bullets mostly in the wrong places and alienate your fan base.

Look at which of your pages on your site benefited after your social media campaigns.
See if the keywords on those benefited pages were or weren't used much in your campaigns. You can target specific keywords in your campaigns. Whether its a good idea to target keywords, rather than fan appreciation is up to you, but you can combine both and benefit.