Well yes. That is entirely true and I whole heartedly agree with you but how do write an algorithm that determines if your content is good (much less give suggestions on making it better)? I'm really looking for ways to help people identify potential problem on their site automatically.
I do work for real estate clients in regards to SEO and yes good content and frequently updated content is very important. Blogs and such are great ways to add new content and I have several sites where I pull an RSS feed into the site to display new dynamic content.
In regards to rules of SEO, these are the things I look for when optimizing a page:
Their is much more I look at, but the best key to SEO is still generating the page for the user and not the spider. While we have important aspects we can influence, I still find it best practice to optimize a page for the reader, of every type.
Thanks. That's what I'm looking for. If you wouldn't mind sharing the "much more I look at" stuff that would be great too! /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" />
I've been trying to think of a way to detect the quality of the content on the page from a users perspective... but that is very very difficult.
Sun'd: I had never thought of that being good from a SEO standpoint. Makes sense that google would value that though - I know that it's very good practice from a web copy-writing perspective (for visitors coming from search engines) as it enables them to immediately see what was advertised.
Thanks for the tip!
I know content is king but that doesn't mean subordinates don't have a lot of influence /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" />
We recently published an article on our Church Marketing Online blog entitled What is SEO? and a more detailed article about on-page optimization, which is the part of SEO you seem to be primarily interested in. I think you'll find those articles helpful.
To be useful the tool would need to count how many times the keyword(s) occur in the the various elements of the page that matter and then score each according to the importance of that element. There may already be some free tool available that do that. You might want to google "on-page optimization checker" or "keyword density tool."
Keep in mind that even if a site scores well for on-page optimization of a particular keyword, that is only one part of SEO. Keyword research, link building, and analytics are other aspects critical to good SEO.
Paul:
Those articles look nice but don't really provide anything I don't know already. Thanks.
That would be a simple implementation of a keyword scoring algorithm... I might do something like that to begin with.
The program I'm working on primarily focuses on identifying problems (potential or otherwise) with your pages/site independent of the internet. There are plenty of tools out there that find how many sites are linking to yours and all that other stuff (most of which info is actually gathered illegally (against google's TOS)). So far the rules I've implemented to date include:
img rule:
Identifies img tags that don't have alt or title defined.
title rule:
Identifies pages with duplicate titles and ones that are very similar to each other.
header rule:
Identifies if prominent headers tags are lacking that are related to the page's title (Elliot's idea /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" /> )
Here's an article by some well-respected SEO practitioners - http://www.seomoz.org/article/search-ranking-factors
A link to an online tool - http://www.seoworkers.com/tools/analyzer.html
And, a web site with plenty of info and help from one of the top SEO dudes - http://www.seobook.com/
Another quick online source - http://www.websitegrader.com/
Yeah, I've seen most of those too /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" />
It's not that I'm knew to the industry, just looking for some concrete definitions of what people call "SEO" /tongue.gif" style="vertical-align:middle" emoid=":P" border="0" alt="tongue.gif" />
I think there is no 'concrete' definition - it varies with each search engine and there are various strategies for optimization. And, is your goal really to improve your search engine results, to get more visits/hits, or to increase the number of contacts you have with visitors?
There are onsite and offsite factors that influence search engine placement - how will you measure them? Some of the sites listed in this thread do plug into the API's provided by various search engines to measure such things as inbound links, links from .edu and .gov sites, age of the domain, etc.
I think it best for web masters to focus on the content they produce - does it really target the audience they are trying to reach? Are the search terms that lead to the site keywords you have selected? How many people visit the site as the result of a search? How many from links on other sites? How many access the pages directly? If 80% of your visitors have your site bookmarked and your audience is growing, do you really need to put a lot of effort into better search engine results?
I decided to release my program under the GPLv2
You can grab it here: http://freehg.org/u/joey/seojo/ if you want to check it out.
For linux users you need python, pygtk, simplejson (unfortunately it doesn't install properly with easy_install so you'll have to put it in your path manually) and I think that's it.
For those unfortunate enough to be using windows (I got vista and it's only a slight improvement over XP, so I know all the pain you guys have to deal with) I'm working on getting it packaged into an exe. I've been running up against some problems though related to the threading and a few other things. So it probably won't work AS well (but still be fully functional) when I do get it packaged.
I would not recommend trying to get it to run yourself on windows unless you already develop with python and gtk AND know enough about threading to fix the problems /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" />
Mac users SHOULD be able to get it running easily enough... I haven't used a mac for about 8 years though.
If you have any suggestions or want to contribute send me an email.
~Joey
Alrighty - I finally got it to work in windows!
The download is slightly bigger (23MB apposed to the linux/source download of >1MB /tongue.gif" style="vertical-align:middle" emoid=":P" border="0" alt="tongue.gif" /> ) due to having to include the GTK runtime etc.
http://joey101.net/static/SEOjolight.zip
I could probably cut it down a lot but it takes a lot of work to test because I have to do so on a windows box that doesn't have all the runtimes installed. So go ahead and use up my... almost 2 TB now... of bandwidth dreamhost gives me... /tongue.gif" style="vertical-align:middle" emoid=":P" border="0" alt="tongue.gif" />
If it doesn't work it will create a seojo.exe.log file that you can send to me that will include the error.
There is one weird problem where on exit it says there was a problem but the log is empty, so don't send me that. /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" />
Rude, critical, worthless and/or helpful feedback is all welcome alike.
(I know that some of those combinations don't make sense but it's better than saying: "rude and/or helpful, critical and/or helpful, rude and critical in possible addition to helpful, rude and/or critical" etc)
Oops, I forgot to mention. After it scans a site I had to disable automatic selection because it causes the thread it's in to freeze (I haven't gotten to the bottom of it yet). So after scanning and rescanning you need to manually select the site for the display to update.
Ok, I did some significant testing and bug fixing.
Unfortunately I'm not sure it works on XP. So if anyone has Vista (I only have one Vista box here) and tell me if it actually works that would be awesome. Also if it works on XP (the computer I tried it on might have had other issues preventing it from running). If launching from the start menu doesn't work (it did that on the XP box when I tried it) please try running the exe directly (C:\Users\You\Program Files\SEOjo\seojo.exe (slightly different on XP))
I created a installer that you can get at: http://joey101.net/static/SEOjolight-setup.exe
(uninstaller included /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" /> )
And the source is still there and updated so if anyone has a burning desire to easily develop desktop applications with one of the best cross-platform widget toolkits and the uber cool programming language python it's there for you to learn from. I've been told I write very clean code that is easy to follow... I can't say it's that way for SEOjo yet - it's not that it's bad it's just there are a lot of hacks the file structure needs to be factored a bit.
Hello all,
It's been a while since I've posted here. I've been around the world and am back living in the USA (Texas; where over 70% of the people speak English! YEAH! (I have nothing against Spanish, I just like to be able to "entiendo gente")). Anyway I have a question that I thought I'd never ask.
What is SEO? I've been trying to nail it down into a set of rules. Let's back up a bit...
This past weekend I brushed up on my GTK skills and created an application to help with SEO work. It's a rule based system where the application spiders a website and passes each page though the rule parers (a rule is not restricted in any form so it can report an ambiguous message to be displayed for the page). I've spent most of the time on re-learning how to use the TreeView and TextView/TextBuffer... I've only spent about 10 minuets on the rules... er rule.
The one rule I have takes all the img tags and makes sure they have an alt and title attribute assigned. One rule I've thought of that I haven't implemented yet is a algorithm to detect prominent keywords on the page. But that's it.
I've browsed around the internet for real definitions and concrete practices that make up this elusive "SEO" (Which is named horribly as it's impossible to optimize something that you can't reliably and accurately measure the results of, but that rant is for another time). The only two I've found have been, guess what - alt/title attributes and keyword analysis.
All other practices of SEO that I have found are either breaking google's TOS (yes, that includes pagerank reporting) OR far to ambiguous, or rather undefined, to be able to put into an algorithm.
So my question is to all you people who do SEO work for their clients (and heck, those that don't), what IS seo?
Thanks!
Joey
My blog about web development.
And my site about game development.
>>> math.sqrt(-1) == joey101
True