What is a Quality Score in search engine ad campaigns, and how do I get a good one?
The Google Ads and Microsoft Advertising platforms use quality scores as a key element at the core of how paid search engine ad campaigns work.
In a search campaign, the ad platform assigns quality scores to each keyword in the campaign. Quality scores impact rank at ad auction, cost-per-click (CPC). The QS is only displayed in real-time, so they can’t be charted historically without resorting to custom or 3rd party software solutions.
The inner workings of quality score are a bit opaque by design and can seem arbitrary at times. As a result, quality scores have for 20 years constantly been the subject of much interest, discussion, and intrigue. Sometimes obsession, and borderline conspiracy theories!
Campaign owners and managers who manage to attain higher quality scores in their search campaigns pay less for clicks, rank better at ad auction, and accordingly are favored for placement onto search engine results pages (SERPs). They win.
Quality Scores in search engine campaigns are primarily associated with keywords. But they are not within the direct control of advertisers. QS provides a mechanism to balance the best interests of the three parties impacted by an ad campaign: the advertiser, the person viewing the ad, and the ad platform. In successful search campaigns, these three interests are in alignment.
How do Quality Scores translate to rank at ad auction, and click cost?
They are measured on a scale of 1 to 10, with 10 being the best QS attainable. A QS of 1 means the keyword is essentially unusable, and click costs will be relatively high. A click cost of 10 brings greater success at the ad auction, lower click costs. The higher the quality scores are relative to competitors at the ad auctions, the greater chance a campaign has of becoming successful.
Rank at ad auction is equal to your bid times quality score. So if an advertiser bids a keyword $2 with a QS=6, their rank at auction is 12. If a second advertiser bids $3.90 with a quality score of 3, their rank is 11.7. The first advertiser wins the auction with the $2 bid! Their clicks for the same ad position cost less.
For good reason, the advertising platforms keep many details of how QS is calculated and assigned under wraps, but here are the basics:
QS is a measure of relevancy
There are three interests to satisfy in PPC advertising, and QS provides a mechanism to balance those interests:
- The advertiser.
- The consumer viewing the ad.
- The ad platform.
In order for all three interests to be successful, ads much be ‘relevant’ to the expressed search intent contained in a search query. The person searching must find the information, service, or product they are looking for after they click on an ad. This leads to success for everyone, and ads that run for a long time with all 3 parties benefiting. Quality scores are designed to encourage relevant ads and landing pages.
Thus a good ad should address and contain the keyword that has put the ad on the SERP page, and the landing page which a click on the ad takes the consumer to must be relevant.
What determines keyword Quality Scores in PPC search ad campaigns?
There are 4 factors that contribute to quality score:
- Ad content – relevancy to keyword and landing page.
- Click-through rate (CTR) versus the platform’s “expected” CTR.
- Landing Page User Experience – UX and conversion rate/efficiency (CRO).
- Advertising and budget consistency.
1. Ad Content
The search platform assesses the semantic relationship between the keyword and the ad content. They must be related in meaning – semantically related.
Since each ad resides in an ad group container (object), this means that each keyword in an ad group must also be semantically related to the other keywords in the ad group. So this tells us how to build good ad groups. All the keywords should have a common meaning. And all should have a strong semantic relationship to the copy in the ad.
We go into more detail on this in other articles where we describe how we create ad group structure by grouping and factoring keywords.
The take-away here is: structure ad groups so that all the keywords mean essentially the same thing, and the ad copy reflects that.
Today, there is no need to entertain “single keyword ad groups” (SKAGs) if you follow this best practice.
2. Click-through rate (CTR)
The CTR for a keyword is a measure of consumer interest in the ad copy. So it is a measure of the intensity of interest in the ad. CTR’s can range from a fraction of a percent to 100%. In a typical search campaign, CTR’s of 1% to 5% are the most common, in most vertical markets.
CTR is the strongest component in Quality Score, and it’s simple in concept: for a given keyword match, the associated ad must draw more clicks than the average ad, or better.
For any keyword being served, the ad platform has a lot of data about average click through rates, which it terms an “expected” CTR for any individual keyword. So if a keyword has an above-expected CTR, it is much more likely to receive an above-average QS. And there reverse is true.
CTR is approximately 50% of the quality score calculation. The bottom line is, must have strong ad copy to win this battle. It also helps to utilize semantically related keywords within an ad group, and pause or replace those that pull poor QS.
3. User experience (UX)
The “user experience” on a landing page associated with an ad group is the least opaque of the three major determinants of quality score. In the early days of search, it was enough to be sure that the landing page had the keyword and semantically related words on it.
This led to a lot of “keyword stuffing”, dynamic text replacement on landing pages, and other ways of artificially gaming the quality score. So the scoring mechanisms have become much more sophisticated.
To assess the UX, the ad platform spiders the landing page and assesses the relevancy. It also looks at what is happening when paid traffic lands on the page. Is it staying on the page, or bouncing? Is the dwell time short, or long? Does the consumer click links on the page? Convert?
In our experience, today the platforms also follow (crawl) all links on the page to check what else the landing page connects to. This ensures that the entire site the landing page is part of, is also related to the keyword, and the ad copy. This means that stand-along landing pages with few links are now at a disadvantage.
User experience scores are the most difficult to control, but if your landing page and site are authentic, it should not be an issue. When we see keywords with poor quality scores due to UX, it is often because the landing page is light on text content. Other mistakes include not tagging images with alt text, keyword stuffing (keyword density over 5%, sometimes less), links to low-quality domains, and other SEO on-page mistakes.
4. Advertising and budget consistency
The ad platforms reward “the steady advertiser”. As domain ownership life impacts the domain authority of websites, account age impacts PPC Ads quality scores. The platforms are unlikely to allow a brand new account to barge into the auction market of established advertisers (accounts) without some time passing.
Consistency is also important. If you pause your campaigns, change the spend dramatically, frequently change ad copy, or are in other ways erratic — then you are unlikely to gain and hold good quality scores. If you want to reduce advertising during Holidays, or seasonally it is better to reduce ad spend by 30%, 50%, or even 70% versus switching the campaign off. If a campaign sets idle for some time, then it will have to rebuild it’s quality scores which can often take 3 to 6 weeks or longer. We saw the impact of this dramatically with advertisers that overreacted during the COVID crisis Q1-Q2 2020.
This not only concerns ad spend and account life, but also ad copy, and ad extension copy changes. When advertisers change ad copy then if possible, the platforms recommend keeping the existing ads (especially if they have good CTR) running in the first few days following the change. This is because new creatives will contribute to the overall keyword Quality Score in proportion to the percentage of time they are being served.
This will blunt the QS loss which can result from changing ads, as existing ads have a longer serving history with much higher impression volumes. So it’s better to cycle new ads in, rather than replace directly when you can.
Two flies in the Quality Score ointment
The biggest flaw in the quality score mechanism is too its reliance on evaluating “expected CTR”, which is derived from exact match CTR, as a benchmark to dole out quality scores. The ad platforms weight CTR as a bit more than half of the quality score calculation.
The expected quality score is an imprecise estimate, but it works well enough. And if it worked perfectly, there would be more attempts to “game” it. It’s a pragmatic solution that works.
There’s a second, more serious, and more fundamental problem with Quality Score, and we call it “the keyword silo overlap problem”. Many keywords have multiple meanings and uses, which we have come to describe as “search intent keyword silos”. The same keywords are often used to express different search intents, or even in completely different markets. Here’s one example.
Example: Different “search intent keyword silos”
Let’s take the keyword “Wi-Fi Router Antenna” for this example. There are millions more of overlaps we could use, but just for sake of showing an example let’s just agree we’re randomly picking this one.
Who might be advertising with that keyword? Here are a few possibilities:
- A Wi-Fi OEM that sells routers with unique or special antennas.
- A ‘system integrator’ or VAR searching for a manufacturer of Wi-Fi antenna components.
- An electronics supplier selling replacement Wi-Fi Antennas for popular routers to consumers.
- A Wi-Fi range booster reseller.
- A student studying microwave antenna design and propagation.
- A consumer searching for a replacement router antenna after her dog chewed off the old one.
Each of these markets is advertising to a different type (persona) consumer who also resides in a different place in global distribution channel ecosystems. So each will have it’s own position in different sales funnels, and different natural, average click-through rates. The same keyword is therefore used in different “keyword search intent silos” (our term for it).
Therefore, some advertisers will “get lucky” and receive higher quality scores than they should, and others may be dismayed that they can never improve their quality scores much, no matter what they do with their ad copy and landing pages. The problem is, they are being “stepped on” by other markets – people searching for different things, with different search intent, but who share the use of this particular keyword.
Despite these issues and others, QS works pretty well and is the best mechanism that’s been developed to date, to ensure that the best ad copy and landing pages are favored at auction.
Something to keep in perspective about QS
Quite a few PPC campaign advertisers and campaign managers become obsessed with quality score. They end up inadvertently spending too much time attempting to optimize QS or even, “game the system”. This has a tendency to work against them. At some level, QS is a fact of PPC life, and you have to learn to live with it. It is that way by design, and it works.
The way to handle QS is to develop the highest quality, most relevant ad copy and landing pages possible, and to build campaigns with highly structured ad groups. It’s ok to do some experimentation, but after some effort QS will become asymptotic (reach a point of diminishing returns) to further improvements. So effort should be focused elsewhere in the campaign, versus obsessing on QS. Sometimes keywords should simply be abandoned, say if they are heavily used in another market that distorts quality scores in your campaign.
The bottom line on QS
QS is a helpful aspect lying at the core of how PPC campaigns are engineered to operate successfully. It is important to keep in mind that QS serves a useful purpose. It’s also helpful to be aware that ad platforms are constantly adjusting how QS works, and they are not going to tell advertisers how they are changing it. So it’s best to focus on building great ads and landing pages, versus trying to game quality scores.