Using Google’s Adwords local targeting features, I feel as if this is never quite accurate and filters out more traffic than necessary. I quite like to separate out high volume area campaigns (cities like London and Birmingham) to monitor the CPCs and traffic and also to identify where most of my conversions come from. However, traffic volumes don’t quite add up to what I had when the campaign targeted all areas:
If all I am doing is excluding London in one campaign and targeting London in another, surely Google defines London the same whether it is included or excluded, then why am I losing clicks?
The same can be seen when gathering traffic information from Google Traffic Estimator tool and using the geo-target advanced options. Just looking at clicks for the keyord ‘weather’, when targeting the UK:
10,099 clicks estimated. Breaking this down to just England:
This suggests 7,884 clicks from England – around 78% of UK volume,
Then targeting all of the UK, except England, this plus the 7.884 of England should end up totalling 10,099 found when targeting the UK. But:
However, 184 + 7,884 = 8068 …short by some 2,031.
Even worse, selecting the 6 parts of the UK separately provides another different result to selecting the UK in a bundle:
Now this is obviously very close to the individual results for England and the other 5 countries. However, why does this not match the ‘UK’ selection? What else can UK possibly include that accounts for 2,000 clicks. This is bizarre and also extremely inaccurate – a reflection on how geo-targeting works within Adwords too. Disappointing.