It had long been the case that Google reported a daily cost of $1.40 and an average CPC of $1.35 – and zero daily clicks.
This led to cost & traffic estimates for clients that were way off of our actual performance and recommend budgets that were never optimal. This always made me question Google’s accuracy. However, Google updated their traffic estimate algorithm in May to provide more accurate results. Now, Google reports decimals in their click estimates (e.g., 79.52 estimated daily clicks).
Even though Google claimed they updated their algorithms, I never actually verified their validity. I decided to run traffic estimates using a client’s account and compare them to Google’s estimates to see how they matched. I was actually surprised by the results...
Note that I limited myself to just phrase and exact match keywords since the Traffic Estimator cannot handle modified broad match - it returns the broad match estimates which would inflate our numbers greatly.
These estimates were actually much closer to our actual performance than we’ve seen in the past. This account in particular was estimated with a $6,550 monthly cost and a $0.71 Average CPC originally. We ended up with an Average CPC of $1.94 and $4,990 cost in the first 30 days.
There are, however, a couple items that may have skewed our data and caused the discrepancies seen above:
- Segmented Match Types – Google will return different estimates depending on your keyword list (e.g., a broad match term’s estimates will decrease if you include the phrase match, and so on).
- Unique Keyword Bids - When generating traffic estimates, Google now requires a Max CPC. With real performance, you never have a campaign/ad group wide bid. Well, you shouldn’t… each keyword should have a bid based on its performance and your goals.
What is everyone else’s experience with Google’s traffic and cost estimates? Just take them with a grain of salt? Do you have alternative solutions for creating estimates and recommending budgets? Please share below!