We have all struggled with goosey analytics for years! Nothing every lines up: server stats, Google stats, publisher stats – for the identical time period they each deliver different data! Frustrating is not the word for it. What has become quickly clear is that even though we can measure everything, everyone has a different yard stick.

Never has this been most apparent as on Google Adwords itself. You can’t trust the CPC they project, because it is based on historic data which is impacted by who is bidding and the events in the moment. When testing, a past test is an imperfect test of what will happen in the future because you don’t know who WILL be bidding and how much they will be bidding the moment you actually roll out your campaign.

Now comes Google’s ACE (Adwords Campaign Experiment) that will allow for realtime testing. Allow me a sarcastic moment: in the demo by Google they show how a side by side comparison for a hypothetical company results in spending more per click. Puhlease. Like always Google = Vegas the house always wins. And again, you may be able to AB test in the moment, but future activity will be skewed by future events. It is still imperfect, but it is better.

The opportunity to do direct comparison testing simultaneously is at last a recognition of how lousy/primitive Google analytics truly is. I used to say that it comes in at the right price: FREE but truth be told it’s expensive competitors like Omniture over promise and woefully under deliver. SO free is the right price, but use it with eyes wide open. And that goes for testing ACE now in beta.

I have seen major advertising success on Facebook and equally terrifying failures. I recommend Facebook advertising to my clients ONLY as a test. If it hits, great, if not RUN.  And then there is the pesky question of Facebook provided analytics. We have seen them change up and down for the same time period, leading us to believe that they are completely unreliable.

But beyond businesses buying Facebook ads, there is the question of whether or not to participate in social activity on Facebook: Where is the measurable ROI?  A company can sink time and money and time and money and eventually maybe something happens, but no one is ever really sure. Gee, sounds familiar doesn’t it? We’re back to the good old days of Public Relations!

But in this modern age of being able to “measure everything” it is time for Facebook to cough up some basic information that would allow businesses to make educated judgments as to whether to spend time and money with the Facebook product. Here is my wish list:

1. Percentage of FB users who are actually active at anyone time
2. Percentage of FB users who are actually active: log on at least once a week
3. Levels of engagement: How many users are adding at least 10 new friends a week How many users are posting more than once a week? more than twice a week? daily? multiple times a day?
4. Levels of engagement by demographic: Here is where it gets down and dirty: who is doing what when? what markets are REALLY and TRULY using Facebook.

These are simple questions for that Facebook can easily provide data. Not detailed personal data, but meta data – in the same way that print publishing historically provided circulation data.

This is not too much to ask.