Let's talk about Bid Caps + Portfolio Bid Strategies in Google Ads.
In this article, there are two questions that I will do my best to answer:
- Why would you use bid caps in automated bidding?
- How do you know what bid cap to set?
I’ve been thinking a lot about frameworks and benchmarks related to this tactic of setting bid caps. It's how my brain works. I like analogies and baselines to make sense of why I'm doing something.
First, to be clear on what we’re talking about ~ there is only one way to set a Max CPC limit on your tCPA or tROAS bid strategy and its through a Portfolio Bid Strategy.
Why would you use bid caps in automated bidding?
You do this for one primary reason:
🔹 Eliminate wasted ad spend by controlling how much Google can lever up your bid in a given auction 🔹
My favorite framing of this tactic comes from Sam Tomlinson.
In one of his recent weekly newsletters “Risk-Adjusted Returns, Expected Value & Media Buying”, he challenges media buyers to think probabilistically. When Google levers UP your bid in a given auction, they assume that the given user has a higher expected conversion rate than normal.
That is generally a good thing. The beauty of Google's machine learning is it able to make those user / auction level decisions way better than I can.
But what happens when, as Sam writes, Google decides to make “a riskier bet", levering up your bid 5x - 10x the mean conversion rate?
That might not sit well with you and it might be too risky for it to perform consistently & on-target. Emphasis on the latter.
I don’t have a finance background so I can’t speak as eloquently on this as Sam does *sign-up for his newsletter*, but I think one principle makes it easy for everyone to understand.
Human Guided Automation
Setting limits on your bids takes riskier auctions off the table while still giving you the power of Google’s machine learning. In most cases, I don't want Google to make a bet of 10x my mean conversion rate.
How do you know what bid cap to set?
Well, I've polled a few colleagues for their advice in our space.
Here is what they said:
- Sam Tomlinson expands on his example “Let’s continue with the same example from above: a client with a $1,000 cost per qualified lead target. But, in this case, Google places a $472.00 bid - which implies an expected conversion rate of ~47%. That’s 3+ standard deviations above the mean for this brand. Statistically, there’s a 1-in-1,000 chance that this click will convert at or above that rate.” He continues to write that is an example of Google placing too aggressive of a bet.
- Jenna Chandler writes that she uses "top performing search terms (as opposed to campaign or ad group level cpc) and use the cpc of those terms as a benchmark. my bid cap depends on the brands goals - volume vs efficiency"
- Michael De Boeck recommends to check your search terms report, filter for keywords that are 3-4x above your average CPC and then check the performance vs your target. If it is below target then you have a better idea of where to set your bid cap.
- Miles McNair & Bob Meijer write that “usually, 3-5X your average CPC should be enough to give the algorithm enough room to push, while restricting it from bidding crazy high CPCs”
Most caveat the same thing:
- Check Cost Per Clicks of your campaign, not the account.
- Higher CPCs does not always mean lower conversion… In some cases it might even be higher conversions (👏 Google).
Last note, I know there are some in our world of PPC marketing that say Cost Per Clicks don’t matter. Let the machine do its thing. Set the right targets, don’t worry about it and move onto bigger and better things.
My opinion is that you should look at what the data is telling you. Look at your Search Terms report at the campaign level & don't blindly trust the algorithm. And vice-versa, just because a $50 CPC doesn’t sit well with you (its ok Harrison, breath... breath), it doesn’t mean it’s right to exclude it.
Hopefully this was as helpful for you to read as it was for me to write.
