Today I'd like to take a deep look inside the minds of the billion worldwide search engine users and share a more comprehensive understanding of their motivations. This pseudo-psychological perspective can help us gain perspective on targeting, click-through rates, ad-serving and even relative value.
First off, we need to realize that search engines are a tool - a resource driven by intent. The search box is fundamentally different than a visit to a bookmark like IHT.com or Reddit.com or moz.com; it's unique from a click on the "stumble" button in your StumbleUpon toolbar or a visit to your favorite blog - searches have a direct intent behind them; the user wants to find... something. That "something" is what I've segmented below:
Navigational Queries
Navigational searches are performed with the intent of surfing directly to a specific website. In some cases, the user may not know the exact URL, and the search engine serves as the "White Pages," passing along the (hopefully) correct location.
Informational Queries
Informational searches involve a huge range of queries from finding out the local weather to getting a map & directions to finding the name of Segey Brin's new bride or checking on just how long that trip to Mars really takes. The common thread here is that the searches are primarily non-commercial and non-transaction-oriented in nature; the information itself is the goal and no interaction beyond clicking and reading is required.
Commercial Investigation Queries
A commercial investigation search straddles the line between pure research and commercial intent. For example, sourcing potential partners for distribution of your new t-shirts in Albuquerque, determining what companies make laptop bags for sale in the United Kingdom or researching the best brand of digital cameras for an upcoming purchase all qualify. They're not directly transactional, and may never result in an exchange of goods, services or monies, but they're not purely informational either.
Transactional Queries
Transactional searches don't necessarily involve a credit card or wire transfer. Signing up for a free trial account at Cook's Illustrated, creating a Gmail account, paying a parking ticket from the city of San Francisco (I got one on my way out of town after getting robbed) or finding the best local Mexican cuisine (Carta de Oaxaca in case you're wondering) for dinner tonight are all transactional queries.
From the above segmentation, we can make a few determinations about the nature and value of the traffic driven by these various searches. Each presents opportunity, but not all are of the same quality or ROI for targeting purposes.
Navigational Queries:
- Opportunities - Pull searcher away from destination; get ancillary or investigatory traffic
- Average Value - Generally Low
Informational Queries:
- Opportunities - Brand searchers with positive impression of your site, information, company, etc; Attract inbound links; Receive attention from journalists/researchers; Potentially convert to sign-up or purchase
- Average Value - Middling
Commercial Investigation Queries:
- Opportunities - Convert to member/sign-up; Sway purchase decision; Collect email; Get user feedback/participation
- Average Value - High
Transactional Queries:
- Opportunities - Achieve transaction (financial or other)
- Average Value - Very High
When you're building keyword research charts for clients or on your own sites, it can be incredibly valuable to determine the intent of each of your primary keywords. For example:
|
Term |
Queries |
Intent |
$$ Value |
|
Beijing Airport |
980 |
Nav |
Low |
|
Hotels in Xi'an |
2644 |
Com Inv |
Mid |
|
7-Day China Tour Package |
127 |
Trans |
High |
|
Sichuan Jellyfish Recipe |
53 |
Info |
Low |
This type of analysis can help to determine where to place ads and where to concentrate content and links.
Hopefully, this excursion into the minds of searchers can help you to concentrate your efforts in the best possible places and think carefully about how to serve different kinds of searchers based on their individual intents.
BTW - I don't have data for it, but experience & intuition are telling me that Google & Ask skew to "informational" and "commercial investigation" searches while Yahoo! & MSN skew "navigational" and "transactional."
Great post Rand and thanks for the kind words Daniel (and you need to check out Daniel's blog) Here's some more stuff on these ideas taking from my Landing Page Optimization Trilogy ;) so hopefully everyone can make some money from them :)
Reference Intent
Transactional Intent
and some recent posts on intent:
Targeting the Two Intentions of Users
How Brand, Product & Community Influence Intention
Rave On!
Going from nothing more sceintific than watching how others search, I would suggest that (in the UK) at least, navigational searches are huge on Google. Because of teh fact that they have 4/5 of the search market over here, and people treate them as the new HTTP - I've lost count of the times that I've seen people type a url or brand into Google when I know that they already know the correct URL - its just how they browse these days....
A scary number of people don't even know the difference between the address bar and the google search box. Google is the internet.
It's also where a lot of commercial etc. searches are done as well though - it's just that no-one(*) in the UK uses anything else
The thing that always surprises me is people who have a different homepage (i.e. not google) but will then go to the address bar and laboriously type www.google.co.uk and then type www.whatever.com into the search box and click search. I'd have thought they'd try typing www.whatever.com into the address bar just once out of curiosity...
(*) pretty much, maybe a couple percent, but I think they either work for another SE or just accidentally set their homepage and now can't get Google back
Uh-huh - you said it girlfriend (or something!)
Haha. I haven't been called girlfriend in a while ;)
This is always a confounding variable..... they start typing thinking they are in the address bar but the web page drops the cursor in the search box... thus a search and at least one more click.
"Google is the internet."
Well, at least that's better than "AOL is the Internet"...
Your right! alot of people do that, its just easier because if you mis type it google will suggest the right one.
I totally agree with you...
I have have a few clients in the London theatre tickets market and I know you are right...
"just how they browse these days..."
I totally agree with you...
I have have a few clients in the London theatre tickets market and I know you are right...
"just how they browse these days..."
In addition to sweet graphics, this is a great post and really highlights what I feel puts SEOmoz ahead of the pack.
This article is like a whiteboard friday without the whiteboard, you really have a gift of putting information out in a very clear and precise manner. So many people in SEO seem to try to "out do" eachother on the whole being fancy smancy and impressive.
To have these concepts so simply explained here and to have insights avaliable that are so useful and timely is so impressive to me and I appreciate it alot.
This is an interesting way of looking at keyword building - having an explicit 'user intent rating' for the potential earning power of keywords could really optimise a campaign.
From a searcher's point of view, I have always found it irritating that when I want to research something (particularly when it comes to consumer hardware), most of the results are just trying to sell the item and actually provide little (if any) content about it.
I don't know if you've seen it, but Yahoo are developing an engine which tries to filter results based on user intent (it's called 'mindset'). When you perform a search, you can use a slider at the top of the page to adjust your intent on a scale from research to shopping. It's still in beta, so the results aren't spot on, but I think the idea is good - once the database has filled in a bit it will be something I'd use quite often.
https://mindset.research.yahoo.com/
Funny MindSet has been around for nearly 2 years and has failed to spark much interest....
Rand,
I gotta ask you: What do you use to make those nifty illustrations? Is it a Mac program?
Secondly, the segmentation and search intent are two things that Jonathan Mendez always has a sharp mind on. I think you've got a great primer here, but for those looking for more interesting ways to look at it, I suggest all y'all check out his website:
https://www.optimizeandprophesize.com/
His latest post focuses on integrating community to the overall search intent picture.
I've had my own take on the idea of search intent and matching that to content offering:
"Acting on this information for PPC and keyword bid management is easy, but this user behavior applies to SEO as well, especially in the need for positioning content for each “step” of the conversion process (from non-branded to branded)."
Daniel - I'm embarassed to say that it's Macromedia Flash. In the '90's I used to be a Flash web designer, so my ability to use the program for graphics remains.
And yes - I, too am a big fan of Jon Mendez. He's actually going to be helping us with conversion rates on SEOmoz!
Good post. I'm digging into link building right now, and you had some good points.
Very interesting post, a good rule of thumb when evaluating kw costs...thanks!
Great post!
Thank you randfish!
I've just posted an Italian version of your study in my (italian) Blog. I've also used your pictures.
I like the breakdown of types of searches, because you can not only optimize your ad campaigns, but you can then also think about the content on your website and how it can appeal to the different types of searching. Thanks!
Hi there,
Informational queries have the lowest value for the direct search->conversion model. That, is a fect!.
but now a days the playground is getting so interesting that i i reckon informational queries will become a huge asset brands will fight fore, as the best way to gain awareness and higher profit from online efforts is to generate a relationship with the customers.
Informational queries are the first step to build these sort of relationship.
So, while they still should be considered of low valuem,, they must be in a long tail strategy, aligned with many other on-page factors.. not trying to sell, but to become valuable as an authority and source of knowledge for prospects.
considering any business is an "expert" within its own niche industry... all could aim for this.
.
This is still an excellent article 2 years on - though people are getting better at searching, the different 'intents' remain the same.
Interesting post Rand. I think the value of each segment might change some based on the goals of the site. For example a publisher who's revnue comes from advertising is probably going to place more value on informational searches than the typical ecommerce site.
I'd also suggest that even navigational searches can have higher value depending on the search. I know if I were the Beijing Airport I'd want to be very visible on in the results for the term. I see it as a branding issue where someone knows they want you so you'd better be there.
But I'd agree that the direct monetary value of both those segments will be less than the other two segments.
Great post. We're working a lot with implicit user self-segmenting and natural audience segmentation clusters at Collarity. We have a recent blog post that discusses our methodology for distilling search and content discovery relevance from anonymous implicit actions here.
A natural byproduct of the Collarity platform is a rich detailed picture of the natural customer communities (segments) that live on a website. Collarity identifies specific “hotspots” of interest based on the measured associations of a web site's visitors with content. These hotspots form the basis for organizing search and recommendation relevance by implicit communities of like-minded users and/or subject matter experts.
From the Desk of Gang0rraZ
The Internet Monday, 11:45 AM
Dear Rand,
Nice article and illustrations but I hate to say this. Nowhere in your article you mention how we can actually make money with this information.
Do you plan a follow up article to address this very important point?Your fan, Gang0rraz
This must be why Shoemoney has 2X the subscribers that we do :)
Gango - there's no concrete answer to your question. I believe it can help you to understand the world of search and the practice of targeting a bit better, but putting a dollar value on that knowledge is almost certainly impossible.
Yeah he buys them.....
Shoe I was only joking... but if the price is right I know you would be buying them!
HiGang0rraZ,
As an economist and keyword researcher I do have some suggestions how you can find the "money" in your keyword report.
As Rand says there is no definite way to find valuable keywords, but your keyword report is able to give some suggestions.
Which keywords have the highest market value?
To get that information you can sort your keyword after highest bid prices.
To get the value you can multiply the bid prices with the number of searcher per day. Then you have the market value per keyword per day.
Unless you are selling advertising, you might not be interested in the market value, but like to know the value for you.
You can estimate the value for you, by multiplying the calculated value with how relevant an averages searcher will find your site. (ex. this keyword phrases is worth 100 USD, but only half of the searchers will be interested in your website, therefore the keyword is only worth 50 USD per day (100*50%=50) for you if you get all the traffic (If you have a conversion guess, It might be a more précis estimate).
To sort after value gives a list of very relevant keywords.
These are 3 simple methods to select keyword based on value. But if you are advertising or search optimizing you might want to find keyword that the market has underestimated or not found yet.
One method is to locate the highest valued keywords and then go though your keyword report to find keywords with similar goal and intent.
I hope this helped. I can offer more information about how to get value from your keyword report here and offer sample keyword reports here.
This method of calculated value isn’t very précis, therefore think of it as an indicator. The two variables: Bid prices and the number of searcher are not that precise and therefore the value estimate won’t be precise either.
I like this article, because I can tailor it more towards what users are looking for. I sometimes had more of a "throw mud against the wall and pick whatever sticks" mentality of keywords. This will definitely help alot.
So I have a new way of looking at site structure. I can do my keyword research and classified them by user intention, that way I know exactly what type of searcher am I targeting on each page. Thanks.
Indeed, very good article. I like your illustrations! Great job
Very solid concepts... reminds me of my product management/development days in that your product may be used differently by different people.
Same ideas comes through here for keyword research and site structure/information presentation.
From a keyword research aspect alone, this really helps categorize search phrases with a whole other dimension, showing that just determining some level of relevancy is only part of the challenge... not just a level of how relevant a phrase is but in what way.
Kind of oldish piece of content, but still makes perfect sense :) Special thanks for the idea of using a table with queries, intent and value. I've been having this problem with some of my clients while suggesting keywords that had less searches because they were transactional as opposed to informational ones. It's a great article to refer clients to.
Great Article! Makes you think alot about how much power a se has!
There's a pretty good discussion of these sorts of query-type categories, among other related things, in Search Engine Marketing, Inc., by Mike Moran and Bill Hunt. Good stuff. :-)