Are Search Queries Becoming Even More Unique? Statistics from Google
May 24, 2010
One of my favorite numbers that Google published in November, 2008 is:
“20% of the queries Google receives each day are ones we haven’t seen in at least 90 days, if at all?” – source.
That number has stood alone for for quite some time being one of the defining numbers of just how unique keyword queries can be.
I just got permission to publish a new number that Google included in a presentation they sent me:
20% of searches each day are new or haven’t been completed in the last 6 months
Now, it is possible that these queries were that unique all along; but think about that number. 1 in 5 searches conducted on Google either haven’t been done before, or at least haven’t been conducted in the last six months. That’s strengthens the usefulness of the new match type: modified broad match.
In the same presentation there is another stunning stat:
70% of queries have no exact-matched keywords
PPC marketers often talk about the usefulness of exact versus broad match, and I believe many of them. It’s not the lack of marketers using exact match as much as the large variety of search queries that occur. This stat does look at all queries and doesn’t just examine commercial queries. However, regardless of your interpretation – it’s a big number.
According to Hitwise (March 2010 stats) approximately 65% of queries are one to three words; which means roughly 35% of queries are longer than three words. According to the last stat from this Google slide:
54.5% of user queries are greater than 3 words
There is a large discrepancy between Hitwise’s 35% and Google’s 54.5% number. These numbers are based upon Google US Internal data – so I’d expect them to be accurate for Google.com. However, it is possible that Yahoo and Bing see shorter queries making Hitwise’s numbers accurate across all search properties.
If you want to see the slide, here’s an image:

Google AdWords Remarketing Campaigns: See how we set up our own campaigns
May 21, 2010
Google remarketing campaigns allow you to serve custom ads to users who have visited your website.
My latest Search Engine Land article covers the basics of setting up a remarketing campaign and gives some examples for setting up remarketing ad groups for an ecommerce site. If you are new to remarketing, you might want to take a quick look through that article and learn the basics of how to set up a remarketing campaign as in this article we’re going to make the assumption that you have some basic knowledge of remarketing.
In this article, we will show you how we are setting them up for our new product, Certified Knowledge (note: the site is not fully live yet, but we’re planning out our marketing campaigns now so this is what it will look like once everything is publicly launched).
The Marketing Campaign’s Objectives
Certified Knowledge is a subscription based AdWords learning, PPC tool, and community site. With many subscription products it takes a few visits for someone to finally convert. Therefore, remarketing is a perfect way to serve custom ads to those who have been on the site once, but have not yet bought a subscription.
There are three main benefits of Certified Knowledge (tutorials, tools, and community); therefore, we want to make custom ads based upon what sections someone has visited within the site.
If someone entered the cart but did not buy, we want a very custom message. For example, if a user started the purchase process but did not complete the transaction; we want to display a custom message or offer specific to the fact that they didn’t finish the purchase process. This group of cart abandoners still have a better chance of converting than someone who has never been to our website before or looked around the site and didn’t enter the cart, so our bids for this list will be higher than for any other audience.
If someone has subscribed we don’t want to serve then an ad for the membership.
Lastly, the further into the conversion process that someone was, the longer the cookie will last based upon our settings (remarketing ads are displayed based upon a browser having a cookie that corresponds to your lists).
The Remarketing Lists
Google offers the ability to create a straightforward list (a cookie on the computer means they are in a list), or a custom combination which uses Boolean strings to combine audiences into a single list. With some of these lists, I think we could use custom combinations instead of straight lists. However, I’ve not been able to test Google’s Boolean strings yet to be confident in their use. What I’d like to do is add someone to list 1. Then, if they also make it to list 2, remove them from list 2. However, that is not currently possible.
In addition, what I don’t know is if someone is on two different lists; and you use one list as a negative list, it appears that the ad will not be displayed to that person at all. (i.e. if someone is on list a and on list b, and list a is your positive list and list b is your negative list – will the user see the ad?)
Therefore, for the initial setup of our remarketing campaigns (this may change in the future); I’m setting up several lists; and then I’ll use negative lists at the ad group level and CPC manipulation to determine which shows (i.e. if a basic list is worth $0.25 and a more advanced list is worth $1, even if someone is on both, the should see the $1 CPC ad copy for that list).
Here is a list of the remarketing lists that we created and where we put the codes across our website:
| List Name | Purpose | Placed | Cookie Duration |
| All visitors | Reach everyone who examined in the offer | Global Footer | 30 days |
| PPC Tools | Reach those who examined the Tools section | PPC Tools pages | 60 days |
| AdWords Tutorials | Reach those who entered the tutorial section | AdWords Video page | 60 days |
| PPC Community | Reach those looking for community engagement | PPC community page | 60 days |
| Pricing | Reach price conscious shoppers | Visited price page | 60 days |
| Shopping cart abandonment | Reach consumers who showed intent to buy but did not finish | All pages except confirmation of cart | 90 days |
| Converted: Free email training | Free 5 email training message for sneak peak. Reach these consumers who showed high interest in system. | Thank you for subscribing email confirmation page | 90 days |
| Converted: Subscription | If someone subscribed, want to make sure we don’t serve them ads | Thank you for subscribing page after billing info entered | 360 days |
Once the lists are created and the scripts placed on the appropriate pages, then it’s time to create the Ad Groups that reach the various audiences.
Remarketing Ad Groups
We are creating an entirely new campaign for our remarketing lists for four reasons:
- Budget. We have a much higher budget than for our placement and discovery campaigns.
- Blocked domains. We don’t want our usual blocked domain list to affect remarketing ads (at least to start)
- Geography: As this product has international appeal (and we’ll do international billing); we are going to start with a few different countries in our campaign list (and refine from there).
- Reach & Frequency: We want a different frequency caps for the ads than our other content campaigns.
Once a new campaign is setup, it’s time to map out the ad groups. The reason to create various remarketing ad groups is when you want to show different ads to different audience lists. At an ad group level, you can add both positive and negative audiences and set a different bid by audience. Therefore, we’re going to build our ad groups from the ads that we’re going to be showing.
In addition, it will be common for some users to be on multiple lists. Therefore, we’re going to use a lot of negative lists to ensure the most relevant ad is being displayed. In cases where someone is on multiple lists, we will also use bids to ensure the more specific ad is being shown (so our bids will be the lowest on general ads and highest on close-to-converting individuals).
Ad Group 1: Non-Engaged Users
This ad group is for users who visited the site but did not enter a specific section of the site. Therefore, these are the absolutely least qualified of all the visitors. This is the reason the cookie duration for this list is only 30 days, the shortest of all the lists.
The positive list: All visitor
The negative lists: All the other lists. If someone is on any other list; we don’t want them seeing our general ad; we want a more specific ad served to that user.
The ad: Showcasing all benefits.
Bid: Lowest of all the remarketing lists. These visitors went to at least the homepage, and possibly an interior page; but not one of the very specific pages that is useful for tailoring a separate ad for that user.
Ad Group 2: Price Shoppers
This ad group is for users who visited the pricing section.
The positive list: Pricing
The negative lists: All others except non-engaged users list.
The ad: Showcasing that your time is worth money, and then show how Certified Knowledge can save you time.
Bid: Second lowest of all the remarketing lists. Price conscious users can be difficult to convince. In addition, I’d rather the more specific ad (the ad groups below) be shown to consumers instead of this ad. If someone happens to be on both the non-engaged user list and the price shoppers list, they should see this ad because its bid will be higher than the non-engaged user list.
Ad Group 3: PPC Community
This ad group is for people who visited the PPC community page. I expect that most people who visit this page will also visit other pages across the site. Therefore, while this will be an ad we show, we’re going to put less importance on what I expect the stronger benefits to be: Tools & Training. However, if someone did visit this page, we’re willing to show them this ad (in addition to other ads they may see from the more specific ad groups).
The positive list: PPC community
The negative list: All others except Price Shoppers and Non-engaged users.
The ad: Showcasing the benefits of having access to an active forum, email, and news system.
Bid: Third lowest. Higher than Price Shoppers; lower than PPC Tools & AdWords Tutorials.
Ad Group 4: AdWords Tutorials
This ad group is targeting consumers who examined our tutorials. As we’re going to have many pages explaining the tutorials, this might be broken down into two lists over time: main tutorials page, all training subpages. That sectioning will help us identify those who just glanced at the tutorials versus looked at our videos and read more about the AdWords lessons.
The positive list: AdWords Tutorials
The negative list: Converted subscription, Converted email training, Shopping cart abandonment. If someone converted, or was closer to converting than just visiting the features pages, then I want them to see the more specific ad.
The ad: Showcasing what Certified Knowledge can teach you about AdWords with more than 50 lessons, and more than 100 coming by the end of the year.
Bid: Higher than PPC Community and lower than the three negative lists.
Ad Group 5: PPC Tools
This ad group is targeting consumers who examined our toolset. As we’re going to have many pages explaining the tools, this might be broken down into two lists over time: main tools page, all tool subpages. That sectioning will help us identify those who just glanced at the tools versus looked at our videos and read more about the tools.
The positive list: PPC Tools
The negative list: Converted subscription, Converted email training, Shopping cart abandonment. If someone converted, or was closer to converting than just visiting the features pages, then I want them to see the more specific ad.
The ad: Showcasing the benefits of the PPC tools we have developed which will save you time in creating ad copies, geographic keywords and keywords. They will help you bid, find quality score issues, find broken links, and help you analyze your site. While most of the ad groups will contain two or three image ads to start (by themes, each theme will have different sizes); I expect this ad group will have more test ads to see what message resonates better.
Bid: Same as the AdWords Tutorials. I expect it will become slightly higher than the tutorials over time as tutorials are one to three time views for most consumers, and many repeat views will help refresh your knowledge. The tools will be used daily, weekly, or monthly depending on what you’re trying to accomplish inside your account; and therefore, I think will be more valuable over time.
Ad Group 6: Converted Free Email Training
This ad group is for targeting those who signed up for a ‘free lessons training’ via email. Many people who sign up for free training do not have an intent to buy as they just want the free stuff. Others want a trial of what you are offering and have a high chance of converting. However, email open rates can be sporadic on free training offers. Therefore, these ads will both remind them about Certified Knowledge, but also serve as a reminder about the emails delivered to them.
The positive list: Converted Free Email Training
The negative lists: Converted subscription, Shopping Cart Abandoners
The ad: Showcase the Certified Knowledge tutorials with messages to remind them about their email subscription. I’m not sure if they will be something like: “Did you enjoy your free email training from Certified Knowledge? Learn how you can access all of our training” as that might seem a bit creepy to people that we know they have the email – or if they’ll be images more aligned with the training aspects of Certified Knowledge.
Bid: To start, a bit higher than AdWords tutorials & tools lists as these users were willing to give us a name and email address to receive the information. Overtime, the bidding for this list might change significantly up or down.
Ad Group 7: Shopping Cart Abandoners
This ad group is for targeting those who entered the shopping cart but did not convert.
The positive list: Shopping cart abandoners
The negative list: Converted subscription, Converted Free email training
I’m making an assumption here that I might change in the future. I’m considering these users more valuable than the converted free email training list as these shoppers were within a submit button of entering their credit card information and becoming customers. I think they will be more valuable than the Converted Free Email Training list – but I might be wrong; therefore, I might test this exact same ad group minus the email training list as a negative list in the future.
The ad: One ad will showcase the Certified Knowledge benefits. One will show a phone number with the benefits that talks about support for any problems. As we start to map out sign-up objections, then we will base the ads around the most common ones.
Bid: The highest of all the lists.
Are all these lists necessary?
Outside of the time to create the actual ads per ad group, setting up one list and one ad group versus seven ad groups and lists is only a few minutes of additional work. We’re using WordPress for this site’s CMS; so adding some custom code to the pages is very simple and will take less than a minute per list. I’m expecting the entire setup (outside of the ad copy creation) to take less than ten minutes. I’ve already set up some remarketing campaigns for the AdWords Seminars and it took less than a minute to setup the list and add the code to the pages for each created list.
The advantage of showing an ad to the consumer based upon their main interests in the site should significantly outweigh the work involved.
The Next Step
Once everything is in place, then remarketing success is all about ad creatives and measurement.
Remarketing has a higher potential of success than almost all other forms of marketing because the user was already involved with your website. Most marketing drives some users to your website based upon a common interest (keyword, placement, email list, etc). Remarketing engages users who have already been on your website and are somewhat familiar with your offerings. This type of campaign can be the final push that someone needs to realize how valuable your service can be to them.
Google’s New Match Type Now Live – Modified Broad Match
May 11, 2010
I was lucky enough to be in the beta of modified broad match and have been using this match type for a couple months now to great success. Google lifted the veil of secrecy today about the new match type and let me know that I can now blog about it.
Broad match increases reach. Phrase match increases relevancy. The new modified broad match gives you the flexibility of broad match with the control of phrase match.
Broad match is useful because: “20% of the queries Google receives each day are ones we haven’t seen in at least 90 days, if at all?” – source.
While the fact that so many queries are unique often led people to using broad match – the returns often aren’t there. This new match type gives you some control over how a broad matched word can be matched.
To use this match type, go to your account and add a + (plus) symbol in front of one or more words in your keyword phrase. Then, the word/s with the +sign must either be in the query or a close variant must be in the query. A close variant is a misspelling (flor instead of floor), plural (flowers instead of flower), or stemmed version (running instead of run).
| Keyword | Possible Matches | Notes |
| running +shoes | running shoes running shoe tennis shoes |
The word +shoes or it’s variant ‘shoe’ is in every query |
| +running +shoe | running shoes running shoe run shoe shoes for running |
Both +shoe and +running must be in the query or have a variant in the query |
| +extra +wide running +shoes | Extra wide running shoes Extra wide exercise shoes Extra wide walking shoes |
All the words are matched or closely matched except for ‘running’ |
The use of the new modified broad match will help expand your possible matches while still keeping those same matches under control. Broad match and negative keywords do work well together, and this new match type will open up some new possibilities for broad and negative match combinations. Just remember that these new matches will still not convert higher than your exact match keywords.
If you wish to try this out, I’d suggest picking a few select ad groups where you are having problems gaining the exposure you desire, and then following these steps:
- Create a new ad group using those same keywords with the new plus modifiers
- If the old ad group has all broad match, then set a higher CPC for these new match types
- If the old ad group has all exact and phrase match, then set a lower CPC for these new match types
- Let the ad group run and collect some data
- Run the search query report examining these two ad groups and their variations
- Set appropriate bids based upon conversions
The reason to create a new ad group in this example is that you can only see search query data at the ad group level. While you can see the match type; you cannot see the keyword and match type combination that triggered a query. Over time, you might get rid of one of these ad groups and combine the keywords back into a single ad group. However, this is a new match type and there are bound to be some odd combinations that you will be shown for.
When trying out a brand new function with AdWords, it can be useful to look at the new features in isolation from other variables.
New Videos for the week of 4/19/2010
April 29, 2010
Here are the new videos for this week. None of the videos are must watch videos. If you are new to AdWords, the basic videos about the new ads are worth taking a peak at:
However, if you want to watch some videos, now is the time to take a look at the At Google Talks videos. These are generally longer videos (40 minutes plus) but they are given by experts and authors in the field at Google’s campus and then recorded for public consumption and generally contain excellent content.
New Videos from lifeatgoogle
New Videos from AtGoogleTalks
- Leading@Google: Brett Johnson
- Leading@Google: Annie McKee
- Leading@Google: Michael Feiner
- Authors@Google: Peter Nowak
- Authors@Google: Steven Weinberg
- Leading@Google: Mike Useem
New Videos from GoogleWebmasterHelp
- How many search algorithm changes were made in 2009?
- How does Google rank sites which run on non-standard ports?
- Can I exclude certain words from Googlebot?
- Should I spend time on meta keywords tags?
New Videos from SESConferenceExpo
- Submit Express exhibiting at SES New York 2010
- SiteCore and Web content management at SES New York 2010
- PR professionals can use search to bolster their objectives at SES New York 2010
- Mobile technology and baby marketing at SES New York 2010
New Videos from GoogleGrants
New Videos from GoogleTechTalks
- MariaDB, the Backward Compatible Branch of MySQL(R) Database Server
- Chemistry on the Web: How Can we Crowdsource Chemistry to Solve Important Problems?
New Videos from FastForward
- In the next three years what do you predict will be the biggest marketing changes for brands?
- What is the most important data to evaluate when it comes to online?
- What 3 things are working with new media, like social or mobile?
- What is the value of co-op marketing and have you seen it work?
- How are you integrating offline and online marketing campaigns?
- What is your vision for the future of eCommerce in the next three years?
- Marketing POVs for CPG
- How has the recession affected word of mouth marketing?
New Videos from GoogleBusiness
- AdWords New Ad Models: Changing the way you think of search ads
- AdWords Ad Extensions: Making a good thing even better
- The AdWords New Ad Formats Initiative: Google’s Next Chapter in Search Advertising
Out Of Keyword Ideas? These 7 Tools Will Help
April 27, 2010
My latest search engine land column is out titled Out Of Keyword Ideas? These 7 Tools Will Help.
It takes a look at some tools which can help jumpstart your thought processes about keyword research.
The tools covered are:
- Microsoft adCenter labs search funnels
- Xenu Link Sleuth
- Google & Microsoft Search Query Reports
- Search Based Keyword Tool
- The thesaurus
- Web Seer
- Wordtracker Labs: Keyword Questions
My two favorite ways to jumpstart keyword research before going down to these tools are:
When those two methods fail me, then the above tools take over to help get me started on new ideas.





ShareThis





