Forum Moderators: open
I have an SEO client who have a state of the art web site. They love it. It was very expensive
(you can probably see where I'm going with this) ...so they don't want to change it.
To make matters worse, I replaced an inept SEO who did a bothched job on this site. This "specialist" simply added a few meta tags, submitted to SE's and took off. No workarounds for the scripted links, dynamic pages, frames. Nothing. Needless to say, the client is far from happy.
So, I'm trying to sort this out.
Cloaking is out of the question as this is a big company with a very strong brand and who are risk adverse. There are also technical issues that make this a non-starter.
I was considering creating a themed satelitte site that would act as a doorway to the main site. Are there any examples of these floating about? How well do you think this approach would work?
Ideally, I want to integrate static, themed pages into the main site but the developers are loathe to accept this option.
Any help and sympathy from SEO professionals who have been there, done that, greatly appreciated ;)
(edited by: feeder at 2:17 am (gmt) on Sep. 10, 2001
The developers are very sensitive so the problems are mostly political. I'm trying to create a situation where the marketing and sales people within the company will be able to over-ride the developers, but this is taking time.
I agree that vanilla pages are the way to go
but wondered if there was any workable alternative.
I really feel your pain, trying to make the MKT people overide the IT guys will make it worse. Many techno stuff is there to help, but the solution must come from inside developpers, if any.
To watch sparks fly and try to control the burn on some meeting will not keep your hands clean but it will lead to results. Results are expected from above...
<rant>In our favour what I find is that the customers are moving on from "we need a website!!" to "what are we doing with this website??". I'm finding more and more people at higher levels understand that marketing strategy and business process is integral to their web efforts and its not enough just to have a great looking site. Unfortunately, this hasn't filtered down to a lot of developers who seem obsessed with using all the new toys and pumping up their C.V's. I don't blame them for this, but when a client is trying to solve a business problem, and paying you good money to do so, then you should focus on their needs. If you want to be an artist, fine. Do it on your own time.
</rant end>
Anyways - something will have to give. Either they want search engine listings or they don't. This won't be pretty....
I bumped into similar situation awhile back and told the client to give one of their products. I developed for that one product and they are making money (forget bells and whistles, it's about the cash). Since then, they gave me the main site and junked the old one with more "product sites" to come.
Any other technical approaches you can think of that I might be missing? The site has got frames so I might be able to work the <no frames> tag a bit. Trouble is, only some of it is in Frames, so I still have linking issues. Any great ways around javascript links without altering how the page looks?
Try to steer clear of "satellite site", even if cheaper, they need some solution for the current site. If you offer some alternate solution on some other site, they will see you shooting at the wrong target.
Spot the second most important keyword (else from the brand name) in the site and make a couple of static versions linked to the homepage. The stats will call the shot.
Its a dynamic site in Cold fusion. Javascript links. Part of the site in frames. All of the site driven from a database. Spiders have already got stuck in it presumably because each request generates a new URL (I would be happy to send you the URL only I am under a client confidentiality agreement for all SEO work). Any other details you need?
I usually handle this situation by creating a mini static themed site within the site itself. Thats what I'm pushing for but it's going to be an uphill battle.
your problem may not be the dynamic content per se, but rather the dynamic URLs.
Pardon me if you already know this, but lets say your site is about widgets, and you have three colours, red blue and green. At the moment you may have a page called products that calls up the widgets thus:
products.cfm?type=widgets&colour=red
products.cfm?type=widgets&colour=blue
products.cfm?type=widgets&colour=green - and so on.
If you create pages such as redwidgets.cfm, bluewidgets.cfm, greenwidgets.cfm and hardcode the database request into the page, you have a static looking page you can optimise which the spiders will like eating, and yet the content can still be dynamic from the database. I do this a lot with no problems and good positions.
just a thought.
Onya
Woz
"If you create pages such as redwidgets.cfm, bluewidgets.cfm, greenwidgets.cfm and hardcode the database request into the page, you have a static looking page you can optimise which the spiders will like eating, and yet the content can still be dynamic from the database"
Not sure I follow but it sounds great. How do I "hard-code the databse request into the page"? Can you point me to any examples?
Cheers :)
Have a look at the site listed in my profile at the end of "Interests" - it's the one starting eith "G", not the linked one! This is a directory with the content coming from a database. The search string looks for information in variable which is usually in the URL string. However, in these cases I have set the variable content in the page code. This is done with each page and so keeps the spiders happy. laborious but it works.
You would need to understand the scripting to get this working, perhaps the IT people can give you a hand.
Onya
Woz
What I tried to do, was get the client to find some obvious breaking point in their content. Sometimes that can be as simple as support related, sales related, tech related, or other logical divisions. Then get them to give you one of those divisions for a fresh site.
The only way I know is to make some hallway pages. After the "?" in URL issue is fixed by our friendly developpers, you will need some HTML links to all the pages hidden behind JS. I try to make those doorways under 20 k and I put NOINDEX,FOLLOW in the robots meta. The text used as a link to the page should be related to it's best potential keyword. I put 10 to 20 links per page plus a footer that cross link all the hallway pages together. I put some hidden link on the home page to one of them and they all link back to it.
This will also help for framed pages, just make sure to redirect them to the proper frameset using an external JS.
Can you give us a rough idea of the size of the site? Also, I would like to know how "clean" is the coding. Is there ten gazillion of JS or CSS lines before the actual body appears?
So many "corporate sites" are developed from the presentation point of view, a lot of fluff about execs and advertising copy that does nothing for the client in the realm of traffic generation.
If you break off specific products or services you can be accountable/rewarded for the efforts you put into SEO. It also gets some of the heat off your back from the internal org. while building it. When the baker sees that "pumpkin pies" are flying off the shelf it is no longer an argument with the IT/Art dept. who is struggling with the main "bakery" site, they can't argue with profit.
We then explained the client the steps taken to make sure they have visitors, and we showed him a 80 pages internet market study. The funny part was that some home constructor was present and reinforced what we said. He got a website 8 months ago and only recieved 1.8k visitors.
My clients now use SEO arguments to differenciate from competition. We picked 20 of the most profitable keyprases and the 60k contract was inked right on the spot. The best part is 10% of it is for basic Web placement for the first 3 months alone.
How about the issue of whether pages with the .cfm extension will do as well as those with .html or .htm extension?
Macguru- I see what you are referring to. Yes, this issue has been sorted. There are no ? in the urls however the spiders still get stuck sometimes.
Brett: when you say that another domain works, do you mean that you have to build a nicely designed second web site? Can you show me an example?
Macguru: "Can you give us a rough idea of the size of the site? Also, I would like to know how "clean" is the coding. Is there ten gazillion of JS or CSS lines before the actual body appears?"
Yes, lots of coding. Site is in excess of 100 pages, but not all need to be optimised, as such. I've dealt with these sites before except my usual methods are much like your own - get some static content in there, make sure it is well linked, build a theme etc.
I just havent come across a client quite this "precious" before ;(
How much are you risking if you've never been in the database to begin with? I have done quite a bit of this kind of work, and I can tell you that if they are a big brand that has competitors that have similar dynamically challenged sites, there's a very good chance that some of them are cloaking to deal with these problems.
While many of the workarounds everyone's mentioned can work, The reality involved with trying to get an IT department to follow through is a nightmare. Even if you do succeed in handling URL formating problems, you still have to deal with all the other design problems that come along with large corporate sites.
What usually ends up happening is that you only end up getting a portion of what you need, yet the client will still expect you to produce top results.
I would set up some secondary sites either with a new domain, or a subdomain of their main site (much better for branding purposes)and use IP deliver at those locations to deliver html 2.0 versions of the pages you want included. If you're worried that you'll get in trouble, just make sure you use the exact copy from the original site (minus al the non content code)and make sure that your visitors land on the corresponding pages.
The IT department is happy because they get to keep all their cool code, and you're happy because you maintain control over all the eliments that contribute to good placement.
WebGuerrilla, I hear where you are coming from. I have used cloaking before and personally have no problem with it. Convincing this client is another matter entirely. IT are loathe to run Perl scripts on their servers, are suspicious of feeder domains they don't understand and have "heard cloaking is bad".
Once again, my problems are as much political as they are
technical. I won't even start on the internal politics this company has between marketing and IT.
(edited by: feeder at 9:51 pm (gmt) on Sep. 10, 2001