We’re all familiar with the option to select Google’s Search Partners as an available network in addition to Google Search. However, how many of us actually break down our click data by each network to assess exactly what results we see from both?
As this blog may demonstrate, decisions you are making regarding clickthrough rate at an Ad group and keyword level could well be being made without seeing the full picture. Making decisions in PPC without fully having all data to hand, as we all know, is a dangerous and risky practice.
So who are Google’s search partners?
Finding a definitive list of all the websites that make up Google’s search partner network is quite difficult to find, and is something I’m sure Google are quite keen to keep close to their chest. However, we do know that sites such as Amazon and Ebay are part of this network, which gives a strong indication of how your adverts will display when this setting is selected.
With that in mind, how is advertising on search partners different to advertising on Google’s content network?
Well, to be honest, when those search partners tend to be websites such as Ebay and Amazon it’s not really that different at all, which is quite an alarming point. Users that search on websites such as these are looking for a different search experience than those that run their searches through Google’s search engine, which is only natural. A user has reached Ebay for a reason and expects to find what they are looking for there – they’re not expecting to be taken off somewhere else to make a purchase. As such, it would not be uncommon to see poor clickthrough rates from large amounts of search impressions on a site of this type.
Take the following search on Ebay for ‘televisions. The Google search partner adverts display at the bottom of the results page as follows:
Now, on the whole, there is nothing dramatically wrong with these adverts – they are relevant to the search on Ebay. However, to compare the relevance of the adverts here on Ebay, to how relevant they would be on Google’s search results page, it’s a non-comparison. Users are clearly searching initially with the aim of making a purchase on Ebay (perhaps even a second-hand television), and it will take some persuading to divert them off elsewhere to make a purchase. Clearly some users will be drawn in by these adverts to click and convert, and for that reason, there is definite value in advertising on Google’s search partners. On the whole though, as this again illustrates, an advertiser shouldn’t expect high clickthrough rates or even high conversion rates using this medium, but would expect to pay less per click than on Google’s search network (as you would expect when running a campaign using Google’s content network).
With that in mind, best PPC practice would say that a content network campaign should warrant its own separate campaign with alternative bids and adverts and managed separately due to the types of users and lower clickthrough rates you would expect with a content network campaign. Ideally, that is what you should be able to do with Google’s search partners. Unfortunately, that’s just not possible – it is either switched off altogether, or switched on in conjunction with the Google search network – there’s no middle ground.
The impact on campaign, ad group and keyword CTRs
This leads me onto the most important of this piece – if the data for the Google search and search partner networks are amalgamated (as is the standard setting in Adwords), surely the results are a little misleading to the untrained eye?
Well yes, they certainly are. The only way to overcome this is by choosing to segment the data by network each and every time you take a look at the data in your campaigns. But realistically, how many advertisers currently do this each and every time that make decisions regarding performance in their campaigns?
Take a look at some data segmented between the two networks from a campaign we are currently running:
If I hadn’t segmented the data, we would see that the campaign as a whole was achieving a clickthrough rate of 1.63%, 14 conversions at a conversion rate of 0.84%. On the whole, not disastrous, but does this really tell the full story?
When the data is segmented, it becomes apparent that the Google Search campaign is running at a far healthier CTR of 3.9%, with all conversions coming from this network for an again healthier conversion rate of 1.2%. Compare that to the Search Partners which has recorded a less respectable CTR of 0.7% and zero conversions – overall, it is this set of data that is bringing the whole campaign down.
Now consider if this data was not being viewed at campaign level, but at ad group or keyword level instead. If you were to see keywords that supposedly were recording thousands of search impressions, but very few clicks and extremely poor CTRs, logically, through fear of impact on quality score, you may have to make tough decisions as to whether to keep these terms in your campaign. This may not be the case at all – it could be that they’re performing extremely well on the Google search network, but poorly on search partners, and as a result, you could be unnecessarily removing profitable terms from your campaign.
The fear of impacting quality score is not something that needs to be considered as intently as for the search partners as you would with the Google search network. For search partner quality scores, Google state that “performance on the specific search partner is used only”. In other words, you acquire a quality score for each individual search partner site which are then allowed to perform mutual exclusively from each other. So fear not – if your clickthrough rate for a keyword on Google’s search partners is poor, it won’t bring down the performance of your quality score on the Google search network.
An additional factor that is far more concerning, is that it’s not even possible to segment the data at an advert level. With this being the case, how can you effectively test adverts? Again, you could be removing adverts that are showing poor clickthrough rates on the basis of affecting long-term quality score, when this really may not be the case.
Summary
This blog is not suggesting for a second that you should rush out to ditch Google’s search partners. The above data is just one example, and in many campaigns, Google search partners are responsible for healthy conversion figures. With that in mind, any medium that can supply conversions at your target CPA should not be disregarded foolishly.
What should be taken away from reading this piece is the message that it is incredibly important to ensure you have all the data available to you when making important decisions within your campaign. Without all the information to hand, it is impossible to make effective decisions – and effective decisions are the key to any successful PPC campaign.
Tip – check your search query report. If you see search queries that look like breadcrumbs (eg for the above example “Electrical: Tvs: LCD TVs” or something) thats often a sign that those impressions are coming from a non-search-focussed search partner (eg ebay) because google actually uses the product categorisation as the search term.
(“Hold on”, you may say, “surely thats content targeting not search targeting”, and you’d be right, but Google is never going to put huge volume strategic sites like ebay on the content network where its much easier to opt out, and ads are shown on every page, not just search results pages).
If you see you’re broad or phrase matching to these breadcrumb style terms, negatively exact matching them will stop your ads appearing on that page. You’ll likely find that they get loads of impressions and not many clicks, but remember its still worth doing because, unlike the content network, search partners CTR still impacts your Quality Score.
Great post – I am a big fan of keeping content advertising in a completely different campaign then search networks. Content networks are a different animal and can affect your overall campaigns performance. I.E CTR and Quality Score at the campaign level. Having it separated also provides you with a clearer and more precise data set.
Peter,
Incredibly insightful blog entry. One of my pet peeves with AdWords is that you cannot pause individual search partner sites. You either opt into all search partner sites or none at all. I’ve spoken with my Google rep about this matter and she tells me that it has been discussed internally.
There is a Google Analytics filter that shows you the referrer network for search partners (ie: the search partner sites that are sending traffic to your site. However, it is pretty useless since no action can be taken based on the data.
OldMatt – very good tip there. Search query reporting can be very insightful to help identify those search queries likely to have arrived via search partners. Quite often you can see quite diverse terms in search query reports with thousands of impressions that you just can’t envisage arriving via Google search – if these supply high clicks but low conversion rates above a targeted CPA, then I would agree on adding these in as negatives. As for the point about search partner CTR impacting on quality score, I’ve added a paragraph into my blog to cover this very important and valid point off – whilst search partners do impact on a Google quality score, it operates exclusively from a Google search quality score and so should be treated separately.
Justin – Thanks for the comment. It definitely is best keeping content network campaigns separate from your search campaigns. Not only will CTR be much lower on the content network, but the keywords and adverts used on the content network often need to be constructed completely differently to those used on the search network.
Matt Umbro – Thanks for the feedback. It would be interesting to see what further news your hear back from Google regarding this – it’s certainly something that needs to be looked into!
“For search partner quality scores, Google state that performance on the specific search partner is used only”.
Thanks for pointing this out Peter, that new paragraph is a great addition!
Google tries very hard to keep the AdWords system as simple as possible. While this makes the system easier to user for less experienced advertisers (the suckers who trust Google) it frustrates advanced users by not allowing them full control over their campaigns. Add in the fact that full transparency might expose some very important partners as crappy PPC partners and you have all kinds of reasons to keep that data abstracted.
Traffic of Search Partner’s are not of that quality as you can get from the google. I usually keep the campaign on the google search if the budget is very tight from client. But if client is not strict with the budget then I test the search partners as well. Results vary with the industry.I can say most of the time it gives lower ROI. But there are industries for which I was able to get high ROI then google even. I tried to find the reasons of that but never got some good clue.
Great post..The issue is that rules for yours ads to be display on the google partner network are not the same than ones on search. ie Ebay ads are mostly irrelevant.. So maybe it is better to be relevant and to have 100% control on Search only, work on your CTR’s and ROI..
You are right…this Google Adwords has changed Internet Marketing a long way..However,i do agree with Jawad Rafique,Google delivers the best traffic,better than any other!
Google hardly discloses anything about their search partners – which means clients like myself are kept in the dark a lot of the time. Our CTR has dropped significantly over our accounts and after trying to figure out ourselves what may have happened (checking negative matching, ad copy, bids) it was only when we asked Google to look into it that it was due to a new search partner being added. Google should inform us about these sort of things, even if they wont disclose who the actual search partners are.
I recently setup three test campaigns for a new client. For the first three days, the click-through % were about identical for all three companies (Google, Facebook, and Linkedin). Then an interesting thing happened on day three. The suggested bid amounts for Facebook and Linked in increased about 500% overnight. I chose not to adjust my PPC bid amounts, and within 48 hours, I was getting zero impressions and zero clicks form Linkedin and Facebook. At the same time, my Google rates actually went down about 8%, so I was able to lower my bid on Google accordingly.
There were no changes to any of the three accounts in an way that would have merited such increases. It looked like Facebook and Linkedin were simply trying to milk me for more advertising dollars, while they were providing terrible conversion rates anyhow. Google stayed the same, kept producing the same results, so I simply went in and paused both the Facebook and Linkedin accounts. I was not going to allow them to hold my advertising dollars hostage while they produced inferior results…on the contrary, I actually increased my Google budget to now include the monies that were originally set aside for Facebook and Linkedin.
Personally, I found that whole process extremely interesting. Now ten days later, both Facebook and Linkedin’s suggested budgets have increased even MORE. Googles has remained constant. Anyone else find this intriguing…?
Hi John,
We found the same thing when advertising on Facebook and LinkedIn. The suggested CPCs kept on increasing every day, even though the CTR/conversions were low. There was no explanation on the increase of CPCs, even though we knew the competition was minimal.
Thanks for this article! After reading it I investigated my own account and found similar results. My CTR for search partners was significantly lower. My CPC on search partners was higher and the ad rank was lower. I am thinking about running two campaigns now. One for Google search which has a promising CTR and a separate campaign for Google search and search partners. Perhaps trying different keywords and experimentation can yield better results with Google’s search partners.
I think a lot of the comments here are blending Search Partners and Content Network, which makes understanding the whole thing difficult for a newbie like myself. In short, you mention to keep an eye on this, and you say that Search Partners have their own quality score? With this in mind, then one would assume that it is worth keeping the Search Partners. However I am personally against it, because if I have a keyword that appears more often than another on Search Partners, then the blended CTR that I see in the Keywords report is completely skewed and I could be thinking that keyword A is performing better than keyword B but keyword B CTR could be negatively affected by a higher number of un-clicked impressions on the search partners network… Or am I missing something? Do you agree?
Thanks for a great post btw.
Hi Dan,
Thanks for your comment. It is frustrating that you can’t see the individual breakdown performance of Google Search v Search Partners at keyword level – this makes assessing accurate performances of keywords difficult with the CTR being blended. You can however break the data down at adgroup level by accessing the ‘Segment’ drop down box and choosing ‘Network’ to see exactly where your clicks and conversions are coming from. If you notice that you’re still getting a spread of conversions from Search Partners, then I would advise sticking with it, regardless of frustrations, as it won’t be negatively impacting your overall quality score, even if it does make it a little bit more difficult to optimise your campaign! In some cases, you could even move a single keyword into its own adgroup, and then when you breakdown at this level, you’ll have the individual keyword detail you need
Thanks,
Peter
I dont do much ppc at all, a friend of mine asked me to look at the campaign a company is running for them. Search partners was 526.76 in the display network, google search 136.07, and google search partners was 98.33. Over 600 was going to non google search, high numbers NO CALLS TO ACTION. But the seo company kept telling him look at all the hits your getting, its exposure blah blah blah. Oh Ya, google search CTR was .48, search partners .012 and automatic placements in display network was .06%.
So, Display Network and Search Patners are kind of the same thing?