Just Say No to Low CTR

Editor's Notes

1. In my presentation today we’ll be talking about paid search strategies. Specifically, I’ll try to convince you that pursuing a low CTR strategy or bidding to a particular position is on average, a losing bet.
2. First let’s define what I mean by a “low CTR strategy” in paid search which I think broadly speaking, falls into two categories: The first has to do with going after broader keywords within your niche – you know, the keywords with stubbornly low click through rates. For example, keywords with informational, navigational or ambiguous search intent and thus tend to suffer from stubbornly low click through rates. Second category is where you employ a strategy to reduce clicks on your ads by either going after lower ad positions or sabotage your ad copy with to discourage unqualified buyers to click on your ads.
3. Now guys just a quick warning here. I’ve got like 40 data-packed slides and just 15 minutes to go through them so try to keep up! After all, this is SMX advanced not SMX beginner.
4. A few years ago, I think pursuing either of those tactics would have worked out very well for you. I’d argue that successful companies like ebay were built by using the low CTRs strategy. There definitely was a time when it was a highly effective strategy. But over the last few years I think things have changed and as a result I believe that it’s time that SEMs adjust their strategies.
5. The first big change I’ve noticed is that Google expects way more performance from your ads. They’ve raised the bar in terms of the quality they expect from your ads. Here I’m showing some research into expected click through rates across a few thousand advertiser accounts and found that the average click through rates for any given ad position is much higher than you might expect -- and keeps trending even higher. So why should you care?
6. It’s because I’ve reverse engineered the Quality Score equation –I don’t have time to go into everything but the key finding was that Quality Score is essentially just a matter of beating the expected click through rate. So, for the rest of the presentation, a “low quality score ads” refers to an ad whose click through rate falls are below google’s expected, average click through rates for a given ad position.
7. Quality Score matters more and more every day. Over the last 2 years has been quietly changing how the ad auction works. They haven’t announced this but we notice it based on our internal research. Basically, today the average quality score is just 5 / 10 whereas 4 years ago, the average quality score was 7/10.
8. As average quality scores drop, the value of having an above average quality score in terms of the discount on cost per click is 200% more than it used to be. The following table shows the huge discounts and penalties that are applied to your cost per click based on your average quality scores.
9. Part of the logic behind the low CTR strategy is that by lowering your Click through rate, you’ll lower your costs. Like this dawson creek guy – he doesn’t want high click through rates because that means he’ll have to pay for more clicks. But the reality is that the up to 400% low click through rate quality score tax imposed by google that you’ll end up paying for fewer clicks at dramatically higher prices. You’re much better off being very picky in your keyword selection targeting options and trying to get as high a CTR as possible.
10. Another rather huge issue with the low CTR strategy has been quietly raising the bar for the quality of ads they want to show in search results. Again, here’s some new research that we’ve done that shows the exact relationship between quality score and impression share. The key takeaway here is that for every 1-point increase or decrease in Quality Score, you can expect on average to gain or lose an average of 9% impression share.
11. I’ve heard people say that it’s not possible to get high quality scores in your specific industry (as if google has a personal grudge against you and your industry). Keep in mind that quality score is based on beating an expected average click through rate for a given ad position. So by definition, in any industry, essentially around half of the accounts out there have above average quality scores, the other half don’t. To prove this, I recently segmented average quality scores by industry and found that in all industries, there were (surprise surprise) both high and low average quality scores for every industry!
12. For example, I looked at specifically at B2B companies (the purple dots in this graph) and found that the calculations used to determine their Quality Scores weren’t a heck of a lot different from any other industry.
13. What about Finance Companies – the yellow dots in the graph? surely google must be out to get wall street? Again, not a huge difference compared to other industries.
14. Same with Ecommerce companies – the orange dots in this graph. Again, their quality score calculations were very similar to every other industry out there. So if you hear anyone say that It’s not possible to get a high quality score – I think that could be a bit of a cop-out – the problem could be their strategy.
15. So we know that quality scores which are based on click through rates – will dramatically increase or decrease your cost per click. But why should anyone care about cost per click? After all if you can sell products or services profitably at 100$ cost per click, who the heck cares what the cost per click is? Totally fair point. The reason why you should care is because there’s a strong relationship between Cost Per Click and Cost Per conversion. Meaning, the higher your cost per click, the higher your cost per conversion. We rand the numbers on hundreds of millions of dollars of ad spend. We found that the higher your CPC, the higher your average CPA. Or put another way, even if you’re running profitable PPC campaigns with high CPCs, it would be even more profitable if you could lower their CPCs by raising click through rates and quality scores.
16. Lots of people tell me: Larry, I Don’t WANT a high click through rate! I don’t want to be in position 1 or 2 and therefore pay for too many clicks. I’d rather be in position 3 or 4 or 5 because those positions somehow get fewer clicks and better conversion rates.
17. Guys, we looked at conversion rates of ads in the first position and second position and third position (etc.) and looked at thousands of accounts and found on average there was no relationship between conversion rate and average search position. So there’s nothing magical from a conversion rate perspective about position 3 or any other position for that mattter. Don’t take my word for it - Google has also published research that confirms no relationship between Avg. position and click through rate.
18. Ok so a friend of mine, sam owen pointed out to me that in one of his accounts, the ads labeled “free” have the highest click through rates (1.1%) but also the worst cost per conversion (!143.75) So what the heck is going on here. My view is that what you’re seeing here is actually a keyword targeting issue or a landing page issue. Alternatively it’s a Change the Keywords. Maybe free is a bad keyword we shouldn’t be going after. The key point here is that the high CTR strategy doesn’t mean to go after keywords that don’t make sense for your business "Free" is resonating with searchers (they’re clicking) but they’re not converting. Try radically changing the offer? (like actually offering something of value that is free) Let me give you an example of this
19. We’ve looked at conversion rates across thousands of accounts and found that the top 10% of advertisers across any industry have conversion rates that are 3-5x higher than the average advertiser out there and are absolutely killing it. But what we found was that the reason for their conversion rate success had almost nothing to do with their average ad position, and had everything to do with their landing pages (their offer, the guarantee, unique selling proposition, etc.) and also their keyword selection.
20. I call these landing pages unicorns because they’re a beautiful and rare thing. And I’ll show you how I created a unicorn of my own.
21. The new offer I created was an adwords report card. It just downloads some data in your adwords accounts and just benchmarks your adwords performance against your competitors. This new thing converts 4x better than the old offer, and the moral of the story here is that if you’re seeing high Click through rates and low conversion rates, it’s possible that this is just a symptom of some other problem – for example, your offer might just suck and need to be revisted.
22. I still haven’t convinced you of the futility of a low CTR strategy, I’ve saved my thunder for the end here. One of the most compelling reasons to drop the low CTR/Quality Score keywords and ads from your account is that they’re killing the rest your account.
23. Take a look at this crappy account here. The dots on the graph are the keywords in the account, which I’ve plotted based on the their click through rate and average ad position. See that yellow line there? keywords that are above the yellow line are beating the expected click through rates and keywords below the yellow line are below the expected click through rate. Now, as you can see, there’s a bunch of keywords that are doing very well and are beating the expected click through rate, and there’s a lot more keywords that are failing to beat the expected CTR. So you’d expect to see a mix of both high and low Quality Scores, right?
24. Unfortunately that’s not the case here as you can see in the account’s quality score distribution. Notice how all of the keywords account are quality scores of 3,4,5 and 6. and they have no high quality score keywords of 7-10. What’s happening here is that the low CTR/Quality Score Keywords reduce your campaign and account average quality scores and click through rates, which in turn, reduces the quality scores of all of the keywords in your account, including the good ones.
25. The reverse is also true. Look at how in this account, most of the keywords are beating the expected click through rate But some of the keywords are below the expected CTR curve, so you’d expect to see mostly high quality scores and a few low quality scores, right?
26. Wrong. Notice how every single keyword in this account has a quality score of 7 or 10, even the keywords with 0% CTR. What’s happening is that by having so many above keywords with high average quality scores and click through rates raises the campaign and account average quality scores in your account, and in turn raises the quality scores of all the keywords in your account, including the bad ones!
27. For example, my boss or client gets mad if he does a search for something he thinks is relevant and doesn’t see our ads. No problem. Use RLSA! Google Analysis of My Customers comparing Regular Search Ads vs. RLSA targeted Ads: 2.5x Avg. CTR, ½ Avg. CPC, and 1.5x higher Quality Score. RLSA = Guaranteed to Raise CTR/QS on even the most Stubbornly Low CTR Keywords
28. At this point you might ask, if I delete the crap keywords in my account, I’ll lose a good amount of my conversion volume. How do I make up for those lost conversions? My boss wants more conversions, not less conversions! The answer is to lean on remarketing and other new display in-market targeting options to scale your paid search efforts.
29. Lets recap here. your above average QS keywords will have low CPCs, lower CPAs, higher impression shares etc. Your below average keywords will be doing terrible and will be making it difficult for even your best keywords to succeed. The idea here is to delete the bottom third of your search marketing campaign (basically the toxic stuff that is killing your account) and allocate this to display remarketing or in-market targeting display campaigns.
30. Here are some typical average CPA’s based on some customer data. Basically the strategy works because, Average Remarketing should beat below average Search, and because by deleting the bottom third of your account, you’ll be helping the rest of your account to be successful.
31. Just a note on averages. in my research here, I’m talking averages. if you happen to see one instance in your account where my advice isn’t working out here, that’s totally possible. What I am saying here is that on average, I promise you that strategies that I’ve outlined for you today will work more often than not.
32. Hopefully in my presentation I’ve convinced you to just no to low click through rates. and to Instead consider use of remarketing and RLSA to scale your search marketing efforts.
33. Here’s the link to my presentation – lets connect on social media! Just mention that you attended SMX.