We love digital

Call 0845 485 1219

We love digital - Call and say hello - Mon - Fri, 9am - 5pm

PPC Click Rate Study – How Click Rate Deteriorates Depending on Ranking

Andy Headington

by Andy Headington on 15th August 2012

PPC clicks vs RankingsToday we present a detailed study in to how PPC click rate changes depending on ranking by Andy Headington of Adido.

There has been a lot of research done into the click distribution of both organic and PPC adverts over the years. When I first started in the industry nearly ten years ago the distribution of clicks on search engines was heavily dominated by natural clicks with around 85% of clicks being generated on the left hand side of the SERPs. Recent research by Wordstream suggested that now as much as 65% of clicks are generated by PPC listings.

This is a huge shift which can be attributed to several factors including Google’s push of AdWords listings to take up more real estate, the blurring of the design of paid ads and changes to user behaviour to trust PPC ads more. It is something that we’ve seen ourselves for our clients and I’m sure other agencies have as well.

What is the drop off in AdWords positions?

With such a shift in the way that Google and Bing now work and focus their listings, getting top spot is even more important. But just how important is it to bid to be top spot? What is the difference between position one traffic and say position three? I’ve been fortunate enough to have access to some real data to show the drop off in CTR on the top positions which I’ve presented below.

Some notes on each of the websites shown here.

  • Website 1 is an established site which has been using AdWords for several years with strong quality score and CTRs on many phrases.
  • Website 2 has been running for nearly a year with a smaller budget than website 1. It has second position on many target phrases but still delivers good CTRs and conversions.
  • Website 3 was launched this spring to try and pick up conversions and searchers which were being missed by websites 1 & 2.

The data below represents a month when all three websites were competing against each other on the same phrases, albeit with different budgets.

[Phrase 1]

 

Website 1

Website 2

Website 3

  Position

1

2.1

4.8

  Impressions

9,545

7,929

6,109

  Clicks

1,951

1,279

160

  CTR

20.44%

16.13%

2.62%

  CPC

£3.56

£1.96

£4.30

  Conversions

217

31

4

 

[Phrase 2]

Website 1

Website 2

Website 3

  Position

2.3

2.3

3.5

  Impressions

2,463

955

1,562

  Clicks

260

155

141

  CTR

10.43%

16.23%

9.03%

  CPC

£2.81

£1.32

£2.17

  Conversions

17

1

2

 

[Phrase 3]

Website 1

Website 2

Website 3

  Position

2.3

1.8

3.7

  Impressions

622

247

424

  Clicks

73

69

23

  CTR

11.74%

27.94%

5.42%

  CPC

£2.48

£1.49

£2.59

  Conversions

8

0

3

 

[Phrase 4]

Website 1

Website 2

Website 3

  Position

2.5

NA

4.1

  Impressions

711

NA

518

  Clicks

34

NA

9

  CTR

4.78%

NA

1.74%

  CPC

£2.3

NA

£2.40

  Conversions

0

NA

1

 

As you can see, there is a big difference here between position number one and number three or four of the same phrases. This drop off rate is considerable as the graph below shows.

PPC Ranking Effect

Things to remember

Obviously we have to remember that doing a test on CTR rates on PPC ads is always going to contain biases. Each ad shown in the listings is different and even the smallest change to any element of the ad can have a big impact on CTR. One small change to ad text, including site extensions, can play a role in increasing or decreasing CTR. I must caveat here that for a part of this data analysis, the site which was listed number one contained seller review stars which could have increased CTR slightly and the ad text of sites 2 & 3 positioned themselves very differently to website one. Despite the variances in the ads, it is my educated guess that the data shown here would be pretty similar across most industries.

We must also remember that ‘average position’ isn’t a straight average- it is a weighted average which is influenced by impression share. Therefore when some of these moved position (i.e. CPC got upped to try and improve CTR) it then changed the weighting of average position over the month. So the final total for the month shown doesn’t always tell the full picture (see phrase 2 which had two sites with the same average position despite very different CTR and CPC stats)

Conclusion

The conclusion we can take from this data is that the top position generates roughly 200-300% extra clicks compared to position three. Is this a good thing? If you are bidding to be top of your selected target phrases then you will take a big part of the possible search volume available. Being in number one position has nearly always generated most clicks due to ‘happy clickers’ and getting more eyeball time then other listings. If you can’t sustain the cost of being top position from a budget or ROI point of view then you shouldn’t dismiss bidding to be top completely.

With a good chunk of quality score being influenced by Click through Rates, and Click through Rates only being significant in the top one or two positions, you should consider my findings when you decide on your bidding strategies.

It’s also worth considering the effect on conversions. Whilst clicks are great, without conversions they are merely missed opportunities. In almost all cases we saw a rise in conversions when the ranking position was higher, expect for the last study which had a very low sample of data. However as all three sites were different, it’s harder to draw a conclusion as there may have been many other factors at play.

Nevertheless if you want to get exposure on new phrases or you are trying to turn around underperforming phrases in your campaigns then adopting an aggressive bidding strategy to achieve top positions (and therefore improving CTR and quality score) is the best way to go. Once you have established a good quality score and CTR you can start to reduce your CPC bids to bring your overall costs down without sacrificing your position too much. If you try new phrases and don’t end up getting over the tipping point to drive up your quality score your ads may end up struggling and you will have to pay huge CPCs just to tread water in positions 3 – 5.

The views expressed in this post are those of the author so may not represent those of the Koozai team

Andy Headington

Andy Headington

Andy Headington is Marketing Director of Adido. He has been working in digital marketing since 2003 primarily in search and social. In this time he has presented to thousands of directors and managers on all aspects of search marketing on behalf of the CIM, IoD, Business Link and many Universities across the South of England. You can read more of his thoughts on the ever changing world of digital on the Adido blog.

down arrow

Your Free Whitepaper

AdWords Beginner's Guide (2nd Edition)

AdWords Beginner's Guide (2nd Edition)

Download this whitepaper now and get a new one every month!

6 Comments

  • Mike Essex

    Mike Essex 15th August 2012

    Thanks for a really interesting study Andy. It’s a shame that the difference in sites means conversions can’t be looked at objectively.

    In terms of ad text were there major variations between the three sites, or were the adverts the same but rotated across all three? That would naturally have a knock on effect on CTR.

    As website 3 had a higher CPC, despite lower rankings was this due to a low Quality Score?

    Reply to this comment

  • Andy Headington 15th August 2012

    Thanks for the feedback Mike.

    Yes, I agree it’s a shame we can’t look more objectively but as I mentioned this isn’t really possible due to many factors.

    Two of the sites had simlar ad texts while one of the other sites had a slightly different message. While this could have affected the CTRs, there were some instances where sites 2 and/or 3 were below site 1 which meant that in some instances the only differences when looking at CTRs was position. As I’ve tried to show above, getting top spot is crucial to get your CTR up and eventually CPC down.

    So on your final point, site 3 did have lower QS due it being last to go live. However, we struggled to get improvements to both CTR and QS without really increasing CPCs (sometimes by up to 200%) to start to get any levels of clicks to help change the situation.

    As I tried to summarise at the end of my post, I think a tip for your readers is to push CPCs as high as possible to begin with to build up QS/CTR and look to adjust over time once you have got significant exposure & history on your account.

    Reply to this comment

  • Martin Macdonald 15th August 2012

    Im struggling with a few things in the post…

    I think its probably quite common knowledge that lower positions = higher CTR, generally speaking as well, conversions happen on the first two results, research happens through the rest of the result set, hence the propensity to see higher CR on top positions.

    That stuff is common sense.

    The data you’ve used though really doesn’t help draw any conclusions at all.

    There’s a shedload of variables you havent brought up which impact the conclusions (what about overall account history/individual campaign history/keyword history for starters?)…

    Just free styling here, but a much better test would be to split out your campaign 24 times, day part them by hour in adwords, then setup bidding rules based on positions to try and normalise them into specific positions.

    This would also help get you an accurate impression count as well, and you can do an actual like for like comparison on ad texts.

    I havent checked, but experiments for adwords would most likely be able to do that on a live account without affecting the rest of your campaign history.

    Comparing CTR’s (even) on ads with identical ad text, but different domains is pointless as well, it varies significantly.

    Another hole in the study, is the number of competing placements:

    In results where you have a three-box above organic, the ad in 4th spot has a traffic boost ABOVE 3rd, and still has the same kind of conversion rate advantage – making 4th the sweet spot in most industries. (but its not that easy to target).

    Additional to that, the actual organic result set makes a huge difference to the total share of clicks that paid gets – start mixing pictures and videos into the organic positions, and your PPC ctr collapses.

    Another factor not considered is the distribution of keywords searched within the ad copy, positioning it correctly on each line is another significant impact driver (Ive seen high increases CTR by ensuring a specific pattern is stuck to).

    Long story short:
    “Click Rate Deteriorates Depending on Ranking”
    <<<< True, but its far more complicated than that. It would be like me saying that "SEO rankings get better with more links", also true, but its only a tiny fraction of the whole story.

    Reply to this comment

  • Andy Headington 15th August 2012

    Thanks for your feedback Martin, you raise some interesting points here and I like your idea re the day parting as that would make our study more conclusive. This might be another blog post in the future!

    You are also very right that there are varying factors which affect CTR and that any one of these could impact and change the CPC/QS each of the ads. But my study is unique in that we’ve managed to get data from three competing sites all on the same exact match phrases which I’ve not seen anywhere else.

    Yes, adtext, domain names, sitelinks etc can all influence the performance of the campaign but the point I’ve tried to illustrate (which may have been lost) is that the top 1 / 2 positions really do dominate any PPC listings dramatically. Typically you will know that #3 might get X% CTR and if you up your bid to appear in #1 your CTR might go up but you can’t really predict how much until you are there and have spent your clients £££s. This snapshot shows how CTR drops dramatically from #1 to #3 for exactly the same phrases, albeit with variations on some of the adcopy.

    We’ve seen site 2 go up to position 1 this week on certain phrases and CTR has gone up by 300% with no other changs to adtext or entrants into the market.

    We all know that top positions get most of the clicks yes, but by how much isn’t always known which is what I’ve tried to show here.

    Reply to this comment

  • Martin Macdonald 15th August 2012

    Hi Andy,

    I guess a lot of my point is, you can spool up the same campaign with identical everything (apart from landing domain), stick it on 10 different accounts (you can have them all under the same MCC), and test this stuff properly (ie. with minimum variables) quite quickly.

    Of course, it helps if you have got, or can build, lots of sites in the same niche immediately…
    …but there is no law that states you have to own sites you send traffic to ;)

    Reply to this comment

  • sarvesh saini 27th January 2013

    Quite interesting and productive…. But I was searching for’how much does google pay us per click in cents or rupees….’

    Would be happy if u could answer that

    Reply to this comment

Subscribe To The Koozai Blog