THANK YOU for creating and sharing.

The most common one I see is the raise to first page bid for brand terms. Or to raise/lower bids based upon conversion rates or cost per conversions (but you can use the conversion optimizer for some of this as well without lots of rules). However, I think most people use the alerts instead of the rules.

Alert me when conversions drop, budget is being used, etc. I see alerts much more often than rules. The alert is essentially an action item to investigate and change after you look at the data.

If Google gave better rules for CPA or ROI/ROAS bidding, then I think there would be more adoption.

I use pretty similar judgements when I am carrying out an audit prior to perhaps taking over an existing account with a view to improving performance.

One thing I would be keen to know from your members is what are their favorite Automated Rules? Which do they think work best?

I want to shed some light to disagree about CTR being everything — sure, with Google’s yield management algorithm in mind, yes — CTR must still be primary because they need to optimize their own profitability on click costs.

However, LP quality now does make an impact BEYOND just meeting the “no penalty” threshold.

The algorithm change to AdWords Quality Score on 10/3 this year made a visible impact on QS for many different accounts at our agency. We’ve particularly noticed a high QS delta % in certain keyword groupings that did not have, for lack of a better term, a “frequent presence” of that keyword on the landing page.

It was almost an accidental discovery as I was using ACE (Adwords Campaign Experiments) to test landing pages, but instead of using 1 keyword to 2 ads (with different landing page URLs), I created experimental AD GROUPS — which means the same keyword and ad combinations were running against each other in experimental rotation.

ACE results showed, to my dismay, no significant difference in the KPI I was measuring (conversion rate) — but interestingly I saw variance in position and CPC. Strange, I thought to myself — especially with brand keywords that are spending over $10,000/day and had low variance in an account we’ve gained historical QS over 5 years now.

Well, when I compared the same keyword pairings, I noticed a pattern where I had QS improvements in the “new” experimental copy of the keyword, especially where in certain head brand terms.

I looked at my control landing page vs. my experiment and when I did a simple search on those brand terms “CTRL+F” — I noticed my brand didn’t show up but once on my control (there were tons of logo pictures, but not in text).

The new page had at least 10 text occurrences of my brand.

This resulted in nearly 15-20% media efficiency in terms of targeting the same per unit media cost (CPC) at the same position.

I wonder if others are seeing the same.

Best,
David Chung
@classdc