Years ago, the world of SEO was filled with mystery and intrigue - vague half-answers from the engines combined with private forum and backroom conspiracies to make for an industry rife with misinformation. Today, those problems still exist in some forms, but the engines are far more forthright and the mysteries of the ranking algorithms are no longer shrouded in dubious half-truths. Sure, we may not know everything in them, and we may not have the right balance, but by and large, search marketers can read through a document like the search ranking factors and feel fairly confident in their wisdom (if not, neccessarily, their abilities). As the old joke goes - "I stole the Google algo!" "What does it say?" "We need links!"
However, all that aside, there are still plenty of straightforward, honest answers to questions, both simple and complex, we'd all love to get from the engines. I'll share a few of mine, and hopefully you can fill in lots of your own in the comments. If we're lucky, one day in the future, these may all have answers.
A few of my questions (in no particular order):
- Does a link from a page with meta robots="noindex, follow" carry less weight? no weight?
- What role do search quality raters play in determining rankings?
- Does your engine ever use the predictive abilities of search keyword demands to profit outside the world of search?
- Some domains move effortlessly to new domain names without a loss in rankings, while the vast majority go into the "sandbox" to languish for many weeks or months - what are the factors affecting the decisions to "trust" some domain moves while "distrusting" others?
- How much impact do the other domains owned by / registered by a site owner have on the way a site is viewed/treated algorithmically?
- What is the purpose/motivation behind obsfucating accurate, precise keyword usage data? (why not simply charge for it?)
- What is the purpose/motivation behind obsfucating accurate, precise link data? (why not simply charge for it?)
- Do better webmaster relations have a direct, positive impact on earnings?
- Why don't you (mostly Google, Ask and, to a lesser degree MSN) refrain from building/owning content portals that could deliver traffic and revenue?
- Google - is your share price overvalued? :)
- Do companies/sites that spend a lot with your engine receive any SEO benefits (free consulting time, a few tricks from an engineer, etc)?
- Do you use any of the following - latent semantic indexing, keyword density, term vectors, term weighting?
- How do you detect cloaking? No, really?
- In less than 100 words, describe why you choose to rank Wikipedia above accurate sources?
If I were Matt Cutts or Tim Mayer or Eytan Seidman or Kaushal Kurapati... I would probably answer - 1, 5, 7, 8 and 9. Seriously, though, I'd really love to get your questions below. Next time I sit down with these guys, I'll put their feet to the fire (or, in my case, a slightly warm pebble).
p.s. I promise - lots of fantastic stuff on China to come ASAP. And if anyone from Google China in Beijing is reading, please drop me a line (rand_at_seomoz.org) - we'd love to see you while we're here :)
I would ask a more general question, one that would benefit a great deal of people were it to be answered .
If you have a crappy site, but are making it better, does it have any chance of being judged for it's actual value, or will it always be "judged" as bad.
More specifically, if you have a thin affiliate site, but start adding quality content, will that quality content ever be given it's due?
Another question I would like answered (come on Matt, I am going to ask you this at SMX, so you might as well let me know) is really boring to most.
I am seeking a detailed explanation of the Google webmaster guideline which states
"Don't use unauthorized computer programs to submit pages, check rankings, etc. Such programs consume computing resources and violate our Terms of Service. Google does not recommend the use of products such as WebPosition Gold™ that send automatic or programmatic queries to Google."
I am not writing about this guideline on feedthebot till I get some info from Matt hopefully to make clear what this means as far as popular SEO web tools go. Are page rank cheackers breaking this guideline? Are there tools out there using api's that are okay?
I am the webs biggest Google guideline geek. I would absolutely adore a breakdown of this from Matt. I am very likely the only person on earth who cares about this but I care about feedthebot and want to get everything right that I can about the guidelines.
It's zealots like you that push organisations like Google in the right direction, ftb. I don't honestly care that much about that particular guideline, but I'm glad you do (I mean that!).
Good one! I want to hear the answer to this one. I am wondering if they are just covering their butt with this statement for in case they have to kick someone out for miss-use or something.
2 great points FTB - AND I am particularly interested in the first part. Is it once bad always bad, or IS their hope. If it is once bad always bad but you have a whole lot of great backlinks, if you .301 to a new domain does the stigma from the old URL pass along with the link juice?
Okay, I'll tackle one: "In less than 100 words, describe why you choose to rank Wikipedia above accurate sources?"
That's kind of a "have you stopped beating your wife yet?" question, dontcha think? :) Here's a simple example that demonstrates a case where Wikipedia is valuable, and why lots of people like it and link to it.
Information need: What was the 2nd book in Terry Pratchett's Discworld series?
Answer: The best resource would be a comprehensive list of Pratchett's Discworld books. If you do the query [terry pratchett], the official site is #1. Click on it. There's a section called books which looks promising, so click on it. Ah, a "Discworld" link! That must be it, so click on it. Whoops, what is Good Omens doing on that list? Death appears as a character in passing, but it's not a Discworld book. And the ordering doesn't look right. Crap, I've just wasted a bunch of time on the official site, and it's not even accurate.
Now go back to the search results and click on the site right after the official site, which is a Wikipedia page for Terry Pratchett. Ah, it's a good page that even has a Discworld section. In that section I see "See the Discworld article for a list of Discworld novels." So I click on that and get a list of books sorted by the date that they were published. That page also links to another page with suggested reading orders for the series. Wikipedia was more accurate than the official site.
This sort of thing happens all the time: a discography for The Cure, comic books with The Sandman or John Constantine, etc. It's surprising how common the "official" sites mess this up. One exception I ran across this month was the John Rain series; search for [john rain] to see a great example from Barry Eisler of how to promote a book series. The order is clear, there are sample chapters, and Eisler gives multiple options to buy the book. Here's the url: https://www.barryeisler.com/books.php . Pretty good stuff.
Of course Google is open to algorithmic changes that improve our results, regardless of whether that reduces how many Wikipedia results are returned in search results. But if you put on a user hat, many Wikipedia pages are a great match for searchers. Sure it's possible to construct searches where it looks artificial for Wikipedia to show up, but for the most part those are also artificial searches that real people don't do very often.
More than 100 words, but not too long. :)
Thanks Matt! Really enjoy seeing you answer one of these! How about doing another one tomorrow?
Maybe some of the SEOs should think about what you explained and see if they can enhance their site to take #1 postion for "SEO" and "Search Engine Optimization" - away from Wikipedia. Pro SEOs have their pants down with wikipedia beating them... but maybe the reason is obvious now?
Too much marketing blah blah and not enough "What is SEO"?
I am off to redo a few topics on my site and see what happens.
I want to know....
1) What data does the Google toolbar report about a person's web activities?... and what use is made of that data?
2) When I have adsense code on my site, what data about my site is collected beyond which ad is clicked? What is that data used for?
3) Same question for what data is being collected when a person is logged into a google service.
I'd like to know what's heppneing to all that data too as well as what are the planned uses of that data in the future.
The benefit to spending millions a month with Google is that you can generally get an engineer to respond directly to emails and phone calls. The bad news is their responses stick to their Webmaster Guidelines almost verbatim. It took about three email conversations and two phone calls with the Googineers to confirm that my time was better spent practicing SEO instead of trying to get them to answer specific questions about the sites I was working on.
There are two ways fro SEO: classical way and technical way. When i mix them up I incline to the former.
Classical way deals with what to do so a website caters to visitors or potentials. Technical way includes a host of factors to analyze.
It's good to make a complete analysis, but simple dos and donts, numerical values, or facts don't add up to a strong SEO.
The point is how these factors contribute to a specific business heavily relying on Internet. As an SEO practitioner, I keep asking myself: do i understand the business and how the factor is meaningful to it? If I fail to consider this, I don't think my work is relevant.
For me, SEO is like translation. Though having been in the industry for years, I'd sweat over unfamiliar topic. In SEO, I'd also sweat over websites about unfamiliar business.
I'm interested in Question 8. Positively. I recall those days without any search engine. I'd like to call PageRank Public Relations factor. And I'd like to do SEO as if there were no search engine.
"Work like you don't need the money, love like you've never been hurt, dance like nobody's watching...
& SEO like there's no search engine."
i love it.
Awsome quote! With the way things change at Google and answers from them being so vague, that's just about the only way you can SEO and stay sane.
Love it.
Nice post Rand, I would add one more question:
How come YouTube videos rank better than sites that actually do have content? (Maybe I would only ask Google this one ;) )
p.l.u.r.
Well, since Google owns YouTube, they push those results before all others. Everybody is doing this for their products. It's like when you sell cars. You're an official VW reseller, but you have Skoda or Seat in your sortiment aswell, of course you'll put VW cars in the front row and the other cars put in the darkest corner of your expo hall. This is how business goes, put your stuff on the first place in the market and keep watching how the rest dissapears.
Skoda or Seat? You must be outside the states :)
How lucky for you to even have those options....
Great questions Sasha!
I'm not google but I can most definately answer with 100% guarantee that if you add on meta NoIndex, follow this does not punish your website. I have on all my websites this tag as I hate when google visitors browse the cashe insted of my own page. From my experience I did not loose any rankings when I added the noindex, follow tag and my new websites that are born with that tags are ranking quite well.
Cheers,
Venetsian
/SEONewsBlog.com/
Oh that's really interesting Venetsian, thanks.
Nice post Rand.
I would add another question based on the latest situation. The questions is; what Google is going to do with Performics Search division?
Mine would be ...
How do you see multimedia such as Javascript, Flash and Ajax and will it ever be different in the future? Do you give weight to multimedia pages with many links, click throughs or what?
Thanks for the quality content Rand, nice topic and post.
I'd like to hear Google's internal take on this question: Do you ever think of the globally connected internet as a brain, in which the neural pathways that are continually etched and reinforced are links, and have you ever thought about the future of the web as the future of an intelligent entity as it develops? And do you think the ranking algorithms of Google and other search engines define the function of this intelligence and its likely success or failure?
Cosmic enough for ya?
Gradiva
5. How much impact do the other domains owned by / registered by a site owner have on the way a site is viewed/treated algorithmically?
I would love to see this question answered.
I think there is a hidden penalty for people that own a lot of domains... not sure how it works... but there is definitely something there.
My question would be:
What are the disadvantages to giving us real search volume numbers? MSN currently does (kinda) with adlab.msn.com, and Overture kinda does, but that is in limbo. It woudl be great for Google to put real numbers on Trends and other keyword research tools.
ABSOLUTELY! i would love to see Googe figures. Wow, *another* resource we can use - that might be nice... ;-)
Seriously, do count links on pages that have the meta tag robots with value noindex, follow at all? I believe to have heard that they don't count for anything. They are excluded together with the page content.
Some faint memory of somebody saying that only links count that are on pages that are included in the index. Pages with the mentioned meta tag are not being indexed. Only the URL with no snippet and no cached version will be shown if you specifically search for those pages.
Paying for tools from the search engines? no brainer at all. We all pay a lot of money to services that do better or worse educated guesses about that kind of stuff, based on a big, but often skewed or not big enough corpus of data (or do the scraping for us). Add to the list tools for competitive intelligence. Who ranks for search term xyz at which position, who is bidding on those terms. Alert me if... etc.
My Question.
How do you deal with the obvious internal conflicts of interest when it comes to advertising revenue vs quality of service to the end user (yeah, they exist)?
Second Question
You know most of the time if ownership of a domain changes and often wipe things like PageRank and other values. Why is it a problem doing the same with all the negative things, such as penalties or complete bans?
Will it ever be possible to rank websites with links playing a zero factor in the algo? If so, what might the new metrics be?
EGOL, how about #1, since it's short.
"Does a link from a page with meta robots="noindex, follow" carry less weight? no weight? "
For Google, I believe such links would carry the same weight as normal links on regular pages.
Woohoo! Thanks Matt!
Matt, thanks for your time (vacation time).
Interesting. Then maybe another SE is doing it different. I wonder how links on non-indexed pages can have the same impact on ranking as if the page they reside on was indexed by Google. In order to evaluate anchor text and the surrounding context of the link to determine relevance do you need the page content.
Does this mean that noindex does not prevent Google from making a copy (or partial copy) of the page and store it on its servers?
In that case would "noindex" only mean, that the page will not be returned in the search results (except for the url itself, without title, snippet and cached copy, if links point to the page). Preventing Google from storing anything of a page is only possible by blocking the crawler via robots.txt. Is that about right?
Thanks Matt! You know we all really appreciate your answers. Now, how about #5, for good measure?.. :)
hmm, so I found this in the answer on the "SEO Quiz" and I guess it made me think. Why would a like from a page that you have told the search engine is not valuable enough for you to want it to be indexed carry the same weight as a link from an indexed page. It doesnt make sense. Would it be passing the weight from that page it was spidered from (since we are assuming it wasn't indexed as requested)?
Or when you (Matt) say "I believe" is that a non official answer?
love your site, you guys are awesome. Maybe someone can answer this bugging question for me.
Why do the search engines still rank sites highly that are not longer in business, phone numbers no longer work, addresses no longer viable, etc.... This to me seems so unrelevant and clogs up the rankings for legitimate working businesses.
Thanks.
takebackourgvt
Least... but not last question:
what's in the next future!?!?!... Hoping for really semantic engines?!?!...
mmm... I dont know!!... Perhaps, the Earth's Pole Shift that may occur in 2012 is more reachable!!! ;)
For me, all Rands questions seem to be about what search engines are doing *right now*.... I would be far more interested in finding out what the big search engines are considering will be relevant *in the future*
Fair point, but like what? Give us some examples?
I would have to ask:
"Do you give high spending PPC customers prefrential treatment in the natural results, especially if they get penalised or dropped with regards to re-inclusion times?"
Because I have my suspisions :)
i'd like to know why when a site ranking first on google gets Sitelinks it oftens gets also second position (and with a link already in Sitelinks).
7. What is the purpose/motivation behind obsfucating accurate, precise link data? (why not simply charge for it?)
Yeah I was wondering this one myself Rand... also especially since MSN have removed this tool altogether and offer no real reason as to why. Looking very bad in my opinion as when its used all that appears is a blank page. At least we can have a more indepth look at 'link' within Google/Yahoo.. they at least appear to be beginning to help the webmaster not hinder...
It actually counts more towards your score because that page isn't indexed because its on a nofollow page so the search engine just collects the data and awards it to the " Special Place " of unfiltered score rank your website is awarded in any search engine.
Regardless every search engine all of them share the same data " Explain why they show all the same results for pictures " it all comes down to Data Warehousing and when it comes to who can store the most information and data securly Microsoft holds all the cards.
I found that it's a great thought exercise if you try to think like a search engine and answer these questions from their point of view.
I've answered a few on my blog: www.dotcult.com
Well, I can give you a possible answer to question 11....and it's a NO.
I've worked with companies that spend $1mil/month just on Adwords and they never got any SEO tips or benefits.
Awesome post - and just in time for SMX. I gotta start looking through my notebook to come to the conference loaded with questions :D
Gptta agree here, I have also worked in the past with an agency spending over £1m a month on clients' behalf and never received preferential SEO information/attention
agency spending over £1m a month on clients' behalf and never received preferential SEO information/attention
Yes, this is true because there really isn't any special SEO info or secret trick to top rankings.
The answer to question #8 can be answered by seeing the results of Google's SEO outreach program and by talking with Adam Lasnik, who says he is searching for love not money. It is almost impossible to fake enthusiasm and trust, and people like Adam have increased trust between webmasters and Google. Indirectly, positive relationships will help a company financially.
My question would be: What do you think of all these SEO's installing nofollow tags all over their sites like obsessive-compulsives incessantly washing their hands? A huge percentage of websites are installing nofollow tags on rss feeds, contact pages, even their homepage links.
Do you laugh at this behavior or wonder why nobody truly understands nofollow? What is the next diversion you will create to give SEO's something to debate and overbill clients for?
I'll concur with that resounding 'No!' And I think this practice is entirely justified
It would be great to know the answers for these questions, but since this whole search engine ranking thing (or whatever) is a multibillion dollar business I doubt we will ever get a clear answer. It's possible that some officials will give us answers and that the majority of us will accept those answers, but my feeling says it'll give us only more question.
I do have questions, but I'm still not throught all articles and all tips and tricks so I guess while reading them I'll get my asnwers.
Ja, so much for ensuring the most accurate and relevant user experience - one might think 6 billion would be enough though mightn't one... Though of course the official mission statement simply says to organize the world's info and make it universally accessible, it says nothing about organizing it for the benefit of the user.
Honestly Rand mine would be several questions, I could ask questinos all day long, but one question would be about the no follows. I have been hearing from long term seos there starting to say they actually give weight, and I seen one site shoe an example of where they was ranking for a keyword, and it was only through blog comments, and etc, so basically all the backlinks were using nofollows.
but I would have several more questions like pr questions, anchor text questions, and I could ask numerous questions :)
Are you referring to the 'Spiderman 3' and 'Piderman 3' example?
That was an interesting article. Makes me want to try my own tests similiar to that to see if I can get the same results.
About the question list...
I wanna know why google won't give keyword search counts like overture gives in their keyword suggestion tool. I don't want to see a little bar graph graphic showing it gets traffic, I want to know how many times a keyword was search for in the previous month.
Wow you've really started something here Rand. I can see this spilling over into a dozen other blogs at least. :) I might even reference these great questions myself, if you don't mind.
I read the questions from EGOL and I remembered one thing that realy bothers me since I saw that info about the search engines. It's about the cookies the search engines use.
Why do search engine cookies expire in the far far future?
- Some Google cookies expire 2008, but most of them expire on January 18th 2038.
- Some live.com cookies expire in the year 2010, 2021, 2022
- Some Yahoo cookies expire in the year 2010, 2015, 2037
Are those dates set on purpose and if, what's this purpose. Did Google take this January 18th 2038 because it's one day before the year 2038 Bug ?
I got more than one question, but the main one is "Why these long expiration dates"
I know that Google owns YouTube but what will hapen if they couple of other big players? Will the top 10 be Google's companies or would they be fair enough to the rest of us?
Why would anyone want to use a search engine that favors it's own companies? I don't think that people in Google are going to manipulate their SERPs since it could hurt them in the long run...
p.l.u.r.
Re. Wiki - Because for most people, that's what they want. If they want more detail, they can skip it. It's only one result. For the rest, who just want a quick bit of information, it does the job.
Interesting point...
Looking at a previous SEOmoz post - the different types of users (commercial, research etc).. I wonder what percentage of searches relate to each type of search and how the search engine uses this data (if it does) to rank its results...
For example, to take a term 'pizza franchise' (thanks Wellwrittewords!)
Does a search engine determine them a 'researcher' - looking for information on how to run/where to get one from? Or does it consider that user a commercial one.. looking for local pizza franchises in their area for dinner.. and therefore how do they display the results accordingly... is the fact that Wikipedia is at the top of the results just cause of their massive research value? and hence we can assume the majority of users are research ones? or are the sorts of searches it appears on merely just the ones a search engine considers 'research'.
ok I stop rambling now ;)
'Re. Wiki - Because for most people, that's what they want. '
Maybe, but in many cases not. It takes up valuable real estate as well. Perhaps there could be a 'include wiki in results' or 'do not include wiki in results' option? probably not, but knpow where the wiki is and will go there with a query if I have one - in many cases I am not interested in the wiki page or i'm not 'researching' and am thus more intested in subjective matter, and constantly getting a wiki result in the results top 10 makes my right eye twitch.
WIKIPEDIA *IS* A RELIABLE SOURCE!!!