Jump to content

  • Log in with Facebook Log in with Twitter Log In with Google      Sign In   
  • Create Account

Subscribe to HRA Now!

 



Are you a Google Analytics enthusiast?

Share and download Custom Google Analytics Reports, dashboards and advanced segments--for FREE! 

 



 

 www.CustomReportSharing.com 

From the folks who brought you High Rankings!



Photo

Search Engine With Emphasis On Psychology


  • Please log in to reply
13 replies to this topic

#1 greggb

greggb

    HR 4

  • Active Members
  • PipPipPipPip
  • 118 posts
  • Location:Oregon

Posted 24 September 2009 - 06:24 PM

Hi all. I've always been fascinated by search technology. I've been thinking for a while now about a different approach to search, which would be fundamentally different than the technology used by today's major search engines. The idea includes a lot of psychology.

First of all, IMO one of the biggest problems with today's search technology is the fact that result rankings can be manipulated by human effort. Coupled with the fact that information is being pumped into the internet at an unnatural rate, by people who are making money on the content they're pumping in, what you get is a collection of now over 1 trillion indexed pages, many of which weren't created for natural reasons, many of which don't have natural rankings (because efforts have been made to increase their rankings). The internet is becoming less and less natural, which is a topic for many other discussions.

The point is that the internet is becoming less natural. I believe this is causing in increasing challenging in providing relevant results. I believe a solution may be to include a strong psychological element in a new search technology.

Every piece of content on the internet was created for a reason, and with a motive by its creator. What if there was a way to determine what that motive was?

Also, consider the fact that every person who uses a search engine is looking for content of a particular nature. For example, they may be looking for purely factual information, or they may be looking for information that is very time sensitive (i.e., important now). Let's look past the nature of the content they're looking for, and theorize that content of a certain nature was created with a certain motive. For example, some people want to (or it's their job to) share what is believed to be factual information.

I could go on, but what it comes down to is matching motive to motive. For example, I need information that was created with the pure intention of sharing factual information, so I'm provided with information whose author sincerely intended to share factual information.

Rankings for a piece of content could now be more determined by its psychological nature than by its keyword placement (though keywords would certainly have to be a factor) or inbound link text. The result is a search engine less manipulated by human effort, and more natural as a result.

So, the question is, do you know if anything like this exists already?

Also feel free to share any of your general thoughts about this idea.

I've already begun work on this and so far I'm learning some really interesting things.

Thanks,
Gregg

PS I posted this in another forum more related to this topic but didn't get a lot of feedback, so I'm reposting here. I hope that's OK.

#2 Jill

Jill

    Recovering SEO

  • Admin
  • 33,004 posts

Posted 24 September 2009 - 06:55 PM

QUOTE
So, the question is, do you know if anything like this exists already?


It doesn't quite exist yet, but it's definitely the way Google is moving and would love to eventually become.

#3 Randy

Randy

    Convert Me!

  • Moderator
  • 17,540 posts

Posted 24 September 2009 - 07:22 PM

Well, I'd have to disagree with the general premise that everything else is being based off from. This general premise being:

QUOTE
The point is that the internet is becoming less natural.


If you had written the statement anywhere from two to 10 years ago, I'd have agreed. But with the veritable explosion of blogs, video, podcasts and other social media platforms in the last couple of years, I'd have to disagree. How many people have a MySpace or Twitter or Facebook account --to name just three of these platforms-- but aren't trying to make money online? How many people are blogging on their own domain or on blogger, etc who have no expectation of earning a living online?

Lots more than a few years ago.

And for those who aren't creating their own content, how many people search and partake of YouTube videos that have nothing to do with e-commerce? Did you know YouTube is the second most used search feature on the 'Net?

As a primarily e-commerce webmaster I'd agree that competition is definitely getting stiffer. But more often than now these days it's not someone trying to earn a living online.

As to the rest, I agree with Jill that Google is certainly moving in the direction of trying to match up the user's motivation with the results they deliver from a search. This is what the whole personalization of search movement is about. It's a very difficult thing to do though, as you might imagine. Especially when they're trying to define motivations from both sides of the equation is something that's very, very difficult to key in on. Doubly so when they're trying to do it with exceptionally incomplete information.

They're working at it though, probably hoping that a 70% success rate is better than the status quo.

#4 greggb

greggb

    HR 4

  • Active Members
  • PipPipPipPip
  • 118 posts
  • Location:Oregon

Posted 25 September 2009 - 10:20 AM

QUOTE(Randy @ Sep 24 2009, 08:22 PM) View Post
Well, I'd have to disagree with the general premise that everything else is being based off from. This general premise being:
If you had written the statement anywhere from two to 10 years ago, I'd have agreed. But with the veritable explosion of blogs, video, podcasts and other social media platforms in the last couple of years, I'd have to disagree. How many people have a MySpace or Twitter or Facebook account --to name just three of these platforms-- but aren't trying to make money online? How many people are blogging on their own domain or on blogger, etc who have no expectation of earning a living online?

Lots more than a few years ago.

And for those who aren't creating their own content, how many people search and partake of YouTube videos that have nothing to do with e-commerce? Did you know YouTube is the second most used search feature on the 'Net?

As a primarily e-commerce webmaster I'd agree that competition is definitely getting stiffer. But more often than now these days it's not someone trying to earn a living online.

As to the rest, I agree with Jill that Google is certainly moving in the direction of trying to match up the user's motivation with the results they deliver from a search. This is what the whole personalization of search movement is about. It's a very difficult thing to do though, as you might imagine. Especially when they're trying to define motivations from both sides of the equation is something that's very, very difficult to key in on. Doubly so when they're trying to do it with exceptionally incomplete information.

They're working at it though, probably hoping that a 70% success rate is better than the status quo.


I'd agree that social websites tend to result in more natural content, but the majority of the content being created at social sites isn't something most people really care about. What would be natural? For someone to have a conversation with a few friends, to mention their trip last weekend, or what their kid said yesterday. This isn't going to be competition with the content the majority of people are looking for while using search engines.

You look at the people making money online in affiliate programs, PPC programs like Adsense, even pay-for-content sites like Associated Content, and what are they doing? They're looking for markets that aren't competitive, that haven't been flooded with artificial content with artificial rankings, so they can flood them with artificial content with artificial rankings. What few natural markets are left won't be natural for much longer, because a natural market is the perfect place to make money with unnatural content. Any new natural markets that develop will be quickly exploited, too.

I'm referring more to the search for information than the search for products, BTW. The e-commerce side of the internet is a whole nother story. In the end it will be ruled by the free market, and that's the way it should be. In many cases relevance is a non-issue in the search for products. Is there any reason that Dell should appear before Apple in a Google Search for "computers"? Aren't they both well-known manufactures of computers? The #1 "spot" (it shouldn't be called a "ranking") should go to the highest bidder. It will ultimately go to the company with the most resources, anyway.


#5 Jill

Jill

    Recovering SEO

  • Admin
  • 33,004 posts

Posted 25 September 2009 - 11:32 AM

QUOTE
What few natural markets are left won't be natural for much longer, because a natural market is the perfect place to make money with unnatural content. Any new natural markets that develop will be quickly exploited, too.


Unfortunately, I'm afraid you're right about that.

#6 Kanshu

Kanshu

    HR 1

  • Members
  • Pip
  • 7 posts

Posted 25 September 2009 - 08:54 PM

If "psychological" can be expressed mathematically, there might be.

#7 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 26 September 2009 - 08:47 AM

Definitely a tough nut to crack. Randy mentioned both sides (searcher/results) and they both make the problem difficult to solve.

One side of the problem could easily be addressed if SEs would put two radio buttons near the search box so that I can indicate if I'm searching for product/service or information. Would that not weed out a ton of crap in the SERPS? Ecommerce is the life blood of the net, but I'm not always looking for a product/service and would love to get some research done on various topics in a more timely fashion. To me that would go a lot further than simply indicating the relevance of a SERP based on my query. Combined, they would be very powerful and useful to a SE I would think.

#8 greggb

greggb

    HR 4

  • Active Members
  • PipPipPipPip
  • 118 posts
  • Location:Oregon

Posted 26 September 2009 - 09:04 AM

QUOTE(BBCoach @ Sep 26 2009, 09:47 AM) View Post
Definitely a tough nut to crack. Randy mentioned both sides (searcher/results) and they both make the problem difficult to solve.

One side of the problem could easily be addressed if SEs would put two radio buttons near the search box so that I can indicate if I'm searching for product/service or information. Would that not weed out a ton of crap in the SERPS? Ecommerce is the life blood of the net, but I'm not always looking for a product/service and would love to get some research done on various topics in a more timely fashion. To me that would go a lot further than simply indicating the relevance of a SERP based on my query. Combined, they would be very powerful and useful to a SE I would think.


My thoughts exactly. There's no way that any search engine can know what you're really looking for without asking you. For that matter, when you start looking at information you'll see that there are different classes of information. Some is very factual. Some is very subjective. Some is intended to be very persuasive. It's not that difficult to make distinctions between obvious classes of information like the ones I mentioned. Then you add these to the radio button, giving searchers the opportunity to include or exclude a particular type of information.


#9 Michael-F

Michael-F

    HR 3

  • Active Members
  • PipPipPip
  • 85 posts

Posted 28 September 2009 - 11:38 AM

I've wanted to see this for a while now, but I'm not sure how it would work out. Most likely, people would game the system in an effort to show up in the results for all types of queries. If they're an ecommerce site, they'll do whatever it takes to show up in both the commerce and information sides of things. It won't matter that the information side might not target the right types of visitors for those sites - people will still do it.

If this search engine checks individual pages to find out if they're selling something, people will have separate pages for ranking in informational queries. If it labels entire sites, people will create separate domains or subdomains to rank on the information side of things. If it's something simple like a meta tag, it will get abused like crazy. If it's something else entirely, someone somewhere will find a way around it (and tell all his friends).

Whenever you create a system for people to compete in - especially one on a global scale like a search engine - you have to expect people to do whatever they possibly can to get a competitive edge. The honor system does not work on the internet, and as long as the internet is a marketplace it will never be natural.

#10 Randy

Randy

    Convert Me!

  • Moderator
  • 17,540 posts

Posted 28 September 2009 - 04:39 PM

Plus the search engine would have to wonder exactly how much they can trust what someone put in a radio select button. People often don't say what they really want. Hell most times they don't even know what they want. lol.gif

Want proof?

On a few of my sites I offer a few of the same things I sell and make 'em available for free. Judging by my own conversion stats people arriving via one of the various "free" keyword phrases for which my decided commercial sites convert to buyers at a higher rate than those who arrive on non-free keywords or other avenues.

As a little marketing exercise my personal goal is to eventually convert 40% of all "free" searchers into buyers. Even though I do have a couple of competitors who do offer the same thing for free, just not at such a robust level as my pay services. My "free" searchers started out converting at something less than 2%. I've been moving the number up incrementally by testing different things over the last few years. As of last month they'd hit a new personal high, with fully 34.6% of searchers using "free" in their search phrase turned into customers.

Yeah, people often don't know what they want until they find it. Then they're more than happy to pay for it.

#11 greggb

greggb

    HR 4

  • Active Members
  • PipPipPipPip
  • 118 posts
  • Location:Oregon

Posted 29 September 2009 - 10:16 AM

QUOTE(Michael-F @ Sep 28 2009, 12:38 PM) View Post
I've wanted to see this for a while now, but I'm not sure how it would work out. Most likely, people would game the system in an effort to show up in the results for all types of queries. If they're an ecommerce site, they'll do whatever it takes to show up in both the commerce and information sides of things. It won't matter that the information side might not target the right types of visitors for those sites - people will still do it.

If this search engine checks individual pages to find out if they're selling something, people will have separate pages for ranking in informational queries. If it labels entire sites, people will create separate domains or subdomains to rank on the information side of things. If it's something simple like a meta tag, it will get abused like crazy. If it's something else entirely, someone somewhere will find a way around it (and tell all his friends).

Whenever you create a system for people to compete in - especially one on a global scale like a search engine - you have to expect people to do whatever they possibly can to get a competitive edge. The honor system does not work on the internet, and as long as the internet is a marketplace it will never be natural.


People will always be trying to game the system, no matter what it is. But the system I'm talking about wouldn't be nearly as easy to game as the current one. In this case your search engine is measuring quality to a much larger degree than it currently is. That is, quality as it pertains to exactly what the searcher is looking for. Provided that the SE really was capable of determining quality, gaming the system would require creating content of higher quality. Compare that to today's search engine algos, where gaming the system is often mass-producing content, spamming, google-bombing, etc. Gaming the system in this case could actually result in a better experience for the searcher.

Let's say that the searcher is looking for pure information, and there just so happens to be a piece of the information he/she's looking for at an e-commerce website. If you deliver this to them as the #1 result in the SERP, and they click the link and immediately find what they're looking for, they're happy. What more could you ask for than to make them happy?

The key really is in knowing exactly what they're looking for, and of course, being able to provide that to them. You need to go beyond a simple text box for keywords - that's not enough to determine exactly what they're looking for. You may even need to help them figure out what they're looking for, as was suggested by Randy. And your content needs to be measured/categorized in a way that when someone tells you exactly what they're looking for, you can say, "oh, I have one of those right here".

An analogy of today's search technology would be walking into an auto parts store, mentioning two to five words, then watching the person behind the counter scramble to gather 50 to 100 parts, and place them in front of you in the order that, based on his past experience, previous customers have chosen them. Can you imagine watching this happen?

An analogy for the technology I'm talking about is more like the real world. You walk into an auto parts store, spend some time describing what you're looking for, and quite often the person you're talking to asks you some questions to help you determine what you're looking for. Once they know exactly what you're looking for, they walk into the back room and bring it back to you.

Search engines need to become more like the latter, because more and more unnatural information is being pumped into the internet. A lot of the information being pumped in isn't original, or doesn't offer anything of real value. It will become increasingly futitle trying to find the best sites/pages by comparing a bunch of sites/pages that are essentially the same, according the criteria you're using to grade them (i.e., keyword placement, link text).

It needs to stop being a question of, "which one is the best?" and become a question of, "what are you really looking for?".


#12 Michael-F

Michael-F

    HR 3

  • Active Members
  • PipPipPip
  • 85 posts

Posted 29 September 2009 - 12:03 PM

It's definitely an interesting concept, and I'd love to see it implemented successfully. It just seems impractical from a technical standpoint.

Prove me wrong and make your millions smile.gif

#13 greggb

greggb

    HR 4

  • Active Members
  • PipPipPipPip
  • 118 posts
  • Location:Oregon

Posted 29 September 2009 - 12:39 PM

QUOTE(Michael-F @ Sep 29 2009, 01:03 PM) View Post
It's definitely an interesting concept, and I'd love to see it implemented successfully. It just seems impractical from a technical standpoint.

Prove me wrong and make your millions smile.gif


I've actually been working on this a little bit. What I've learned so far is there are some types of content whose nature is really obvious. Persuasive writing, for example, uses certain words and phrases much more frequently than would be found in factual information. Subjective writing is likely to use the word "I". These are obvious things that anyone could figure out.

Unfortunately, this kind of leads you down the path of statistical modeling, and isn't a whole lot different than modern SE semantics. This could very likely be manipulated by keywords placement/usage.

Still, I'd venture to say that there's no disguising certain types of content. Or, in using/not using certain words/phrases (which might be done for SEO purposes) the power of your writing will be significantly diminished. For example, can you imagine a political pundit, trying to persuade you that his/her view is correct, not asking a rhetorical question, or a question for that matter?

Look at some of the problems with search today. In a search for information you might come upon a page with a salespitch that contains the keyword you're searching for in optimal places, and a strong link campaign behind it, resulting in its high ranking. Well, if you knew for starters that the person searching was looking for information (i.e., radio button), and you knew that this particular piece of content is very persuasive in nature, you can probably assume the searcher won't be happy with this result. How would you know that this piece of content is persuasive in nature? Look for things like questions. Look for heavy use of the word "you".

If this were part of an SE algorithm, what would someone have to do to game the system? They'd have to stop asking questions, which are important to their persuasive process. They'd stop having to personalize their writing with use of words like "you" which also really help in their persuasive process. So, if they've successfully gamed the system, they've dramatically decreased the persuasive power of their writing.

This is one really simple example of what I'm talking about. I believe this is something that is very practical (and doable) from a technical standpoint. If you'd agree that this example is doable, then you have to consider the possibility that this methodology could be used in more complicated examples.


#14 greggb

greggb

    HR 4

  • Active Members
  • PipPipPipPip
  • 118 posts
  • Location:Oregon

Posted 29 September 2009 - 01:03 PM

As far as making this happen, one of the things I've considered is making it a project where anyone could contribute, and making the entire process, and all of the findings, completely public. I mean, no secret technology. I think there are a lot of brilliant people out there, many of whom don't even realize their brilliance in an area like this. I can't think of a better way of bringing them together in a kind of "think thank" environment than with an open project where you wouldn't have to worry about being screwed.

In the best-case scenario, where a few brilliant minds came up with something brilliant, and it was public knowledge, well... it would really start to level the playing field for all of the search engines. Not that I think there are any serious problems with Google pretty much running the show (I think they've been really responsible with their power so far)... I just like the idea of not having to answer to Google as much as we have to. Do you realize how much of our livelihood is controlled by one company? If Google de-indexed your company site, wouldn't you be in big trouble?

And not to say that Google still wouldn't be the most popular search engine for sometime as the result of the patronage of its existing users... but this would be the biggest field-leveler I can think of.

For me, personally, it'd be cool to be part of something like that. I think there are others who'd feel the same way.

What are your thoughts on this idea? I mean, a project where any one could contribute to the development of new search engine technology? It wouldn't have to center around the idea I mentioned here, either.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

SPAM FREE FORUM!
 
If you are just registering to spam,
don't bother. You will be wasting your
time as your spam will never see the
light of day!