joi, 17 mai 2012

9 Lessons from 1,000 SEO Questions

9 Lessons from 1,000 SEO Questions


9 Lessons from 1,000 SEO Questions

Posted: 16 May 2012 12:08 PM PDT

Posted by Dr. Pete

I spend a lot of quality time in Private Q&A here on SEOmoz, and I recently passed a milestone – 1,000 private questions answered since we re-launched the system (just over a year ago). Not surprisingly, we see a lot of the same questions and concerns pop up over time, and I’d like to think I’ve learned a few things along the way (please tell me my suffering wasn’t in vain). This post is an attempt to distill the biggest lessons from those 1,000 questions…

1. Dogma Will Get You Killed

You finally got your head around SEO best practices, and then you tackled your first e-commerce site, only to find that nothing worked the way the blogs told you. Search is algorithmic, so we assume it follows the same rules for everyone. In theory, it usually does, but those rules are incredibly complex and situational. Google claims over 200 ranking factors, many of those factors are probably multi-part, the algorithm is changing more than once per day, and there’s occasionally a manual intervention to really screw things up.

It’s good to know the basics (and there are some best practices), but you have to learn to roll with the punches. Even something as “simple” as de-indexing a few dozen pages rarely goes as planned, and can take weeks or months. Measure, evaluate, and adapt. If one tag or tactic isn’t working, consider your options.

2. One-trick Ponies Make Good Glue

I wrote an entire post recently on this topic, specifically link-building vs. on-page SEO. People naturally get comfortable with one aspect of search marketing (link-building, on-page, social, etc.) and then want to “perfect” it, but at best they hit diminishing returns fast. At worst, they’re putting band-aids on URLs while they bleed to death from a huge link wound. I’ve seen sites with spotless on-page SEO that have been stuck for months suddenly leap through the rankings because they’ve acquired a few good links. On the flipside, I’ve seen sites that were a total mess but had solid link profiles miraculously improve when their on-page problems were fixed.

3. A Link, by Any Other Name…

…might still stink. In the rush to build links, too many people, especially people with brand new (read that “highly vulnerable”) sites, make the mistake of thinking that all links are equally good. It’s no mistake that my most linked to blog post in Q&A is Rand’s 2010 post “All Links are Not Created Equal”. It’s not just a question of spam and penalties – link value varies tremendously with the page, placement, density of links, and on and on.

Case in point: I can’t tell you how many people I’ve seen spend months on a DMOZ link only to have it buried on a page that has little or no internal PR or isn’t even indexed. Link-building is not just a numbers game. I’m not making a white-hat argument – it’s just SEO fact. Some links are better than others. Don’t waste your time on junk.

4. You’re Not a Black-hat Genius

Sorry to break it to you, but better to hear it from me than Google. First of all, if I can spot your paid links and gratuitous spam in 5 minutes of looking at Open Site Explorer data, how hard do you think it is for Google, who can essentially see the entire link-graph at a glance? Obviously, they don’t always get it right, and plenty of spam slips through the cracks, but the algorithm isn’t stupid, either. Ethics aside, the practical problem with black-hat SEO isn’t that it doesn’t work – the problem is that 98.7% of people do it badly.

At the risk of kicking you while you’re down, I also have to add that your link circle/wheel/tetrahedron isn’t brilliant, no matter what your mom says. Just because you’ve cross-linked 157 Squidoo lenses doesn’t mean that you’ve built an impenetrable web of black-hattery. If your link wheel were a Disney movie, the theme song would be “The Circle of Crap.”

5. On-page Is Getting Messier

I keep wanting to write a post on Google’s recent advice about pagination (and rel=prev/next), but then I get so angry I’m afraid I might turn green and start fighting alongside Iron Man – not that that wouldn’t be awesome. The problem isn’t that they’re wrong (although I think the advice is horribly over-generalized and often ineffective), but that they’ve put a tremendous burden on webmasters. Implementing a proper canonicalization + pagination scheme on a dynamic site with hundreds of thousands of pages is incredibly complicated, and requires not only substantial development resources but stellar communications between the SEO and dev teams (if you’re lucky enough to actually have teams of both). Add in HTML5, schemas, and the whole mess of other new options, and it’s only going to get more complicated.

6. Check Your Headers

Sorry, that wasn’t particularly helpful, so here’s an easy tip. When something isn’t going right and you don’t know why, check your page headers. Job #1 is to make sure that crawlers see what you see (or think you see). It’s unbelievable how often a problem comes down to a bad redirect, status code, or other crawler accessibility issue. There are tons of header checkers, from web-based to bookmarklets – I still use this header checker over at SEOBook.

7. Use Basic Tools Well

There are some great SEO tools out there, but I see the same issue in SEO that I do in writing, time management, and basically every single 21st-century human endeavor. We’re so busy chasing shiny new tools and the perfect app that we don’t bother to learn how to use any of those tools effectively. You can go a long way with a solid header checker, Google’s “site:” operator, a link analyzer (like our own Open Site Explorer) and a desktop crawler (I highly recommend Screaming Frog, but Xenu is still great, too). Master the “site:” operator and learn how to use it with “inurl:” and “intitle:”, and it’s amazing how many on-page problems you can diagnose. Stop chasing every new tool and learn how to use a handful really well. You’ll save a lot of time, money, and holes in your drywall.

8. Learn When to Be Patient

Patience may be the toughest skill any good SEO eventually has to learn. There are times when you’ll need to react quickly to a problem, especially a technical problem (like a bad redirect or site outage). There’s a fine line between reacting and over-reacting, though. One of the most common mistakes I see in technical SEO is when someone makes a change, it doesn’t immediately improve their rankings 24 hours later, and so they revert it or make another change on top of it. Even if it doesn’t make the problem worse (and it usually does), you’ll never be able to measure which change worked. Make sure your changes went live, that Google has acknowledged them (i.e. crawled and cached), and that you can measure the impact or lack of impact. Don’t change your strategy overnight based on bad information (or no information).

9. Stop Scheming & Get to Work

This post was originally “8 Lessons…”, but when I wrote #4 I got so annoyed that I had to follow it up with maybe the most important SEO lesson I can teach you. Are you ready? Here it is (warning: this may be inappropriate for younger readers)…

DO THE FUCKING WORK.

The most frequent excuse I hear in Q&A is “I don’t have time to…” Let me ask you something. Isn’t this your business we’re talking about? Isn’t it your livelihood? Isn’t it the thing that puts food on your table and clothes on the backs of your children? You’d better damned well find the time. If 80% of your traffic is coming from Google, and you don’t “have the time” to do the hard work of improving your product, creating unique content, and participating in your industry, then here’s the simple truth: no blog post is going to save you.


Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!

Niciun comentariu:

Trimiteți un comentariu