Buying Google AdWords Ads on Brand Keywords?

Is Paid Search Incremental or Cannibalistic?

Earlier this month Google referenced a "study" they did which showed that 89% of AdWords ad clicks were incremental (meaning that they were clicks that the website would not have received if they relied on organic search results alone).

As part of that "study" they stated that "indirect navigation to the advertiser site is not considered." Why did they chose to exclude that segment of traffic? Because they advertiser would have got almost all of it anyway. They never really defined what indirect navigation is though, so you are left to guess as to what qualifies as being part of that segment.

The "study" also stated:

A low value of incremental ad clicks may occur when the paid and organic results are both similar and in close proximity to each other on the search results page. This increases the likelihood of a user clicking on an organic result as opposed to a paid result. Close proximity occurs when the ranking of the organic result is high, placing it near the paid results. Organic results triggered by branded search terms tend to have a higher ranking on average and this may lead to a low IAC value.

So which keywords should you advertise on? And which keywords are buying the milk when you already have the cow for free? According to Google:

A low IAC value does not necessarily suggest a pause in search advertising is in order. In fact, for many advertisers with a low IAC, it is still profitable to invest in search advertising. To evaluate the economic benefits of search advertising, an advertiser must run a calculation incorporating their individual IAC, conversion rates, and conversion revenue. The below equation can help determine whether search advertising is worthwhile on a case by case basis.

Is an Experiment Required?

Google later suggests that a more rigorous test would include a split test experiment that compares a control group against an ad group with paid search ads held back. They then suggest that "many advertisers are adverse to conducting such experiments due to the setup costs involved and the potential revenue impact from having a hold-out group."

What I don't buy *at all* is the suggestion that such studies need to be rigorous & expensive. On the organic search front, Google Webmaster Tools already offers organic search CTR stats by ranking position & ranked page:

And since Google is heavily promoting adoption of the +1 button, they also offer A/B split data for how that button impacts search performance.

If Google provides this data for free for organic search then why (other than protecting their own revenues) do they suggest this data is hard to attain for paid search? If Google respected their advertisers & wanted the advertisers to advertise based on complete data they would make this data available automatically, like they do with the +1 button data.

No "Study" Required

Here is my big problem I have with Google suggesting that I need a quantitative study to know if I should buy my own branded keywords:

  • I know I am going to get almost all the clicks anyhow (Google removed "indirect navigation" from their study for a reason, and 3rd party studies have shown how directly cannibalistic these ads are)
  • the whole point of building a brand is increased affinity with users & not needing to pay for incremental distribution driven by brand demand. To spend money to build brand only to have to keep rebuying the existing brand equity is quite a futile exercise.
  • in the bid auction Google sets arbitrary pricing floors at the keyword level to squeeze advertisers (almost nobody is bidding on "seo book" but if I do Google will want $2 or $3 per click)
  • even if I go through said "quantitative study" I end up needing to re-test it every so often as Google arbitrarily juices the ad prices to increase their revenues
  • when Google offers the enhanced long sitelinks they are doing so because they think the search query is primarily navigational, yet they still put ads above the organic search results, which IMHO is pretty dirty
  • and the dirtiest bit of it all (that smells the worst) is that competing against you in the ad auction is not only arbitrary pricing floors, but also Google itself, which buys keywords against your brand (using their own monopoly money)

Larger Sitelinks Drive Down Competition

Google recently expanded sitelinks in the organic search results to make them take up a huge portion of the above-the-fold screen real estate, driving down attempted organic search brand arbitrage & negative reputation issues.

Driving Down the Search Results

Using features like Google Instant, a Google+ promotional bar & longer AdWords ad copy, Google has been aggressively pushing down the search result set so fewer listings appear above the fold.

Each month there is another test of some new feature that pushes the organic search results downward.

Zero Moment of Burning Ad Budget

Google promotes a concept called "the zero moment of truth" suggesting that you need to advertise just about anywhere late in the conversion cycle to "be there" and reinforce your messaging.

However, with enhanced organic sitelinks, the brand owns so much of the search real estate that it will lose limited traffic to competitors if it doesn't buy AdWords on its branded keywords. Further, given the ability to block certain sitelinks & edit the page title & meta description you should be able to control the copy on your branded organic listing to make it look and feel like the ad copy you would use in AdWords.

There are some nuanced exceptions though, as brands are not always well aligned with how people search...

When You Should Buy Your Brand Keywords

Short Term Specials & Promotions

If you have an event coming up that you need to promote for a short duration of time then running AdWords ads is a great way to instantly get exposure for that campaign.

Certain Misspellings

In the past if you misspelled keywords Google would put the spell correction right at the top of the page. More recently they have decided to put it below the AdWords ads. So on this type of ad (where Amazon already ranks #1, but has the organic search results pushed down by the ad & then a spell suggestion) I think that ad is burning money.

In other cases, like where you don't rank high in the organic search results, buying AdWords ads on common misspellings is a much smarter idea. For instance, I think this is a smart ad buy by Agoda.

However, in the longrun, if I ran Agoda, I would point a few misspelled links at my website to boost my rankings for common misspellings.

One way to reach misspellings and longertailed searches for your brand is to use an embedded match, where you bid on agoda and then use -[agoda] as a negative keyword.

Brand is Shared By Multiple Companies

Mercedes Benz is burning a bit of their ad budget by advertising where they are irrelevant.

Certainly it makes sense for them to buy exposure for the branded keywords, but in the above examples they should put -kingston as a negative keyword.

When Google Runs Negative Ads

In some cases Google ads promote negative messaging. For instance, while using Gmail I saw an ad suggesting that I should "uninstall McAfee" in a computer that did not even have it.

Buying branded ads in those cases would likely make sense, if for no other reason to compete with & block out risky negative ads that could go viral. Whether Google should even allow such ads is another question for debate.

Big Money Markets Full of Spam

Google was recently clipped by the DOJ for a half-BILLION Dollars for running illicit ads promoting Canadian pharmacies. The DOJ went so far as highlighting that Larry Page knew what was going on & intentionally allowed these ads to run:

Mr. Neronha said those efforts amounted to "window-dressing," allowing Google to continue earning revenues from the allegedly illicit ad sales even as it professed to be taking action against them. Google employees helped undercover Justice Department agents in the sting operation evade controls designed to stop companies from advertising illegally, he said.

"Suffice it to say that this is not two or three rogue employees at the customer service level doing this on their own," Mr. Neronha said in an interview. "This was a corporate decision to engage in this conduct."

After the above instance, Google is perhaps going to be "guilty until proven innocent" where they are running sketchy ads.

In the short run it is likely appropriate to still run branded keyword ads while the issue is getting sorted out, but if you see anything like the following on your branded search results it probably makes sense to fight it on the public relations front in the background while opening the wallet to protect the brand publicly.

And since most major pharmaceutical corporations are routinely fined for running illegal ads, I don't understand why these pharma corps don't have a black hat SEO (or 3) on staff to help manage the search results.

If Google wants brand then give it to them in spades. ;)

Published: August 29, 2011 by Aaron Wall in pay per click search engines

Comments

JamieMac
August 29, 2011 - 6:38pm

Interesting post Aaron. My data is a bit dated (several years old) but I did pretty extensive testing on this subject with a Fortune 200 brand - actually did it with 3 of their brands. We set up fairly well controlled tests to gauge the effect on Clicks, CTR and most importantly sales and ROI when running adwords campaigns on branded keywords vs. not running the adwords campaigns. We even took into account email and display campaigns to normalize the data as much as possible. For all three brands there was a net incremental jump in clicks and esp sales of 15-30% depending upon the brand. Again, the data is old, but the results were consistent. The incremental increase far outweighed the canabalization. A lot of that was due to the conversion rate strength of branded keywords. Food for thought.

August 29, 2011 - 9:55pm

...wouldn't you expect that data to skew much lower in the current market (outside of perhaps on brands where the brand is a generic industry keyword...like credit cards for CreditCards.com)?

JamieMac
August 29, 2011 - 10:47pm

...but I still think that two factors will lead to net incremental sales gains by running a ppc campaign.

1. there are still xx% of people that prefer to click paid links for whatever reason
2. Branded keywords have a very high conversion rate.

This is relatively easy to test out (and ultimately that is the only way to know if it fits each campaign), but I think for major brands this is easy incremental money. From a purely PPC centric standpoint there are also other reasons to have a brand adgroup that has a relatively high CTR and conversion rate, but that is off topic to your post.

September 23, 2011 - 1:38am

...these ad units get before more merchants start to notice a parasitical nature to them? Google recently pushed out AdWords ad units with 10 sublinks in them! Oh, and the Interflora case in the EU is looking pretty ugly for them.

MANO RAME
September 7, 2011 - 7:42am

The eye view chart you posted here is very helpful. I also admire your SEO strategies. No wonder you have got so many pages indexed on Google. You certainly know your stuff.

mattcommins
September 13, 2011 - 6:02am

I recently launched a new adWords campaign for my company's brand name and there were zero search traffic for my branded terms (exact match) and my min bids were $1. I will provide this cavet, the account I used was brand new.

Bidding on branded terms allows advertisers, specifically in-house or agencies, to be more aggressive with non-branded keywords by blending the performance of both keyword types. I believe in bidding on branded terms is an essential branding initiative that is worth more than the costs you pay. That being said, every advertiser should test and see if bidding on branded terms is a viable option or not. I

Add new comment

(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.