vlexo, that's pretty dramatic, and obviously an important observation. It's also not the way that PPC and organic used to work together, though I'm sure there's always been a degree of cannibalization. At one point, there was a symbiotic relationship, where simultaneous top organic and top AdWords positions would augment each other, adding clicks in both categories.

I'm immediately wondering whether you can relate your observation to specific changes in serps display... in particular I'm thinking about the dropping of the AdWords display on the right... and also do you tie it to desktop vs mobile?

Additionally, would you clarify this comment...

...we're spending 3x as much on PPC for certain commercial keywords

When you say "3x as much on PPC", the question comes to mind, "3x as much on PPC" compared to when or compared to what?

Google's big study is a little dated now, but the headline was that 89% of the traffic generated by search ads is not replaced by organic clicks when ads are paused [googleresearch.blogspot.co.uk]. An obvious criticism would be that if the PPC keywords don't have a corresponding organic rank, that would be wholly expected.

eBay found the opposite in 2013 [conference.nber.org] (PDF) in what could be described as a more controlled study. They have the 'big brand' element, of course:

We find that new and infrequent users are positively influenced by ads but that existing loyal users whose purchasing behavior is not influenced by paid search account for most of the advertising expenses, resulting in average returns that are negative.

eBay suggested then that brand keywords were making a loss on PPC, and non-brand had a "statistically insignificant" effect on sales.Their methodology is something that might be of interest in your own analysis.

The quick summary is that the overall effect is difficult to measure, and is likely to be specific to your site, your audience, and which keywords you bid on for PPC and rank for in SEO, There are numerous other variables, of course. You might have great PPC ad text and mediocre organic listings, for instance.

In terms of your own data, is the CTR from Search Console information? That data itself can be problematic.

A site search for...
adwords organic cannibalization site:webmasterworld.com

...brings up, among others, this result, which includes the most recent comments I can find relating to the 89% study. It will include more quirks on methodology... and I'm assuming that everything is dated since the most recent SERPs display change and mobile.

Does Adwords / Organic Symbiosis work efficiently for you?
April, 2013
https://www.webmasterworld.com/google_adwords/4567262.htm [webmasterworld.com]

Whitey leads off with a reference to the study and to a post I'd made earlier about the symbiosis effect...

There's a discussion involving brand bias and the influences and behaviours of Adwords campaigns in conjunction with organic serps.

Does anyone have any experiences about the added performance efficiencies of running both in conjunction with each other. References to the Google research studies and a post by Robert Charlton are included here....

More about that link in a moment... but first, within the thread, this comment from eWhisper, our AdWords forum mod, which responds to the title question. There's also a follow-up post about the 89%, the most recent I can find (but still before the SERPs display change), dated July 15, 2013...

I'm a big fan of doing both at the same time.

I often see CTR & CR go up when you are twice (or more with universal results) on the page. There is some cannibalization, but in most cases, the 'free' traffic you're now buying is worth the additional traffic you get from paid.

It use to be really easy to do these tests yourself; but its much harder now with (not provided) as you can't see traffic volume on keywords for organic anymore to test the difference between buying and not buying certain words. You can use destination URLs as a bit of a proxy to see the traffic difference to particular URLs with and without the paid traffic, but that proxy isn't 100% accurate.

However, overall, I see a benefit 90%+ of the time.

The discussion Whitey asks about is found in this thread, and I believe in part talks about both studies Andy mentions...

Big brands do not have the upper hand - Matt Cutts
- copy and paste url into address bar -
https://www.webmasterworld.com/google/4554058-10-30.htm

As best as I can tell, Whitey is referring to a discussion starting with his post on Apr 21, 2013 and continuing on into a discussion about the 89% and beyond that....

On multiple results paid / organic, - brand folks I speak with that have very large accounts, used to hold the view that multiple results paid/organic were a cannibalization concern, which prompted my question....

Worth mentioning that eBay had a significant problem with how it conducted its AdWords campaign, which was so badly targeted that it was literally a joke, and destined to fail, making its study fairly meaningless. I'm sure there's a discussion of it somewhere in the forum.

@Robert Charlton: This test was conducted after the right hand side ads – to make this a like for like comparison – as much as we possibly can. I also picked keywords where we ranked in position 1 consistently throughout both a control period and the test period.

I haven’t looked into the impact mobile has when it comes to this, but that’s obviously a next step for me.

This would be 3 times as much versus last year, where our CTR for a certain keyword was significantly higher. However, we’ve switched off our PPC efforts to that of similar levels to last year to see what impact that would have this year. So we are comparing date ranges in 2016 and not 2015.

@Andy Langton: That’s really interesting, thanks for sharing. I’ll give it read for sure. I read the conclusion and this is what stood out to me as why the experiment they conducted might not be an accurate measurement:

“Other factors such as the ranking of the organic search result or the strength of brand awareness of the search term could influence the IAC estimate”

This is a pretty flawed test – like you point out. At the very least the test we’ve conducted had organic rankings, and meta data at the same points through the control and test period. The only difference was turning down our PPC efforts then calculating the impact. But if in their test a keyword is fluctuating quite a bit, then it could go either way. Plus the brand aspect for us is quite a large aspect. Yes, the CTR data is from Search Console, but we are comparing with the same data, so statistically, something is happening whenever we tone down our PPC efforts. The impact can also be felt via Google Analytics as we see an increase in traffic and compare our paid vs organic traffic levels on the days we do these tests.

This test was conducted after the right hand side ads...

So, that could well shift the percentages cited in the old studies.

As you'd mentioned in your first post....

We've done some calculations and actually, even though we are paying for more traffic, we are actually making a higher return as PPC are picking up additional traffic, outside of what we both would have received.

This is in line with what eWhisper had posted, and the question now, after the side results have been dropped, is how much the percentages might have changed (if you can even reliably determine that).

Also... here's a WebmasterWorld discussion I've found on the eBay study. I haven't read the study, but the thread is worth checking out...

EBay Study Questions Advertising On Google
March, 2013
https://www.webmasterworld.com/google_adwords/4554401.htm [webmasterworld.com]

Members posting were almost all critical about the quality of the eBay ads, feeling as I did that their low quality skewed the conclusions. One poster disagreed, feeling brand entered into it a lot more.

Buckworks posted ths excellent suggestion about managing campaigns which include client brands...

I like to add the client's brand name to the negative keywords for most ad groups, then I make a dedicated campaign to focus on their brand name. For my biggest client, their average CPC for the brand name campaign is two cents.

My question is will the PPC hurt your SEO efforts at the end? Because click through rate is a ranking signal for Google so in theory if you run a PPC campaign and your CTR for SEO drops from 30% to 20% Google might think you are not the right answer for that query so sooner or later you got the 2nd position organically instead of the first one.
Have you noticed anything like this?

Because, I have a client and no matter how hard we tried we couldn't get ranked on the top 4 for a certain keyword, but we used PPC to target that term, however the the four competitors we couldn't outrank haven't run any PPC.
So my suspicion was, the reason we can't outrank those four because of the PPC and as a user sees the same domain he/she clicked through PPC he just won't click again organically, this means lower organic CTR and this means lower ranking...

The problem is he didn't want a test period to stop PPC because that term made 60% of his profit and stopping the PPC campaign for 2-3 weeks for a test won't be the smartest idea for him. Probably it's a bit different problem but I hope it make sense to post here.

I too look after several brands in a large company group and can confirm that for several months we have noticed PPC spend rate affecting organic visits. I don't look too closely, as we find that other marketing activities, e.g. TV/radio ads, display ads, EDMs etc can send people to a search engine as a method of entry. They might click an ad or an organic result. Hence a search engine visit can be triggered by many factors.

I fail to see what the question is here. What you are seeing is clearly the result of cannibalization of organic traffic. The users who normally search for and find the organic results now see both the ad and the organic result and a portion of these users choose the ad instead of the organic result. The total impressions remains unchanged, because the same users still see the organic result, but since it is the same as the ad choose the ad. Before ad, 100 users, 30 click on organic result. After ad 100 users 10 click on ad 20 on organic result. Net difference zero!
What you are really paying for is hoping to influence the 70 users that don't normally click on your site, I assume that the ad drive more traffic to the site overall so clearly there is a benefit.

As to increasing engagement, from my experience as a user. I take companies more seriously if they are top of organic ranking and advertising, to me it says I want your business.

@ehorvath. The CTR should drop across the organic serp when ads are present, so there should be no significant change between the relative CTR from position 1 and position 2. say without ad the ctr is 30% and 20% then with the ads 10% CTR go to the ads, so position 1 and 2 go to 27% and 18% respectively, the CTR relative to position 1 and 2 remains mostly unchanged.

What you are seeing is clearly the result of cannibalization of organic traffic.

Not for me. I bid on phrases for which I am ranked #1 in Google. The effect is an increase in total traffic. I receive more total traffic bidding PPC on phrases I already rank for than if I hadn't.

My presence on the SERPs is amplified. I now have two chances to convince a user to visit. It's like owning positions 1 and 2 and effectively turning my competition's position 2 into a position 3. I bid on PPC to strengthen my position and to weaken/kill my competition.

My question is will the PPC hurt your SEO efforts at the end?

No, it does not. For years I have spent on PPC for phrases I already rank for and it hasn't hurt my SEO.

Because click through rate is a ranking signal for Google

CTR is not a ranking signal. It never was. You need to do some research about what a ranking signal actually is. Once you understand that, then go read actual scientific research on CTR (stay away from blogs, most SEO bloggers research their CTR articles from each other and end up repeating each others mistaken assumptions). Only after you've done real research will you then understand that CTR is not a ranking signal. I'm not going to link to anything, you have to do your own homework on this. I'm just setting you on the right path.

Good luck.

mb

@martinibuster,

What you are seeing is clearly the result of cannibalization of organic traffic.

I think you misunderstood my point, or maybe I misunderstood the ops point. What I meant by cannibalization is that since you now have two links in the serps, one ad and one organic, the CTR of the organic link drops, the reason is that some users that normally click on the organic link are now clicking on the ad. Clearly overall the there is increase total traffic as a result of the PPC, otherwise why would you be paying for it.

NickMNS, yes! We are in agreement. :)

I've run branding campaigns with PPC, and the reason for it was, as mentioned, to improve the brand name, and to scoop up extra traffic by having both a top organic listing and a PPC listing. As soon as we dropped the PPC, in tests, overall, we lost traffic to the site. Turn the PPC back on and the traffic went up. Typically, we saw a 10-15% increase when the PPC was on. The organic ranking remained constant throughout.
I was interested in CTR to the site, and then ROI. The ad spend was set based upon the budget, which, once set, didn't change much at all. In other words, we didn't end up spending more.

Personally, when I see a PPC and an organic listing click on the organic result.

Was the 10 to 15% increase on organic traffic only or to total traffic?

Was the 10 to 15% increase on organic traffic only or to total traffic?

Total traffic. Let me try and qualify that a little more. The figure was so close to 50-50 between the two and was tough to measure any closer than that. Also, dayparting was used so that visibility and, importantly, paid clicks were not wasted outside of the target times. The organic didn't get many referrers outside the core times: It was mostly robotic traffic.

The key thing was the 10-15% increase of quality traffic.

Note: this was prior to the loss of the right hand ads, although the ads, on the most part, appeared at the top, and not to the right, so the ads were in the prime positions.