• CD

    Sorry, I don’t buy it.

    Anyone out there a statistician? I’ve worked with many and have just enough knowledge to be skeptical.

    The problem I see with the line on the graph is that I could just as easily have averaged it a different way and drawn that line down and right. Just because the data looks like it supports a conclusion doesn’t mean it does. To me the data looks completely random, the dots plotted really don’t mean anything in relation to each other because we don’t know when they were collected and what was turned on and off at that time.

    My bet is that if this graphed showed the two variables plotted independently by day they were recorded that the results would look completely random. That’s just a guess but, I good one based on my experience.

    This could be a really interesting and valuable analysis, but I believe more statistical rigor could be applied.


  • jjmullz

    It’s the gradual budget increase for paid search ads that makes the data ‘trendable’ and gives it order – from what i can gather at least

    It seems to me that the further a point is to the left the less budget assigned to paid search listings (and of course vice versa) so this graph shows that as more budget is assigned to paid search, the CTR rises for paid search (obviously) but also, and crucially, for organic search.


  • jjmullz

    On second thoughts – that is clearly wrong – I am inclined to agree with the first comment.

    Can anyone shed more light on this?