Brands are Spammin' and a Cover Up?

30 comments

A few weeks ago, a guy named Robert Medford sent out a press release regarding Fortune 500 companies who were using hidden text to apparently spam the search engines.

It was posted in a thread on my forum by one of my admins, and we discussed it a bit. Interestingly enough, not one other forum picked it up, not even the white hat supremecists over at IHY.

Well, this weekend, the author wrote to tell me that he posted a comment on Matt Cutt's blog in a thread where Matt was discussing hidden content.

He saw his comments appear on the blog as they should, but noticed later at a friends house that they were gone. Back at his own computer, they were still there.

I told Robert perhaps it was just a caching issue, but he showed me his saved html file of Matt's page with his comments still intact, along with others that were posted afterwards. (This is all posted in the thread on my forum.)

So, a few question for the TW'ers:

Is there any logical explanation for why only Robert can see his comments on Matt's blog? And also, would the hidden text used by the companies Robert posted about, simply not be considered spamming anymore and I missed the memo? (Perhaps the old black hat saying that if it's relevant it isn't spam, actually is true after all?)

And why might Matt apparently not want this discussed openly since he is supposedly a fighter of all that is spam?

Comments

any logical explanation ?

Jill

To be clear on this, you are suggesting that there is more to it than just Matt removing a comment from his blog (for whatever reason)?

For the life of me, I cannot see Matt taking the trouble to do anything more nefarious than that.

easy answer

When you first post a comment on Matts blog you can see it immediately but no one else can until Matt approves the comment. That explains the seeing them one place but not another issue. Why Matt may not have approved them I'm not even going to guess.

While the subject's raised I think 'outing' people on Matts blog is getting out of hand. I have no idea if your guy is right or wrong Jill but someone did 'out' a well known UK company for spamming in his comments a week or so ago and all they were doing (unless I missed something more nefarous) was geo targetted delivery of different sites to the US and UK. Matt didn't comment either way on it but given the number of citizen activists that get worked up over there I've been expecting to see some form of DOS attack against the poor guys ever since.....

As to why it shouldn't be discussed openly. Well being English about the only one of those companies I've heard of is GM, so I looked at that one. That doesn't look like a deliberate act of spamming to me, it looks like they were doing what they thought should be done and might have got it wrong. Perhaps the main site has CMS but changes in that haven't changed the copy on the noscript? Perhaps they are evil keyword stuffing money grabbers with no morals? Perhaps many things but I couldn't be bothered to get worked up about that and the entire article looks like link bait to me :)

Its polite to warn people who may have made a mistake before outing them, and I cant see where it says that was done in this case. I disagree when Matt makes OTT examples of people in his blog and I disagree when other people do it too. Sorry but the entire thing is a non-story IMHO. He should have timed it for August when we were all desperate for something to write about :)

yup

The comment's in mod show for the submitter. Clear cookies and they disappear until approved.

As for hidden text=spam, that should be a judgement call because of all of the potentially legitimate ways it could happen. Positioning is a wide open option, as are those nice in-line detail insertions that go along with hide/nohide anchors. Perhaps that is why such "outings" on Matt's blog are so interesting... it really can't continue like that forever.

Restoration Hardware

If you look at the Restoration Hardware they talk about hiding keywords with CSS, when pressed earlier this year Matt said while they have identified stuff like that they aren't automatically penalizing it (I assume that means a hand edit would be required)

http://www.threadwatch.org/node/4313#comment-26923

I'm with Gurtie mildly amusing to look at, but looks like link bait to me. Yes using Matt' site as an outing tool is childish. Next thing you know people will be giving their competitors sites at conferences for site reviews, oops did I say that ...

Not denying that it's link bait

Ok, I completely agree that the article was designed as link bait. Medford so much as admits that. And I can understand why many forums didn't pick it up, because of this. (Although, I see other link bait picked up all the time without a problem, so I do have to wonder why in this case it didn't work.) I never heard of Robert Medford before having read his press release. But what he wrote was indeed true stuff. (Worse, untrue stuff gets picked up all the time.)

We don't allow outings on my forum either for so many different reasons, and I wasn't even going to discuss the article either, but Alan felt it was worthy of discussion and posted it. The subsequent discussion was a good one.

All that aside, regarding Matt's blog, as others have mentioned, people are outing stuff on Matt's blog all the time, including Matt. So why is this one so different that it didn't get approved by him? Look at the crap he approves? It generally seems like he approves anything!

I posted last night on Matt's blog about this too, and will see if it's been approved (although I probably can't tell...can anyone else let me know if they see it?).

Like I said, I figured there were logical explanations for all this, and I'm not one to buy into conspiracy theories, but this one just kept snowballing and seeming to get weirder and weirder!

I do think it's a shame that any large company should hire an SEO firm who hides text on their page (whether they know about it or not) in this day and age. It's not only unnecessary, but just plain stupid, imo.

I don't see anything sketchy going on...

I'm no SEO expert, but those 'outings' seem like minor technical violations at best. I was expecting to see massive keyword injection...

Minor technical violations?

Purposeful hidden keyword stuffing is hardly a "minor technical violation."

That said, the sites may certainly be ranking well in spite of the spam, as opposed to because of it, but that's another story.

Hidden text, yes, but spam?

Take that hidden GM text snippet as an example. There's absolutely nothing in there that would not have been totally okay to post - verbatim - in bold *visible* letters on the same page.

It's a pretty neutral description containing the essense of the site's contents. Like a meta description. As for keywords, well, of course a text on the GM website would mention "GM Cars" - omitting that phrase would be weird.

I agree that it's stupid to hide that text. I don't agree that it's bad though - just stupid, like "silly". It should have been visible of course. However, I see it more as a pretty common (and, perhaps the only) way to get the message across when an ignorant graphic designer team have too much power.

The page is clearly designed for "the looks" more than "the message" - that's just stupid, graphical design. Far too common. It's the sort of thing you get when you hire web designers that would rather design for magazines.

---
Added:

"Lillian Vernon", "ShopNBC", "Restoration Hardware" likewise, IMHO. Dang ignorant graphical designers with too much to say. For the "Netjets" site, the graphical designers haven't even left room for decent site navigation it seems. Incompetence, I'd say. OTOH, "Briggs&Stratton" have your usual links galore at bottom of page, so even though that site is also severely handicapped by graphical design, the "solution" is a bit different, and not quite "clean", IMHO. I haven't seen the other examples.

For most, if not all of these cases, I do agree with Robert Medford that :

I believe that their rankings could be improved by a ground up SEO approach that emphasized content over form. Such a site would have no need for hidden text or links.

(Said about Restoration Hardware)

intent

we do keep hearing that spam is now all about intent, and I reckon the intent on the GM one wasn't spam.

Harvard, Fairmont, Harry & David

Now, the Harvard and Fairmont sites, and especially Harry and David - that's seriously bad planning and use of screen real estate (bad IA). If people built houses like this the kitchen would likely be located on the neighbour's roof.

You have a perfectly fine page (even with lots of textual content for some of them), yet you feel the need to write a short story (and then some) in hidden text as well. The amount of hidden text there suggests that the whole page should have been planned differently, as that hidden text is a whole missing section. That's even worse than too much graphical design, it's simply bad information architecture.

(wonderful way to spend January 1st, honing your critical eye :-)

I think these cases illustrate quite clearly that (especially in large organizations) SEO, and even basic IA is often only considered when it's too late, ie. after the graphical designer (perhaps - and most likely - an out-house agency) has more or less ruined the informational content of the page. And of course, as soon as the boss has said "wow, looks good" then you can't really change much.

I've found personally that getting into the process early and working with/alongside the graphical designer(s) yield much better results. No need for hiding stuff when everything essential is right there on the page, and the non-essential stuff isn't. And pages can still have that "wow-effect" graphically.

Right, claus

I'm one of those who believe that there's never any reason to hide content. You can write text that works for people and search engines and not screw up the look of the page as long as you don't treat SEO as a final step that can't undo any of the previous ones.

Now, regarding whether the examples here are minor enough to be ignored... If that's the way Matt sees it (and I wish he'd comment here even if he doesn't say anything about this on his blog) that's certainly his prerogative. The example he cites in that post is certainly hiding a lot more than the F500 examples in the article.

But it's kind of inconsistent, in my opinion. Back in October, he wrote the following in his Unwise Comments post:

Quote:
don’t recommend that people use CSS to hide text, and I don’t recommend that they document it, either.

I hope that's not intended to mean that a little hidden text is OK as long as you don't document it.

He later responds to a comment to that post by saying

Quote:
Search Engines Web, you said that ‘That site has a perfect right to use that “invisible” text and the h1 tags’. I agree. In turn, Google has the right to decide not to return that site in our search results, because we feel that hiding text that is not visible to users is deceptive.

No indication there that there's some threshold where a little hidden text that's ok crosses over into being deceptive.

I'd just like to know where Matt stands on the issues brought up in the article.

I kind of like that. More

I kind of like that. More power to them. Nice to see the big companies showing the search engine who is wearing the trousers :-p

I wish I had the power to do that.

There is no way blue chips can be ommitted from an unbiased set of SERPs. All the SEs can do is try to unbias the tricks they pull.

That's true

It would only reflect badly on the SE if a site like
GM's failed to come up for relevant queries, but by the same token, that would mean that there's absolutely no reason for them to bother trying to fool the SE. They can put whatever they want on the site, even leaving text out of it altogether, and still not worry about organic rankings. So why do they bother?

So why do they bother? My

Quote:
So why do they bother?

My guess is dumb SEOs.

Some uses for hidden text

There are actually good uses for hidden text. Usability issues, like on page tabs where the content on one tab is hidden until you click it and so on. So, the text is hidden only to become visible when the user does something.

I remember that the first year I had my own firm the homepage was actually just one single "flat" HTML page, but it appeared as (IIRC) seven pages because the individual "pages" were just layers on the same physical page being toggled on/off. An extremely fast page that was :)

Of course, the hidden-to-visible ratio for a user would at all times be 1/7 but for a search engine the whole page (7/7) would be visible at all times (and of course the same for a user without CSS+javascript).

But what you're talking

But what you're talking about is temporarily hidden text. All the user has to do is mouse over something or click a link and it appears. As such, it's not a case of hiding something from the user and showing it to the SE.

There are lots of uses for text that's hidden under certain circumstances. I don't think anyone is questioning that.

Oh yes, but I'm only saying

Oh yes, but I'm only saying it because if it's automatically detected as "hidden" then it's "hidden" regardless - something like "intent" is very hard to program *lol*

or...

My guess is dumb SEOs.

Or dumb Fortune 500 companies not listening or letting the SEO's implement the changes they are recommending, then forcing these dumb SEO to find "alternative" ways to perform under given contracts with them. Many of these websites have spent thousands if not millions on their current look and feel and changing them for the sake of SEO is just not and option for them. I am sure we have all had a run-ins with this type of feedback from prospects.

Or maybe it is all a conspiracy and Google does let some things go with the bigger websites. Of course the fact remains Google feels it is a good user experience for searches to find the GM website when some does type in "gm customer service" so they may turn a blind eye to the hidden text.

Who knows, just guessing here.

There are actually good uses

Quote:
There are actually good uses for hidden text. Usability issues, like on page tabs where the content on one tab is hidden until you click it and so on. So, the text is hidden only to become visible when the user does something.

Yep, agreed. And of course none of that is spam. If that's what these companies were doing, we would definitely not be discussing it. (At least I sure wouldn't be!)

separating opinion from "fact"

Well Jill, that is not how I read you comments above. You seem to mix your opinion with the "factual" definition of spam. Granted, there is no actual definition of spam LOL but still we have to be careful not to assume that what we think is spammy is actually spammy. Think lawyers and trial judges. Every plaintiff thinks they are right, but sans precedent, it goes to argument. We all know arguments are often won and lost on things other than fact, right?

What you consider a legitimate use of CSS to hide text may differ from another's. As claus suggests. I saw Matt once refer to a media=print offset as "spammy" (WTF?)

I doubt we can say "of course, none of that is spam" because some of it may be judged (by Google) to be spam. When and if they make such clear statements, we can move on. Until then, well let's just say this is leaning towards that ethics thing and we don't want to go there on TW, right?

Nope

This is not one of those "ethics" discussions. At least that's not the perspective I'm coming from. The question as I see it is whether the importance of these companies -- that a relevant SERP that didn't include such sites would be deemed less than useful by the general public -- allows them to get away with things Google (or at least Matt) would not want other sites to do.

If it's a simple "the sites are relevant despite the hidden text, not because of it" that's fine. But for Matt to choose to point out tactics that he wants us all to understand are a bad idea, but to only do so on Mom and Pop sites, does raise the question, IMO.

well

Well I think there are perhaps more important notes from this.

For example, if this is a legitimate publication, the SEs have succeeded in getting many mild tactics to be commonly referred to as "black hat" and to get Black Hat SEOs to become known as "spammers". Here it is clearly written on an SEO site, and openly accepted on this forum. "10 of the 12 sites" deployed noscript technique.

Also notice the DANGER of listing clients on your SEO site. Some of the players here continue to publish "guidelines" for hiring SEOs which include looking for or even talking to past clients. Really?

Finally, the writer mentions "ethics" himself:

Quote:
Apparently the prime real estate at the top of the search results pages has grown valuable enough to justify ethical lapses at all levels of the food chain.

and suggests the reader file spam reports:

Quote:
Want to help Google, Yahoo! and MSN keep their indexes free of spam? You can report it here:

The word troll comes to mind, with a sense that the SEO community is the troll bait.

Interestingly, a liguistic analysis would probably put this writer on the outer fringe, style-wise... it should not be difficult to compare the article with publications from known seo talking heads.... I wonder who would match?

SE Spam is only what an engine says it is

Quote:
I doubt we can say "of course, none of that is spam" because some of it may be judged (by Google) to be spam.

Search engine spam of course can only be judged as such by a search engine. It's easy for me to provide my opinion of what spam is, but it's only what a search engine says it is, either by their action in penalizing or deleting it, or by what they publicly state.

This is exactly why I'm hoping to get Matt Cutts' opinion on the matter so we can put to rest once and for all whether those sites are indeed spamming Google. At this time, it appears that they are not, since they do very well in the engines for the keyword phrases that they are hiding from most of their users.

No cover-up. Read about

No cover-up. Read about how comments work on my blog at
http://www.mattcutts.com/blog/comments/

The first time someone posts, I require it to be approved. As I mention at
http://www.mattcutts.com/blog/powerless/

I haven't had power for going on two days. So if I don't approve new commenters over the New Year's holiday weekend when I don't have power or broadband, it's a cover-up? Sheesh. Tough audience. :) Lemme check... yup, comment was posted 2005-12-30 14:15:02. It just hasn't been approved yet.

In fact, I'd seen this press release when it first came out, and planned to circle around on it after the holidays. I had already heard harryanddavids.com mentioned, so when I saw the press release I checked it out again. The extra text is in noscript tags, which our indexing system handles very well (not unlike if someone tried to keyword stuff in comment tags). I usually don't talk much about spam that doesn't affect our system much/at all, but I was planning on checking out Robert's list. In general, the answer to Robert's question is that large sites can be removed for spam just like any other site. For example, one of the largest auction sites in Spain was cloaking and was removed in December. We just recently reincluded it after it stopped cloaking.

Hope that helps, Jill..
Matt

???

The extra text is in noscript tags, which our indexing system handles very well (not unlike if someone tried to keyword stuff in comment tags). I usually don't talk much about spam that doesn't affect our system much/at all, but I was planning on checking out Robert's list.

So does that mean what harryanddavids.com is doing is OK or not advisable? You say the Google system "handles this well", which doesn't really tell us much on if what is being done is right or wrong according to Google guidelines.

Still not clear on this.

Wordpress ...

(Re the blog software, I believe that's the way WordPress 1.5x behaves: the first time someone posts a comment, it's held for approval. Not to mention that we've just had the yearly longish holiday period and, if Matt's in Northern California, there's a little problem with storms there right now.)

...

Quote:
The extra text is in noscript tags, which our indexing system handles very well (not unlike if someone tried to keyword stuff in comment tags).

I think that means it's ignored. You can stuff as many keywords in there that you want, but it won't make a difference.

If it doesn't work, the search engine has no reason to label it spam, so there's no need to ban/penalize.

I hope you look at the rest of the list, Matt. I just looked at one of the sites, and its home page has hidden top- and second-level headings, all within a single span (which isn't even proper code), and without a noscript element or a comment. View the page without any styles, and they take up about a third of the page.

Thanks Matt!

Quote:
Hope that helps, Jill..

Yes, it helps quite a bit, thanks! (We had already surmised over at my forum that you had simply not been around due to your missing power.)

I guess I can finally take off my tinfoil hat now. I was just getting used to the way it felt though!

I do hope you'll look closely at that "spam." It appears that there's a good chance that 1 company is responsible for a good portion of the sites that Robert mentioned in his original article.

Outing Spammers and SEO Business Models - the New Threadwatch

Do we really want to be outing spammers and exposing flawed SEO business models on TW? If yes, then let's take a close look at some of your client contracts __put your handle here___ so we can chat about where your profit margins are.

Of course not

TW is absolutely not the place for such things. I'm not naming names, and neither is Jill. I figure she brought it up here because I've never known Matt to post at HR (where we also don't name names), and she wanted to know why those comments hadn't appeared on the blog.

Question answered, problem solved (at least the perceived problem that led to the creation of this thread).

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.