SEOmoz released its latest SEO quiz today, which is now getting some buzz as folks poke at the accuracy of the questions. Below, my own spin through the test, which constantly had me disagreeing with many of the "answers," plus some links to more of the chatter about it.
Some of the disagreement is in comments in a post about the quiz at SEOmoz itself. Over at our Sphinn forum site, we have multiple topics arguing about the quiz (1, 2 and 3) Former Googler Vanessa Fox took a swing in her Why the SEOmoz SEO Quiz Is Completely Wrong post, giving me a big chuckle at the part where she found herself getting an answer wrong, only to have the proof of her wrongness be a quote from herself!
When it came out this morning, Barry and I talked about going through it, and Barry said he might do that tomorrow. He still might, but I ultimately decided I wanted to go through it myself. I intended to go through each and every questions, but I lost energy after the first 10. Still, I think it will be a useful look over why it can be so hard to come up with a quiz with exactly the "right" SEO answer.
Let’s dive in at question 1:
1) Which of the following is the least important area in which to include your keyword(s)?
You could chose from Meta Description, Body Text, Meta Keywords, Title, Internal Anchor Text. Me, I went for the meta keywords tag, since as I explained just recently (Meta Keywords Tag 101: How To "Legally" Hide Words On Your Pages For Search Engines), only two of the four major search engines even use content there for retrieval purposes.
2) Which of the following would be the best choice of URL structure (for both search engines and humans)?
OK, so you get five choices. I know the answer the quiz is after. It wants you to go for one of the three choices that use keywords in the URLs. In particular, the "right" choice will be wildlifeonline.com/animals/crocodile because to a human, that suggests an order (here are all our animals, and within the animal area, our crocodiles).
The other two "keywords in the URL" choices aren’t as neat. Then you’ve got two URLs with no keywords. One is using parameters, you know, like wildlifeonline.com/blahblah?id=1234. The other is a nice short URL with a number like this, wildlife.com/563.
Well, avoiding parameters in URLs is good advice, though search engines are much, much better at it over time. Keywords in URLs have commonly been consider to provide at best a very slight ranking boost, so I suppose one of those three choices makes sense. BUT! You also want to make sure your URLs don’t change, if you can help it. Saves you from maybe getting 404 errors, losing link love without setting up redirects and so on. Now say you use blogging software like Movable Type. Keywords in the URLs are all nice and good until you discover changing a post’s title can change the URL. In that case, maybe you want what I call rebuild safe URLs. That can mean numbers, and so numbers could arguably be better for search engines and humans (since humans don’t like broken URLs, either).
3) When linking to external websites, it’s wise to use the keywords you’re attempting to rank for on that page as the anchor text of the external-pointing links.
It’s a true or false. The answer is supposed to be that if you want to help your own page, link to the other page with the words you want your own page to be found for. I think. It’s a bogus question, though. When linking to external sites, you should use anchor text that explains to those clicking on the link what to expect. Alternatively, if you want the external site to rank for certain terms, then make sure those terms are in the anchor text. But this stuff about getting your own pages to rank by the outbound anchor text? Makes my head hurt.
4) Which of the following is the best way to maximize the frequency with which your site/page is crawled by the search engines?
The answer is put your head between your legs and kiss…. I mean, you don’t have a lot of control over this. One answer is to "frequently add new content," which might help search engines realize they should keep coming back. But then again, if you’re some new site with no authority or reputation, they ain’t going to come a callin’ just because you add a lot of stuff. That would mean they constantly are letting time get sucked down by sites that would just post anything in hopes of getting listed. It’s probably the best of the choices, but it’s not guaranteed. Saying "best way" would be better qualified as "a way that might help."
5) Which of the following is a legitimate technique to improve rankings & traffic from search engines?
"Submitting your domain to the top 2000 search engines on the web" isn’t a choice I’d go for — but neither is it an illegitimate technique. I mean, using one of those tools that submits to 2,000 search engines isn’t going to bring you traffic when only four or five of those search engines has significant traffic. Nor does submission generally do much to ensure ranking. But it’s not spamming to do this. And adding "keyword-dense meta tags" isn’t wrong if the density means the tags contain a variety of different but relevant terms. Again, probably won’t help you — but it’s not spamming.
The right answer is clearly supposed to be "re-write title tags on your pages to reflect frequently searched, relevant keywords." Though, that’s not right either — it’s really, rewrite title tags to include the key terms you want each particular page to be found for, with the assumption you also use those terms as part of the body copy of the page.
6) "Don’t Make Me Think" is a must-read book about:
I admit it — I didn’t know. And I didn’t know I should know the book as part of my SEO knowledge. Nor do I think everyone would agree you should. FYI, I guessed it was about site design and usability.
7) Which of the following is the WORST criteria for estimating the value of a link to your page/site?
Sigh. Worrying about whether the link is useful at all, perhaps. I mean, it’s a link. If someone’s going to give you a link, take it. If you’re asking should you request a link — or buy a link, then sure, you might want to evaluate whether it is worth the time or cost involved. Of the choices, we’re suppose to say the Alexa score is the worst criteria, since Alexa ratings are commonly seen as unreliable. OK, I can roll with that. The better answer is that all the other criteria (is the page ranking? is the page including in an index) all relate back directly to the search engines themselves.
8) Why is it important for most pages to have good Meta Descriptions?
Interesting — I totally agree with the right answer, that "they serve as the copy that will entice searchers to click on your listing." But, sometimes they aren’t used, so that’s not entirely right. And you know, we don’t use them here (that’s actually going to change in the near future). So saying it’s important might be better phrased, "How might meta description tags help with SEO?"
9) Which of the following content types is most easily crawled by the Search Engines?
Gosh — Windows Player Media files (Windows Media Player actually plays a variety of different files), executables, Flash, XHTML and Java applets. I’m going with XHTML, which is basically (to my understanding) a cleaner, more-standards compliant form of HTML. But you want to get technical? First, which search engine are we talking about? I mean, there are media search engines that will crawl media files. Second, crawling isn’t necessarily the issue – it’s INDEXING. They can crawl a link to a media file, but they might not be able to see inside to index it, store the content for retrieval.
10) Which of the following sources is considered to be the best for acquiring competitive link data?
You’re supposed to say Yahoo, since Yahoo Site Explorer provides really good backlink data for any site. Google only gives you a sample. Ah — but what’s more important, a sample of links that’s probably skewed toward the ones that count more or a big chunk of everything.
Beyond The First 10
After the first 10 questions, I thought I’d then only comment on specific questions I had problems with. Continuing on, that led me to:
16) Which of the following is NOT a "best practice" for creating high quality title tags?
So I’m supposed to say "include an exhaustive list of keywords," but I disagree that "include the site/brand name at the beginning or end of the title" makes sense. I’ve seen too many pages where the site name in the title actually detracts from me wanting to click through. Sometimes it helps, but not always.
17) HYPOTHETICAL (for this question, assume that "link value" refers only to the Google search engine): If placing a link from page 1 to page 2 provides page 2 with X amount of link value, what happens if two links are placed on page 1 pointing to page 2?
Seriously, hypothetical — and yet there’s a correct answer? According to what, the PageRank formula of 1998? I guessed that the link value might remain the same.
And Enough! Focus On The Wrong Answers
At this point, I ran out of energy. Some questions were clear cut (PageRank was named after Larry Page), while other questions had me sighing because there simply was no correct answer, in my opinion — only opinion (such as the choices for "Spammy sites or blogs begin linking to your site. What effect is this likely to have on your search engine rankings?").
There was actually a lot of sighing as I continued on the quiz. Eventually, I made it: 184/257 points, 72 percent. I suck at SEO. Or do I? Let’s look at some of my wrong answers:
For #3, I was told "Linking out to other pages with your page’s targeted keywords in the anchor text can cause keyword cannibalization and reduce your page’s chances to rank well at the search engines. It also creates additional competition for your page in the search results, as you give relevance through anchor text and link juice to a competing page."
Well, I could maybe agree a bit with the second part of that, but the first? Sorry, doesn’t win me over.
For #17, I was told "Page 2 receives an amount of link value greater than X, but less than 2X. In testing from multiple parties (SEOmoz and a few of our friends), we’ve found that Page 2 will receive more value from two links on page 1 than from only a single link, but nowhere close to "double" the link value. This appears to be consistent across all of the engines."
Um, OK. You asked a hypothetical question that apparently really did have an answer. I guess it should have been phrased like, "We’ve done some testing in in these cases, which do you think we found?" And I haven’t tested this particular situation, but if I did and come up different than SEOmoz, would I still be wrong?
For #20, I picked Yahoo as patenting "TrustRank" rather than Google. Dang it – I knew it was Google! But it was late, and I questioned myself.
For #22, the idea here was what solution was needed to avoid duplicate content issues when you’re in a blogging environment, where a story moves down and then off the page. I said there aren’t issues. Instead, I’m supposed to add "noindex" to the paginated pages. At this point, I’m like — "what the hell are you talking about." I mean, what’s a "paginated" page? At Search Engine Land, each story has precisely one URL. It gets listed one the home page, then moves down and off. It’s not going to get duplicated in its entirety by any other pages. So looking at that environment, damn right — I didn’t see any issue.
For #24, I said if you were worried about "keyword cannibalization" (which I had to look up — it means competing with yourself for the same term, and it’s not phrase I’ve typically heard in the industry in my years (and at only 1,300 matching pages, I don’t think I’m alone in this), you should prevent your less important pages from being indexed. Here, I was trying to think like someone who is being really ruthless about SEO, kill of the weak. The "correct" answer is apparently to link from those "weak" pages back to the stronger ones. I suppose….
For #25, the "de-facto version of a page located on the primary URL you want associated with the content is known as the ‘canonical version’," says the quiz. Bull, sez I. See, canonicalization is special to me because of my inability to pronounce the dang word, for one. I’m much better at it now. But I hated it because I had to talk about it and how it means that the search engines will pick what domain version to use for your site (like you have www.mysite.com and mysite.com, they’ll go with one of those). But see, that’s the search engine making the choice — not you. The choice you want is not the canonical version. I don’t know what it’s called across the board, but Google calls it the "preferred domain." And that’s why I answered "preferential version" as the closest of the bad choices I had. And got it wrong.
I got more wrong, some of which I could argue successfully against, and a few where I’m like "Oops! You got me."
Overall, I appreciate the intent of the quiz, which was to further educate folks. And I’ve generally found these types of quizzes fun. But I guess I’m all quizzed out now.