Search Engine Optimisation :: Frequently Asked Questions
SEO FAQ Page 5
2. Why doesn't my site rank well for the keywords I target?
4. How do I optimise the site copy? Where do I place my keywords?
5. How do I build links pointing to my site?
6. How do I optimise the <title> tag contents?
7. What are meta-tags? Are they really important for SEO purposes?
8. What is search engine spam? Why is it a bad idea? Is there a SE spam classification?
10. How do I optimise the site's navigation for SEs? What is anchor text?
11. How soon will the search engines find my site?
13. Such and such a site ranks #1 for my targeted keywords, but it's no better than mine - why?
17. How do I create a spider-friendly site map?
18. Do spiders follow image links?
19. Do spiders crawl dynamic sites?
Why is it believed that all the directory submissions should be done manually? There are so many auto-submission programs around - and manual submission will take a lot of time, won't it?
Directory submission - as with everything in SEO - is only good when done artistically. It means varying titles and descriptions; it means choosing the most relevant category for your site in all cases. The last task is not as easy as it may seem: all directories have different category structures.
It is good style to read the site-specific guidelines before you start submitting your site. Like everything else, guidelines differ slightly from directory to directory; your auto-submitter is, obviously, incapable of reading them. Many directory owners attach a "test for humanity" to their submission forms: you will have to enter certain characters, shown to you as graphics, manually. That's done in order to stop auto-submitters, as such practice is commonly recognised as spam.
How do I create a spider-friendly site map?
Make it using pure HTML, not JavaScript. If you like JavaScript-powered site maps, add a static alternative in the <NOSCRIPT> tags. It will be visible to search engines, as well as to the Internet users who prefer to turn the scripts off for security reasons.
If you wish to make it in static HTML, it is good to indent entries according to your site's structure: let's say, your home page is Level 1; pages directly linked from it are Level 2 (with a 5-pixel indent), then comes Level 3 pages (with a 10-pixel indent) and so on. Such indents will improve the usability, and you can use CSS to achieve this.
Be sure to include keywords in your site map links. It will happen naturally if your pages are well optimised, as on-page optimisation includes headers; if you've done everything properly, they should already be keyword-rich. Just reproduce the headers of all pages on the site map - and everything is done. Do not, though, include multiple links to the same page with the sole purpose of including slightly different keywords: it is a spammy technique.
Do spiders follow image links?
Yes, they do. To improve your rankings, add an alt attribute to such images; but we would recommend using plain keyword-rich text links whenever possible.
A mini-sitemap containing links to your most important pages placed at the bottom of every page is, therefore, a good practice; just do not overuse it.
Do spiders crawl dynamic sites?
Again, yes, they do. They do not like parameter strings with too many parameters (generally, more than two parameters at once). And they hate session IDs, though modern search engines can handle such situations.
So, avoid showing session ID parameters to spiders whenever possible. Try to use POST instead of GET. Make sure different URLs do not point to the same page (spiders will treat them as duplicates); if you think you cannot avoid this, add a <META NAME="robots" CONTENT="NOINDEX,NOFOLLOW"> tag dynamically to all the excessive copies of your pages or hide them from spiders via the robots.txt file. There is no need to waste spiders' bandwidth by making them crawl the same content several times: they will filter such pages in their SERPs anyway.
And remember that SE spiders do not support cookies, so if your application requires them, it may become non-spiderable.
Do I hurt my rankings by linking to other sites?
No, you do not, unless you have too many links to bad neighbourhoods. The Internet is supposed to be interlinked, so an attempt to penalise for linking outside would kill the very idea at once. Quite on the contrary, many experts will tell you that if your site is reasonably linked to good, relevant resources, it will, quite probably, help you with your rankings. Our own experience confirms that.
There is nothing wrong in being generous.
What is the "Google sandbox effect"? Can it be avoided?
There are several different Google-specific effects recently noticed by SEOs which are somehow connected with the time/age factor. For simplicity, they all come under the umbrella name the "sandbox effect", but they are not all the same as each other.
Most often, when people say "sandbox effect", they refer to the effect experienced when new sites don't receive any decent Google rankings for any valuable keywords during the first months of their existence. This effect is also known as "aging delay" or "aging filter" and was first noticed after March 2004. Sites that had been launched earlier weren't affected.
Some people say that the aging delay effect can be avoided. So far, I know of only three examples of sites that actually avoided the effect completely. Two of them were pure blogs, and the third one a pure forum, which, I think, shows preliminary evidence of a pattern. Regular sites (neither a forum nor a blog) do get delayed, though in the non-competitive niches they can appear in the TOP 10 for some of their targeted terms quite soon (in about three months). In more competitive niches (like SEO or real estate) the process usually lasts much longer (8 to 14 months).
The aging delay effect wears off step by step and requires a lot of patience from the site owners.
Another type of sandbox effect is the so-called redesign sandbox. When the site gets completely redesigned it is in danger of losing a lot of previously acquired Google rankings for an undefined period of time, but it doesn't mean it happens to all sites.
Sometimes the complete redesign doesn't affect the rankings at all. It is recommended to preserve old URLs of pages, if possible. If not possible, a 301 redirect from the old locations to the new ones (per page) is very strongly recommended. If it doesn't help to avoid the redesign sandbox completely, it should at least minimise it.
The third manifestation of the sandbox effect is the trial period applied to all new inbound links. It means that when the new inbound links are added, they don't immediately start giving any ranking boost to the site they point to. They start working partially in about a month and a half and then, step by step, their weight increases.
People who like to jump-start their linking campaigns sometimes complain that their sites suddenly don't show in the TOP 1000 of Google for any terms, and assume it is because their links trigger yet another type of Google sandbox. It may or may not be true, but jump-starting link campaigns is not a good idea any more. It doesn't comply with natural linking patterns, which are now valued most.
The last type of Google sandbox effect comes into play when webmasters use the 301 redirect to merge their several websites and point them from the more authoritative sites to the less authoritative one. In such cases, it can take up to three months before the full authority of the redirected site gets transferred to the main domain and the rankings recover.
See also
SEO FAQ Index
[ Page 1 | Page 2 | Page 3 ]
[ Page 4 | Page 5 ]
Featured SEO Articles
Avoiding Bad Neighbourhoods: How to Link Out Safely
301 Moved Permanently: How to Use It Correctly for SEO



