AW Bid Strat Test - Fail?
Ran a little test to pit two types of AW bid strategies against one another.
Conversion Optimizer using CPA bids for one campaign, and replicated that campaign (but geotargeted to a different state in similar population size) and ran the second campaign on a flexible bid strategy, again using Target CPA. They were both targeting a -30% reduction in CPA, but after running for over a week they are overspending and the CPAs have doubled for both on average.
Here's a breakdown when comparing to past:
Conversion Optimizer (Target CPA): +830% in cost; +385% in conversions; +91% in CPA
Flexible Bid Strat (Target CPA): +940% in cost; +389% in conversions; +113% in CPA
Some pains I noticed:
- They have to be applied at the campaign level (AW doesn't support it at the keyword level) which is far from ideal
- Once applied you lose max CPC visibility in AW. Completely. I've reached out to our reps and they say that the ad groups/campaign works towards the target CPA collectively, which makes sense, but I thought that could only be achieved with keyword bids fluctuating according to the algorithm to reach that target. The bids are still the same as when I applied flat bids across the board when initiating the campaign!
(Please note that these are not new campaigns that needs account/campaign/QS history to accrue and historical data is there).
I applied it to two different locations with similar masses to avoid wasted spend, but the doubling in CPA when I was seeking a reduction of a third highlights that bid strats in AW are inadequate.
DoubleClick Search's bid strategies however show incredible results for the CPA targeting feature. And it allows for keyword-level visibility so you can see how the bids change proving that the portfolio really works together to hit that target.
Thoughts? Perhaps I could've done something differently?
Re: AW Bid Strat Test - Fail?
Also, given that there needs to be, IMO, significant conversion data for any such optimizer to work, when you duplicated the campaign, did you amass enough conversion data before starting on this bid strategy test?
Bottom line stats don't tell the whole story by any means and even though you're looking at similar geographies, there are so many factors that come into play when you switch locations that it could easily skew your data.
Do you have more data and info to share?
Did you find any helpful responses or answers to your query? If yes, please mark it as the ‘Best Answer.’