AdWords Tips

Going Unicorn Hunting: The Secrets Behind Ads with 3x the Average CTR

By Larry Kim February 11, 2014 Posted In: AdWords Tips Comments: 16

What’s the average click-through rate for a Google AdWords ad? When people ask this question, it’s often because they want to benchmark their own ad CTR’s. Unfortunately, way too many advertisers are content to reach an average click-through rate and leave it at that.

We recently held a well-attended webinar packed with crazy techniques you can use to triple your AdWords click-through rate. Sounds unbelievable, right? It’s not at all – in fact, we walked step-by-step through a process that helps marketers improve their ads to the point they’re performing in the top 1% of all ads across the platform.

We’ve had a number of requests for more information on this webinar, so here we go! This post is a monster – use this handy guide to help you get around:

First, we had to establish one very important thing: a lot of the "experts" you’re currently turning to for advice don’t know what they’re talking about.

What is a Good CTR?

There are blog posts, articles and podcasts about CTRs littering the web, but no one really likes to give a direct answer to, “What is a good CTR?” Much depends on your industry, competition and goals. However, I can tell you this: marketers who are satisfied with 2-5% CTRs, despite conventional wisdom telling you this is a good thing, are underachieving.

In our webinar, we first looked at two ads side by side and asked the audience to identify the “winning” ad.

which ad wins

Not surprisingly, the ad with the higher CTR was selected as the winner by an overwhelming 81% of people.

I wasn’t being entirely fair, though; it was a trick question. Neither ad had a good CTR and therefore neither were winners.

We can’t see our competitors’ data, so we don’t know if their CTRs are similar to ours. What we do know is that whether or not your CTRs are “good” largely depends on your ad position. Ads in higher positions naturally have higher CTRs than those with less prominent placement.

good click through rate

The above chart reflects the average CTRs per ad position, based on our analysis of nearly 100,000 advertiser accounts representing $3 billion in annualized spend. We have a pretty good idea of the bigger picture when it comes to the CTRs real advertisers are actually experiencing.

So the ad in the second position has an average click-through rate of 4%. That’s good right? No, that’s average. When you were a kid, did you lie awake at night dreaming about how you were going to achieve your goal of growing up to be average? Of course not. Even in the third and fourth positions, 2-3% CTRs are just average, so if this is what you’re seeing in your account, you have a lot of room to grow.

Averages are great, but we needed to dig into specific accounts to see just how much variance there is between advertisers. What we found is that CTRs vary widely between advertisers, as you can see in the chart below. Remember, these are not keywords – each blue dot indicates an advertiser account. To keep it clean, we eliminated low activity and low spending accounts.

Ad CTR Winners

In this chart, you can see the expected or average CTR curve in orange, but I’ve also added the green and purple curves, to show you what you should aim for if you want to achieve CTRs two to three times the average.

We can see from this data that there are, indeed, a few stand-out accounts who are killing it and outpacing their competitors by far. What does it take to be one those stand-out advertisers?

The bottom 50% of advertisers are achieving CTRs of average or lower. If you want to be in the Top 15% in the very competitive Google AdWords landscape, you would have to achieve an average CTR across your account two times higher than the average (we call these Awesome Advertisers). Three times the average CTR makes you a Super Awesome Advertiser, which about 1 in 20 advertisers accomplish.

Why not shoot for the stars, though? If you want to be in the top 1%, you need an average CTR six times higher than the standard – we call these “Unicorns.” They’re rare, but I promise you, AdWords Unicorns exist. They are not inherently better than you are; they simply work harder and optimize better.

Here, you can see in more detail what it takes to be an AdWords Unicorn, with the required CTR for each ad position listed in this table:

below average ctr

So how do they do it? First, we need to understand…

Why Unicorns Matter & How AdWords Works

Understanding how the AdWords auction works is critical if you are going to optimize and improve your performance.

AdWords is a live auction, much like eBay. Google needs to determine the ranking position for each ad, as well as the cost of each click, in real time for each ad in the auction.

Google uses Ad Rank to help determine your placement in the auction and your CPC. This is calculated by multiplying the maximum cost per click bid (the most you’re willing to pay for a click) by your Quality Score (Google’s rating of the quality of your ads). (Note: Google recently updated Ad Rank to include the expected performance of ad extensions, but the basic principle is the same.)

ad rank formula

In this example, you can see exactly how much Quality Score can affect your performance in this live auction. The top advertiser, who achieved the #1 ad position, had a Max Bid $2 lower than his competitor, yet his Quality Score and therefore Ad Rank are far higher, resulting in a better position and lower CPC:

average ctr

Google does this because they don’t get paid if no one clicks on their ads, so they give preference to the ads most likely to succeed.

Wow, Quality Score Sounds Pretty Important…

Yes, yes it is.

We’ve found that the average Quality Score is roughly 5.1 out of 10.

average quality scores

If you have an above average Quality Score, you can get up to a 50% discount on your cost per click. Conversely, if you have below average Quality Scores, you could pay up to 400% more. Ouch.

CPC is just one metric, though. What about the actual cost per conversion? Here, we’ve also found an inverse relationship, across billions of dollars of ad spend, that as your QS increases, your cost per conversion decreases because your clicks are so much cheaper.

cpc vs. quality score

So how is QS calculated? In general, the higher your CTR, the higher your QS. As you can see in the chart below, though, there is a lot of variance, thanks to ad positioning. As we discussed earlier, ads at the top of the page are predisposed to getting more clicks than those in lower positions.

expected ctr by position

The blue line at the bottom, for example, is for keywords in position 1.0 to 1.5. Quality Score isn’t calculated by your CTR alone, but by whether you’re over- or underperforming compared to the expected CTR. In this next graph, you can clearly see just how important your CTRs are to your QS calculation.

expected click through rate

It’s all relative.

Your Quality Score, which is the key to getting those prominent ad spots, low CPCs and low CPAs, is not entirely based on your CTR. Rather, you need to have ads and keywords above the expected average. That’s the key to high Quality Scores.

The Top 1% of PPC Ads and What Makes Them Tick

So what about those unicorns?

We’ve learned so far that an above average CTR will help you get there, but to be the best of the best takes more.

three times average ctr

When we see unicorn accounts in the AdWords Performance Grader, they share some common traits:

  • Every keyword in their account has a high QS.
  • They are effectively using negative keywords to reduce wasted spend.
  • Their high impression share demonstrates that Google loves showing their ads.
  • They have those super-high, impressive, double-digit CTRs than outperform the average by many times.
  • They’re targeting highly specific longtail keyword phrases.
  • Their account activity is high – they continue to optimize and improve on an ongoing basis.

If you can home in on the ad and keyword combinations that have these above average CTRs, the rewards are huge – higher impression shares, greater ad positions, lower costs per click and conversion.

Hunting Unicorns

Now you know what AdWords Unicorns look like and should have a pretty good idea why they’re so desirable. Let’s look at each of the characteristics so you can emulate this success in your own account.

Unicorn Keyword Strategy

The Top 1% of all AdWords accounts tend to focus on high commercial intent keywords. This is not surprising, as people tend to click more when they’re already motivated and have an intent to purchase. For example, “best internet marketing software” or “stainless steel dishwasher” are keyword phrases that demonstrate an intent to purchase. (Something like “Lincoln’s birthday” does not.)

Branded keywords are another fantastic opportunity for advertisers, but we’re not just talking about your own brand. Target partner brands, competitor brands, or other brands your target market may searching on when they’re in buying mode.

A third type of keyword that does well is local keywords, where the intent of the searcher is to find your business – or that of your competitor (e.g. “San Diego personal trainer”). These tend to get high CTRs, as local searchers often have a high commercial intent.

Avoid informational keywords, where possible. These are often phrased as questions, for example, “What is a dishwasher?” or “Where is Chicago?” These searchers tend to be looking for Wikipedia-type information and aren’t demonstrating an intent to buy, which can result in a lower CTR or wasted spend.

A Caution on Dynamic Keyword Insertion

If you want to achieve unicorn status, don’t overuse Dynamic Keyword Insertion, or DKI. This is a strategy whereby you use a certain syntax in your ads that allows Google to automatically substitute the keyword that triggered the ad into your ad copy or headline.

DKI is beneficial, but only to a point, as illustrated below:

dki average ctr

Here, we have two curves: the red curve is ads with DKI and the blue is those without. The red curve shows that ads with DKI are relatively more abundant in the top 15%. However, as you move up into the top 5% of ads right into the unicorn ads, you’ll notice the red curve actually dips below the blue.

Dynamic keyword insertion ads seem to hit a wall. This tells us that it should definitely be part of your toolbox as it tends to produce slightly higher early returns, but it shouldn’t be overly relied on.

Ad Extensions – Is This the Magic Bullet?

Google seems to be releasing new ad extensions almost every week, but what’s the impact of ad extensions usage on the Quality Score of ads and will it help you become a unicorn?

ad extensions impact

This graph shows that impact and it’s not a game changer. We can see that the accounts using sitelinks on their ads did slightly better than those that did not, which is not a surprise. Extensions provide an incremental lift in CTR, but it’s not going to remedy boring copy. It’s a modest increase, but won’t double or triple your CTR rates. Again, use them, but don’t rely on this alone to shoot you to unicorn status.

Most Ads Suck. Create Emotional Ads.

below average ctr

A lot of search results pages look just like this, with boring ads that all look the same. This is exactly why position can make such a difference; because the only differentiator is where an ad appears on the page. Borrrrrring.

But wait – this is a huge opportunity for you, Unicorn Hunter.

When you see a bunch of boring, dry, similar ads appearing for a query, you have a chance to break out and be different. I like to call this an “AdWords Jackpot.”

AdWords CTR

The alternative is to come up with something that stands out from the crowd, that uses emotional triggers.

Creating Emotionally Charged Ads with the Swiss Army Knife Blade Brainstorm

Respected marketer Perry Marshall devised a clever, structured method of brainstorming persuasive ad copy called the Swiss Army Knife. You take different entities such as your customer, something that your customer loves, something they hate, their best friend, their enemy, etc. You then determine how your brand is connected or related with each entity and build out from there.

ad writing techniques

It’s a simple, structured brainstorming process that can help you generate more compelling ad copy for testing, by mixing and matching those concepts, like this:

above average click through rate ad

This is a perfect example of a unicorn ad, with over 30% CTR, that stands out against all of the boring divorce lawyer listings.

Your ad copy has to resonate on an emotional level and it must be different to stand out. You must test the big things in your ad messaging, meaning stuff like your guarantee, your unique selling proposition, and your offer. Not just punctuation.

Understanding Relative Abundance

We’re still on the hunt for unicorns and by now, you should be much better equipped to find them. There’s still a missing piece though: effort.

Unicorns wouldn’t be all that special if they were abundant. In fact, only 1 in every 100 ads is a unicorn. The chances that you’re going to find it on your first try are slim to nil; on average, you need to try out 100 different ads to find your unicorn.

Above Average CTR

The more ads you create and test, the more likely you are to hit on that one that outperforms the average by 2x, 3x or even 6x!

You’re Not Testing As Many Ads As You Think You Are

But Larry, I already have THOUSANDS of ads in my account! The last thing I want to do is write more ads!

That may be true, but I can almost guarantee you aren’t testing as many ads as you think you are.  Certainly, they weren’t all created using the insight and process outlined above.

Have a look at this example – can you tell the difference between the top right ad (circled in red) and the one below it?

Test More Ads

This is a real-world example where an advertiser adds a comma in the second ad and calls it a variation for testing. In fact, this example is testing six copies of the same ad over and over.

Let me tell you a secret: the bar, especially for SMBs, is incredibly low. The average SMB AdWords account has just 18 ads. Maybe you can’t get to 100 ads this month or even this year, but if you work away at this and test even 40 or 50, you’re already doing way better than the majority of your competition.

5% of Ads Account for 85% of Impressions

Here’s some more good news: you can get rid of your underperformers and boy, there are a lot of them. Across accounts, we’ve found that the top 5% of ads in an advertiser account make up 85% of their impressions. It’s not necessary to test 100 ads for every ad group in your account.

Instead, choose your top two or three ad groups and really focus your efforts there. Test those like crazy, ideally reaching 100 ads to test in each  of those top performing groups.

Bonus Tip: Mobile Ad Optimization

Mobile ads are obviously important – almost 50% of searches are going to be from mobile this year. I want you to ignore everything I’ve just said when it comes to mobile ads.

Our research has found that the calls from mobile ads convert 3x higher than clicks to a website. When you’re testing mobile ads, you need to base your findings on call rate, not CTR or conversion rates.

I repeat: go all-in on call rates and do everything possible to get people to click the click-to-call button.

What Does It All Mean?

Congratulations, you made it! Let’s recap and then I’ll send you on your merry way to go back to AdWords and slay your competition:

  1. Above average CTR keywords = success.
  2. Below average CTR keywords = trainwreck.
  3. So… delete the bottom third of your account and re-deploy that spend to remarketing. These are your low CTR, low impression share, junk performers. Lose the dead weight, it’s dragging you down.

Crazy, right? Your boss is going to freak out if you delete a third of your AdWords account. However, the idea here is to maximize the budget you’re now spending on junk ads.

As you see here, the cost of the average click can be substantially lower across a number of industries in the Google Display Network when compared to Google Search.

Average CPC by Industry

Similarly, the conversion rates for display ads are comparable to those for search ads, especially when you’re using remarketing. In some industries, like automotive and travel, the display network conversion rate can be even higher.

Average Conversion Rates

If you can get rid of the bottom third of your more expensive search ads and shift that spend to even average performing display and remarketing ads, that would be a brilliant way to begin to more effectively scale your AdWords efforts.

Life is so much better for the unicorn. We did segment results by industry and didn’t even show it, because we found that these unicorns exist fairly evenly across industries, and for advertisers of different sizes and levels of spend. YOU can be an AdWords unicorn.

Of course, you can get started with your own instant, free account audit using our AdWords Performance Grader – this is a critical first step in determining the health of your AdWords account and identifying optimization opportunities.

Want to learn more? See the entire webinar below:

AdWords Performance Grader




If you enjoyed this post, please consider leaving a comment.

Comments

Tuesday February 11, 2014

DGM (not verified) Said:

Hi Larry,

I thoroughly enjoyed the tip on the various variation. Many marketers do not test this PPC technique and just a little comma, can increase conversions, and in some cases, a few % or more!

Take care! 

Genuinely,

 

Jordan 

 - Dove Global Media

Wednesday February 19, 2014

Larry Kim Said:

yes. certainly punctuation matters. like having correct capitalization and stuff. but this is table stakes. the very minumum and will make a small difference. However, to be a unicorn you must test the big things in your ad messaging - meaning stuff like your guarantee, your unique selling proposition, your offer. Mastering this picture stuff (also known as "marketing") is the only way to get 2, 3 or even 6x higher CTRs.

Wednesday August 06, 2014

Bryan (not verified) Said:

Larry,

In your "advertiser 1" calculation of CPC, where did the numerator of 16 and denominator of 10 come from?   It appears that the 10 might be the quality score but I'm not sure where the 16 came from?   I'm just trying to figure out the sources of the $1.61 CPC calculation for advertiser 1.

 

 

Wednesday August 06, 2014

Elisa Gabbert Said:

Hi Bryan, It's the ad rank of the advertiser below you.

Tuesday February 11, 2014

Peter Martino (not verified) Said:

Larry, just watched your On Demand Webinar about Unicorns.  Some awesome stuff in there—mind blowing statistics which are hugely valuable. It’s clear you’re a mathematical genius and in our office your name is used as many different parts-of-speech and is generally synonymous with taking over the world. For example, “Let’s get all Larry Kim on that Ad Group”. (Can you sense the BUT coming? Here it is)…

But, there’s one thing you do consistently (without qualification) that I think is a mistake.  You try to set broad-sweep standards for all accounts and only use some metrics to do it—especially CTR and Quality Score. If we’ve learned anything from the Sabermetrics era of baseball it’s that you can’t use 1 stat to evaluate Player A vs. Player B and you certainly can’t do so if you don’t understand what type of player they are or what their strategy is at the plate. With PPC, this is especially true when considering that maybe some Advertisers have different approaches. Maybe some are targeting searchers with a very singular & linear purpose. For example, I sell Widget A, my keywords, ads, and landing page are all about Widget A, I’m only looking to show ads to people who want to buy widget A today. Well of course I should be consumed with a better CTR, Higher QS and lower CPC related to Widget A. But what if I sell Widget B? Widget B is different from Widget A and in our experience is a better Widget. Yes I’m marketing to people who know about Widget B, but I also have chosen to use a slant approach to convert Widget A users into Widget B users, and have the data to show that I can be profitable doing so. We will have some casualty clicks--people that are diehard Widget A users and want nothing to do with something as awesome as Widget B and that will affect our CTR, CPC, and QS. But if a substantial enough amount of people search for Widget A but end up buying Widget B we might actually have a successful campaign despite our sub 3% CTR and sub 7 Quality Score. Naturally we should always continue trying to improve all our measurables to reduce CPC and CPA, but I think this is just one example of why it’s dangerous to set universal standards and evaluate advertisers with an “apples to apples” method without understanding their approach.

I say all this because one of our most successful and most profitable PPC Campaigns is exactly as described above. If we were overly consumed with CTR, CPC, and QS we may have paused that Campaign long ago. Yes we are still working to improve those metrics, but I doubt our CTR will ever be 5%+ despite an Average Position under 2.0. Yet this Campaign brings us tons of leads and tons of profitable business.

Keep up the good work! We’re extremely satisfied customers and Wordstream evangelists. 

Wednesday February 19, 2014

Larry Kim Said:

Hi. Thanks for this note. Couple thoughts here:

  1. I’m no expert on sabermetrics (I’m a Canadian – talk to me about hockey, eh?) but my general understanding of books like “moneyball” was that the industry was trying evaluate the “quality” of baseball players without objective benchmarks and comparative data on metrics that matter – instead they were making distinctions like “that guy is tall, handsome and likable” or whatever. I think my approach here is very much in line with the intent of “moneyball”-type analysis in that I am attempting to define metrics that matter and quantifying the range of values and the average costs/benefits of being above/below those values. It’s a data/quant approach to PPC benchmarking. Most other blogs i read are more like the "this is a hansome likable PPC campaign and i have no data to back it up" kind of thing.
  2. You outlined some scenarios where an advertiser shouldn’t be concerned with better CTR or higher QS/lower CPC and make the case that ultimately all that matters is CPAs – essentially, that there are some exceptions to the rules. I think we’re on the same page here in general. But like you said, even if you are making a profitable target cost per action on “widget B”, things could always be better with higher QS/lower CPC, which is exactly my point. Don't give up on the goal of 5% CTR at position 2! try a different offer, like a guarantee or different unique selling proposition, etc.

 

 

Thursday February 13, 2014

eva (not verified) Said:

Fanatstic Post! Thank you so much for taking the time to explain all the important variables in a concise and easy to understand way. It is amazing how easy it is to waste your budget if you don't drill down to the details.

Thursday February 13, 2014

Drew Bay (not verified) Said:

Hi Larry,

Nice post. You've covered 'the essentials' in a clear and illuminating way.

1 thing that's always bugged me though about Quality Score: 'Ad Relevance.' Presumably, this is referring to how relevant the ad copy is to the keyword that triggered that ad. All fine and good. However, far too often than not, an Adwords advertiser will have an 'Average'-- or even more infuriating-- 'Below Average' score for this metric, despite a few mentions of the keyword in the ad copy itself.

Now, what I'd appreciate, is a full 'deep-dive' into the 'Ad Relevance' portion of Quality Score. What comprises Ad Relevance score? Shouldn't mention of the keyword or variations thereof be 'enough' to constitute an 'Above Average' grade for QS?

This leads into the Unicorn Ad that Perry wrote: I love it. Coming from a direct-response copywriter like myself, that kind of copy is such a refreshing reprieve from the 'usual,' uninspiring ad copy you see for lawyers (or anyone for that matter). However, wouldn't it miserably fail the Ad Relevance portion of the QS caluclation? Certainly the keyword that is paired with that ad in the auction has nothing to do with the copy Perry used. 

Now, I'm sure it's not the end of the world, and you could likely still squeeze an 8 or 9 on QS for the keyword if your other metrics were bang on (Exp. CTR and Landing Page Experience), but isn't an ad like that detrimental to QS simply due to 'Ad Relevance?'

I'd LOVE to get your take on this illusive metric and why it never seems to make sense for the given keyword and ad in question.

Thanks Larry!

- Drew

 

Wednesday February 19, 2014

Larry Kim Said:

Hi drew,

this may sound surprising but i don't believe the keyword/ad text relevancy matters much in general in calculating quality score. There is such a strong relationship between Expected CTR and Quality Score that i'm somewhat skeptical of any claims of substantially increasing quality score that weren't mostly based essentially entirely on raising CTR. meaning, if you can have a high CTR ad that doesn't contain the search query, you can still get very high Quality Scores provided that it has above-expected CTR for it's ad position.

an example of this is competitor trademarks. ads on competitor trademarked terms is allowed, but you're forbidden from including the trademarked keyword in your ad copy. Nevertheless, these ads can get high Quality Scores desipite not having ad copy/search query alignment.

 

 

 

Wednesday February 19, 2014

Drew Bay (not verified) Said:

Thanks Larry!

Alright, that makes sense... I suppose it was more of an aggravation on my end than something that seriously debilitates Quality Score and therefore the keywords (and ad group!)

I'm a little confused on your last point;  "ads on competitor trademarked terms is allowed, but you're forbidden from including the trademarked keyword in your ad copy."

This sounds like "you can use your competitor's terms in the ads, but don't use your competitors terms in the ads!"

Could you clarify?

I mean, we have a client or two that are bidding on competitor keywords and have them in the ad copy and landing page, and I spoke with Google and they said... unless they specifically requested trademarking with Google, you can indeed use those terms in the ads.

I must be confused, eh? Thanks Larry. 

 

Wednesday February 19, 2014

Larry Kim Said:

you can bid on a trademarked term.

however you may not use the the trademarked term within the ad copy.

 

Thursday February 20, 2014

Drew Bay (not verified) Said:

Makes total sense, thanks!

Monday March 24, 2014

Shana (not verified) Said:

What a well written and informative article! This gives us insight and information to improve our Adwords CTR!! Unicorns here we come!!! 

Thank You! 

Wednesday August 06, 2014

Shawn Duffy (not verified) Said:

Wow! Thanks for all the great info and tips! I am new to adwords and I want to start off right. Your article is full of awesome info and ideas on how to rock adwords. I am off to hunt some unicorns! Thanks again.

Monday August 25, 2014

Stephan (not verified) Said:

Hi Larry,

Good suggestions with the emotional ad copy. The example you provide is all about seeking marriage revenge, an incredibly emotional subject. What I noticed about the ad, however, was that it didn't seem particularly keyword rich. If the headline of the ad is primarily filled with keywords, how do you creatively get around that limitation?

So to stick with the marriage revenge example, I could imagine people are typing in things like "marriage private investigator", "husband cheating on me", "divorce lawyers", etc... None of those phrases appear in the ad copy. Despite the creativeness of the ad copy, wouldn't the lack of keyword relevancy kill quality scores?

Stephan

Friday September 05, 2014

Brandee (not verified) Said:

Wow Larry, very valuable and robust information, so intriguing when applied to ecommerce!

Thanks for sharing!

-Brandee

Leave a Comment

Type the characters you see in this picture. (verify using audio)
Type the characters you see in the picture above; if you can't read them, submit the form and a new image will be generated. Not case sensitive.
 
Free Keyword Tool

Get thousands of relevent keyword suggestions - more,
faster, free!

Free Keyword Niche Finder

Discover profitable pockets of keywords for your
business.

Free Negative Keyword Suggestion Tool

Identify wasted spend before it happens and increase
your paid search ROI.

Contact Us | Company | Support | Site Map | Trademarks | Privacy Policy © 2007-2014 WordStream, Inc. All rights reserved.