Why your SEO advice sucks so bad

I was going to write another “rel=’nofollow’” rebuttal but, let’s face it, the Ministers of Misinformation are out in force and the SEO community no longer seems to be interested in the facts. Okay, that means I discovered a really biased, wacky explanation of the use of nofollow on internal links today after thinking about it for a while I decide to go looking for troub– more current discussions of the issue. In the process I came across some very badly written SEO advice.

At the risk of inflaming prissy egos and starting another flame war, I’m going to link to some of the really terrible articles out there and explain why they are so bad. I’ll be using nofollow on the links, so click on them at your own risk. If I had to judge these people’s knowledge, SEO skills, and talent based solely on these articles, I would not want to work with them. However, as you’ll see below, these articles may not be good examples from the SEOs that wrote them.

Advanced SEO Techniques: Website Design, Internal Page Rank and Nofollow.
This article tells us that spiders leave your page at the first link. I’ve literally watched major search engines crawl my servers from start to finish through deep crawls and never once have I seen a spider jump off the front page.

Spiders can work in mysterious ways, I suppose, but I’ve only seen them work in two ways. Way One: The spider fetches a URL (page location) from a list of URLs, fetches the page, hands the page off to a queue, and fetches another URL from its list of URLs. Way Two: The spider fetches a page at a URL, parses the page and extracts all the URLs, then adds those URLs to a list from which it fetches the next URL, repeating the process.

Way One, so far as I know, is how most major search engines crawl the Web. They have multiple copies of their spiders running at the same time so they can fetch pages from your server very quickly while fetching pages from other servers.

Way Two is the classic Deep Crawl process. Most people in the SEO community have probably never been deep-crawled, or they were deep-crawled when they weren’t looking. To notice just how intrusive a deep crawl can be, you have to have enough content to keep the spider occupied for at least a few hours. Assuming the spider fetches a page every 2 seconds (and I have seen them fetch faster than that), a deep-crawler will go through 1800 fetches an hour. If you only have 400 pages of content on your site, the deep-crawl will be over before you know it’s happening.

The article goes on to talk about how tables slow the spider down down, so you should use CSS to speed it up. The whole tables-versus-css debate died down a long time ago but obviously some people never got their facts straight. The faux arguments that some Web designers used to shoot down the use of HTML tables never proved a single point. Anyone can make a design technique look bad by abusing it and distorting the facts, and that is precisely what anti-table designers did.

Spiders crawl tables just fine. It’s not like the spider is a complex mechanical device stumbling across jumbled furniture in your living room. I have written spiders and page parsers. All you have to do is filter out the HTML code. Where tables cause problems for search engine optimization falls into two categories: an improperly coded table may cause a page parser to miss some text; and a very complex page layout structure may cause a page parser to linearize your text in a most confusing way.

Linearization just refers to what your text looks like after all the HTML code has been stripped out. You can nest tables from now until doomsday without impeding a search engine’s ability to linearize your text correctly. You can also avoid the use of tables completely, rely solely on DIVs and SPANs and CSS, and just totally botch up the linearization. Eschewing tables doesn’t make it any easier to ensure that text is linearized, especially given the way designers like to embed DIVs near the top of the page that are shown way down at the bottom. Nor does using tables make it any more likely that a search engine incorrectly linearizes your pages.

Bad Web habits afflict both table-loving and table-hating designers in equal abundance. They’re all schmucks if they strike up the band and start claiming to have the holy writ of Web design on their side. And SEOs are schmucks for repeating this crap philosophy.

PageRank dilution
This wonderful piece of StinkBait tells us that you’ll want to employ redirects so as not to squander any link juice (PageRank) that your site has acquired.

PageRank is the most stupid reason you could possibly have for implementing a redirect. The whole point of the redirect is to get people to where they are supposed to be. Redirects predate PageRank by several years, and some redirects — believe it or not — DO NOT CONFER PageRank. If you’re fussing over how to redirect your PageRank, you’re not someone I would want to work with.

Should an SEO advise a client to implement a 301-redirect if the client retires a URL? Absolutely. Should the reason the SEO cites be PageRank? Absolutely not. If a page URL goes away, it may still have links pointing toward it. None of those links have to confer PageRank in order to send traffic to that URL. Your first priority as a search optimization strategist is to get people to where they need to go. The spiders may or may not care about the links that lead to your page.

In fact, most links do not confer PageRank, so justifying the use of redirects on the basis of shaping the flow of PageRank pretty much shows you don’t have a clue when it comes to understanding how search engines work.

You redirect for people, not for PageRank.

The Ultimate Guide to Internal Linking

So this tutorial wants to gives us a guide on how to design and impliment internal linking that’s elegant and works (spell-flame: “impliment” should be “implement”).

Trick Question: Can internal links ever “not” work? (Think about this one for a moment)

Answer: No. Why? Because even if the link leads nowhere, it can still be clicked on. That is why I use “target=’_blank’” as a courtesy to my visitors. They have no idea of what is going to happen when they click on a link. Links ALWAYs work. They just may not take you anywhere useful.

The tutorial goes on to provide a concise explanation of Classic 1998 PageRank. Then it follows with “filtering”. The author recommends you filter out your Privacy Policy, Accessibility Policy, and Legal Terms. Hm. Never mind making that stuff accessible to people who want to find it (through site search).

Okay, so the first real advice is flawed pro-nofollow advice that doesn’t take into consideration the visitor’s needs.

But wait! There’s more. He also tells you to use your robots.txt file to block shopping carts, search results, thank you pages, and 404 pages.

Shopping carts make sense. Block them with robots.txt. Point links to them using nofollow. But “thank you” pages? Why do you have links pointing to your “thank you” pages? Those should only be generated pages that people see after conducting a transaction. How many purchases do search engines make in a day?

The indexing of search results is controversial. At first Matt Cutts struck fear into the SEO community by saying, “GOOGLE WILL DELIST SEARCH RESULTS!” (ominous thunder rumbling). Duh-dun-DUN!. Then he came back and said, “…but we might let them stay if they add value and people find them useful….”

Okay, if your site is just a search results regurgitator, you’ve got a problem. However, if you have dynamic content that people are linking to from other sites without any incentive from you, don’t block that useful content from being crawled.

Dudes, I click through Google’s SERPs to indexed search results every day of the week. I don’t want Google to remove those pages. I USE THEM AND I WANT TO FIND THEM.

Finally, our dispenser of sage SEO advice tells us to block 404 pages from being crawled in the robots.txt file.

Um…last time I checked, search engines didn’t index 404 pages. Oh, wait, perhaps he means “block dead URLs in your robots.txt”? Okay, but frankly if I had links pointing to a non-existent URL, I’d either fix the links or redirect the URL to a relevant substitute. Help the spiders (and your visitors) find the content they are looking for.

Dead URLs are not ‘droids, and you’re not a Jedi Knight, and search engines are not storm troopers, so you don’t need to be putting out “These aren’t the ‘droids you’re looking for” messages.

Tell Google How to Treat Your Content : Disallow, Nofollow, NoIndex
Ann Smarty has been causing quite a stir on Sphinn with her Search Engine Journal posts. As near as I can tell, she is repackaging SEO advice she likes in most of her articles. There is nothing wrong with this approach to becoming an SEO. Aaron Wall did that when he created his SEObook. A lot of people like Aaron and I get the impression a lot of people like Ann.

Still, I had trouble understanding the point of this article. Okay, I’ve written my fair share of incomprehensible gibberish through the years, but this article got 40 Sphinns and it makes absolutely no sense whatsoever.

It’s like the opening paragraph of what promises to be a great tutorial. We get some bullet points, a few pictures, and then BAM! (sorry, Emeril), we’re done — oh, btw, there’s a link to a Sebastien article.

There have been days when I’ve read Ann’s articles and thought, “Hey! Didn’t I just say the same thing yesterday?” Of course, she says it more concisely. If I were to rate her SEO judgement on a scale of 1 to 5 (1 being bad and 5 being great) I’d give her at least a 3, maybe a 4. She picks some good topics to repackage and obviously her readership appreciates her work.

Nonetheless, once in a while I find myself commenting on Ann’s articles because I cannot contain myself. She leaves out something, or maybe advocates a position that she doesn’t have enough experience to know the pitfalls associated with it.

Don’t get me wrong. I like Ann Smarty’s articles and there are plenty of SEO repackagers who should wish they did half as good a job as she does. But this article just left me wondering what the point was.

Conclusion

Okay, we all know I can be pretty harsh when criticizing my fellow SEOs. The first time Todd Friesen participated in one of our SEO classes here at Visible Technologies he cautiously approached the white board and said to the group, “This may not be something Michael would show you, but it works.”

True enough. But Visible Technologies hired Todd because he does stuff that works. He’s earned the right to get up in front of people and tell them stuff I might not show you. In fact, I would share a white board with Todd Friesen any time, any place.

There are a lot of ways to do what we do.

Nonetheless, if you’re going to write about SEO then you’d better be prepared to take your lumps. There are plenty of critical idiots out there who will slam you just because you don’t rubber-stamp the nonsense their online friends spout. But then there are more discerning people who really want to know why you think something works or what your point is. If your SEO experience doesn’t extend much beyond title tags and asking for links, you have a LOT to learn about search engine optimization.

Most SEO advice is no better than average, and the average SEO advice tends to be repackaged ideas that have caught the community’s attention. I personally find no value in an SEO article that says nothing new, unless it finds a new way to say something — a better way, even.

SEO advice usually sucks so much because so much of it is written for Sphinn, for SEO back patters, and for the sake of obtaining links from other SEO blogs.

I probably didn’t need to pick on anyone else to say that, but I thought it would be helpful to provide some examples.

That said, you could do a lot worse than read Ann Smarty’s stuff. The other guys, I just haven’t read anything else by them (that I can remember).