Why Quality Score is One of the Least Important AdWords Metrics
This article by Howie Jacobson was originally posted on Search Engine Watch.
Recent discussion among search marketers over whether lowering a keyword bid can lower quality score reminded me of a totally different topic: the way pharmaceutical companies sell us unnecessary and harmful drugs that keep us from getting healthy naturally.
Drug companies make a lot of money by inventing diseases so they can sell drugs to people who would otherwise be considered healthy.
For example, “high cholesterol” is now defined as a disease, rather than a risk factor associated with heart disease. Everyone with total cholesterol over 200 is encouraged by drug companies and doctors to take statin drugs to lower that number – despite the fact that lowering cholesterol artificially through drugs doesn’t seem to have any impact on heart disease or longevity. In other words, both your cholesterol level and your likelihood of developing heart disease depend on your lifestyle: diet, exercise, sleep, attitude, environment, and so on.
Yet the marketing scare tactics have us all running around trying to manipulate our cholesterol through drugs with nasty side effects, rather than addressing the root causes of our declining health.
Quality Scores are Like Cholesterol Numbers
Similarly, AdWords advertisers tend to get paranoid about the quality scores of their keywords. And AdWords educators and consultants encourage this obsession by writing myriad blog posts deconstructing Google’s ad rank algorithm. Doing the math, they point out that the higher your quality score, the less you have to pay per click to outrank your competition.
The logic seems airtight: if your quality score is 7, you should do everything you can to get it to a 10. And I was guilty of this until I got schooled in the subject by my colleague Joel McDonald. Quality scores, he avers, are to CPC what cholesterol is to heart health: A single number related to the end goal, but not a cause of it.
“Contrary to popular belief, you are better off turning the entire quality score column off and ignoring it entirely,” according to McDonald. “Advertisers spend way too much time and energy trying one trick after another to increase their quality score in the hopes of reducing cost per click. Yet quality score is one of the least important AdWords metrics.
“Pay attention to the other more important metrics that matter, and you’ll see a far better ROI than you could ever see by finding that one little trick that will get you a better quality score.”
Below are a couple of screenshots of extremely competitive and profitable keywords from an account that McDonald manages. A few things to note about these screenshots:
- The keywords in each screenshot are identical; same word, same match type.
- The only difference is the campaign: who he is advertising that keyword to, and what he is willing to pay for that specifically targeted audience.
- Note that despite being the same keyword, quality scores vary from 4 to 10.
- Quality scores as high as 8 or even 10 don’t save any money compared to the same keyword with a quality score of 4.
- The Avg CPC on the QS10 version of the keyword is 66 percent of Max CPC, and the QS 4 bid is only 3 percent of max CPC. If high quality score really meant lower bids, you’d expect the opposite effect. But here you can see that the “worse” quality score keyword require bids that are a tiny fraction of what I’ve agreed to pay.
- For the QS4 keyword in the first screenshot whose Avg CPC is $1.78, getting that quality score up to 10 would not lead to any benefit. The cost per click wouldn’t drop, and the average position wouldn’t improve (you can’t get higher than 1).
If quality score is so important, McDonald should be paying less for keywords with quality scores of 10, and more for keywords with quality scores of 4. In the above examples, the exact opposite is true.
What’s going on here? To understand this seeming paradox, we need to remember a couple of quality score facts.
Quality Score Facts
What’s going on here? To understand this seeming paradox, we need to remember a few quality score facts:
- Google uses quality score not to tease or torture advertisers, but to maximize its profit on every single search. When you understand this underlying motivation, you realize that little tricks to improve quality scores aren’t going to work, at least not for long. If you want stellar account quality, create campaigns that make Google money today and don’t endanger their business tomorrow.
- The biggest contributing factor to quality score, by far, is click-through rate. Which stands to reason: more clicky, more money for Google. (The one exception is a quality score of 2 or less, which reflects a landing page or website that Google finds objectionable, and a risk to its business model. I want to make it clear that we’re not talking about that kind of poor quality score in this discussion.)
- Keyword quality score is graded on a curve, relative to other advertisers bidding on the same keyword. That’s why the same word in the above screenshots can vary so widely; it’s like McDonald playing ice hockey in his kid’s junior club league or getting on the ice in the NHL. Same player, different competition.
- Quality score affects the minimum bid to have our ads show up. This is perhaps most important (but hardest to wrap our heads around). If you’re running an AdWords account obsessing over how little you can pay for a click, then I hope you’re competing against our clients.
Bid for Maximum Profit, Not Maximum ROI
You see, if a keyword becomes profitable, you don’t get extra points for an amazing margin and ROI. Your bank balance doesn’t care about that. Rather, you want as much of that profitable traffic as you can get, to maximize total profits.
(Again, an exception: once you reach capacity, you do want margin rather than volume. But I haven’t come across too many advertisers in that situation.)
When McDonald took over management of this account, the client didn’t understand this principle. They refused to pay more than $3 per click for any keyword, no matter what the return.
After discovering the right keywords, optimizing ad copy for those keywords, and pairing them with the right audiences at the right time, McDonald was able to raise their bids by as much as 10 times, which generated 10 to 20 times the traffic. And even at up to $30 per click in very specific cases, bring in 150 percent to 300 percent ROI.
Although McDonald has no objection to reducing what they’re paying per click whenever possible, he’s not going to waste his time trying to “trick” Google into thinking the quality score is higher than it is.
Now, if you’re going for penny bids, then by all means worry about quality score. But for everyone with a serious business, please realize that just because you can get away with nickel clicks if you can score that perfect 10 quality score, there’s no reason to do so. You might want to brag about your 500 percent ROI, but your traffic will be anemic.
If you forget about cheap clicks and quality score, and “overbid” (at least according to the Google algorithm) by paying one whole dollar, you’ll probably get 1,000 to 10,000 times the traffic. True, your ROI will “plummet” to maybe 150 percent, but you’ll still bring in a lot more revenue than one or two conversions per month at 500 percent or even 1,000 percent ROI.
Let’s say your quality score on that keyword drops from 10 to 4; calamity, right? After all, Google now insists that you pay much more for a click – perhaps 50 cents instead of 5. Oh wait; you don’t care: You’re already bidding a dollar, because that’s the most profitable bid you can make. The floor doesn’t matter when you’re already floating way above it.
The Right Way to Improve Quality Score
If you have high cholesterol, you probably want to do something about it. And that something, if you’re smart, involves improving your lifestyle. Eating better. Exercising. And so on.
Similarly, if you suffer from low quality score, you want to do something about that as well. You want to improve the most important things about your AdWords account:
- Write compelling ads, that connect tightly to keywords, make the right promises and stand out from the rest of the search results page.
- Continually test and improve your ad copy so your ad is more likely to earn money from Google than your competitors.
- Segment your audiences so you can adjust messaging to connect with different groups’ mindsets and language.
And after that, if you’re profitable, forget all about quality scores above 2. Track conversion diligently so you can bid based on performance instead of “how little you can get away with.” The right two dollar keywords will be a lot more profitable than the wrong nickel keywords, every time.
The more traffic you generate, the more you can segment your audiences and adjust bids based on the differential profitability of each segment’s conversions. McDonald suggests segmenting and assessing different performance levels for as many of these as possible:
- Search network computers, mobile phones, and tablets.
- Google search vs search network partners (if partners performs worse than Google).
- Display network computers, mobile phones, and tablets.
- All of the above sliced and diced by geography (country, state, region, time zone, or whatever makes sense for your business).
Even if you only have one keyword, you can easily bid on it a dozen different ways, as you saw from McDonald’s client screenshots. You’ll quickly be surprised to find the drastic difference in quality score for that one keyword. Don’t be surprised, however, that quality score doesn’t correlate at all with profitability.