This post is part of the Hero Conf Los Angeles Speaker Blog Series. Jay Stampfl will join 50+ PPC experts sharing their paid search and social expertise at the World’s Largest All-PPC Event, April 18-20 in Los Angeles, CA. Like what you read? Find out more about Hero Conf.
Bidding is the most important lever to affect performance. The simplest way to execute is a strategy “bid up if performance is good and down if it is bad” strategy that will get you into trouble in some situations. While there are a great many ways for make mistakes in bidding, I want to talk about one specific, prevalent issue that comes from not understanding the incremental cost of bidding.
Bid up for the good, down for the bad?
Scale v. Efficiency: The Premise
First things first, the most fundamental relationship in paid search is as CPC goes up so does CTR. Since in any given auction CTR is not affected by average position, that is the same thing as saying conversions increase as CPA increases. You can have more conversions but they are going to be more expensive OR you have fewer conversions but they are going to be with a better CPA/ROAS. You have to trade on side of the equation for the other.
What does that mean for me you ask? Ok, let’s take a standard arrangement.
- Client or boss comes to us and says $10 dollars is our goal CPA. I am happy with anything underneath it and sad with everything above it.
- And then we go to our top exact match KW and see a 5 dollar CPA. Hey, right on that is less than $10. Let’s bid that guy up.
- Ok we bid up and the numbers have come in. Cost has gone up as well as conversions. CPA is still less than $10 at $6.50 so what do we do? We bid up again.
- This time CPA ends up at $10 dollars. Whew, nailed it. We stop here. Any higher bids will push our CPA past $10.
Displayed another way…
But that is not is not what is really happening. The CPA for each of those increase in bids are averages so some conversions will be higher than the goal and some will be lower. Let’s take a different view.
From this perspective, we can see the cost of the second and third batch of conversions exceed what we considered a good target.
On the surface, it may look that you are hitting goals but the conversions that you bought by doing this are all outside of your desired goals. We measure this by taking the increase in cost divided by the increase in conversions.
Incremental CPA = Delta Cost/Delta Conversions
If most of your CPAs are around your target goal, then almost certainly a heavy chunk of your spend is buying bad conversions. It is easy to measure incremental cost with Google experiments or through careful tracking of how your efficiency changes with CPCs. Either way, incremental measurement is something that a vast majority of paid search programs are not tracking and therefore represents many opportunities for improvement.