Dynamic spider killer to static spider friendly?

[I, Brian]

I've got some potential clients who have a database driven PHP/MySQL ecommerce site that is choking search engines. No unique titles on unique pages either.

However, I'm a link builder - webdev is not my speciality. So I can set up some good static keyworded landing pages for them (useful for human users as well as spiders), and then set up a substantial number of links to their site.

I've advised that getting the entire site pages is certainly preferable - but I'm not altogether sure if webdev specialists would feel they have to rip everything out to repair what's there.

So I'd like to ask for some advice on how likely it *might* be to correct the issue.

To illustrate: the page URLs are set up like so:

Product section:
shopping/mainpage.php?uniq_rand=af730df1019&hidcat=22

Product listings:
shopping/typesearch.php?uniq_rand=16605068&hidcat=227&txt_searchstring=227&txt_from=2

They look awful from a SEO perspective, but I'd like to invite some thoughts from those with webdev experience as to the viability of turning their sites over to being more spider friendly just on the information provided.

Obviously, those people have already paid for a site - I'm looking to see if the existing sites can be modified, because I don't think they'll be happy to be told to spent a few grand on a new online catalogue.

Ideally, I would want to look at a mod_rewrite solution for static pages, too.

Thoughts/comments/questions, welcome.

[DaveN]

Ideally, I would want to look at a mod_rewrite solution for static pages, too.

Bingo thats exactly what you need to do , then for the title grab them from the database or cat ( you most probably displaying them on screen already) and auto generate the titles " sitename - cat heading" and "sitename - Product Description"

once you have the mod rewrite working setup a good site map ..

DaveN

[I, Brian]

The question, though, is how difficult such a solution may be to implement.

[DaveN]

it depends on

1) what server if linux very very easy, if NT you need a rewrite mod plugin....

2) is it their own server ??

DaveN

[Mikkel deMib Svendsen]

The question, though, is how difficult such a solution may be to implement.

With all respect, I think the question is wrong :)

I would ask: How much is it worth to you to gain - long term, the kind of visitors full indexing and optimization of your database driven content could generate?

Usually such database driven sections of a site is where all the great content is: Dedicated product pages, category overviews (with generic terms) and descriptions of it all.

So, the first question is how much content is "hidden" behind this dynamic website that is not indexed?

Then I would look at what kind of content it is, and what kind of related content that are available from other (internal or external) resources.

If this is extensive - say 1000s of products and pages and plenty of content, then I guarantee you that it is worth the work of getting indexed and ranked. In fact, I am almost ready to dure this under my new program: Google Paid Inclusion - you pay me $25 per URL I get in :D

Not joking, for a good "dynamic case" I would do it! Does that tell you something about how easy it is ? :)

[seomike]

I've used category names and product names to query the db instead of product id #'s and cat id #'s.

Having said that though names would need to be unique.

[I, Brian]

With all respect, I think the question is wrong :)

In an ideal world, you'd be right - but when a client has paid a few thousand already for website development, they expect to be ready to go out and makes sales. They do not expect to have to suddenly fork out a few thousand more on redoing their site for indexing purposes.

Most smaller businesses simply cannot afford to budget for paying thousands for a site design twice over. And if they can already get targeted traffic and conversions from PPC and organic listings on landing pages, how would spending a few thousands more to completely renovate a site makes sense today, and not tomorrow? The immediate cost is difficult to justifty, unless the changes are small and significant, rather than ripping out the entire MySQL backend out and restarting from scratch.

Thats the reality I tend to deal with - and the small businesses I usually work with. Certainly I'll advise full indexing - but where the site already suffers, any corrective modifications absolutely have to justify themselves in the immediate short-term - or else be left to long term goals and planning.

[Mikkel deMib Svendsen]

I have yet to se a single case where it made more sense to build landing pages by hand - one by one, rather than optimizing the system. Often a combination can be used but I think you will limit yourself from too much of the actual search inventory if you entirely rely on such hand build lnding pages.


rather than ripping out the entire MySQL backend out and restarting from scratch.

It's usually never like that. As an example, I am just about to get approx 300,000 pages indexed with great content for a local client. We estimate that all the work required (from me AND the internal engineers) will be 1-2 days. The kind of content will probably give us 1/2 - 1 visitor per month per indexed page - thats 150,000-300,000 new visitors a month. You just can't beat that ROI with ANY other marketing :)

[I, Brian]

Certainly when you're talking about 300,000 content pages then there is great sense in prioritising the spider-friendliness of the index. :)

In the example I originally refered to, there were far fewer pages involved - a couple of hundred.

Landing pages that work as product lists, with keyword headings, would be built for the link campaign to create targeted linkage to - rather than simply to leave as standalone on-page SEO.

[DaveN]

Quote : but when a client has paid a few thousand already for website development, they expect to be ready to go out and makes sales.


In All Fairness the website can make sales it's just not SEo'd, if a client asked for a SEo'd website then the orginal WebDev team should pick up the Bill if the client made the mistake they should expect to pay to get it fixed....

We have monthly SEO Retainers just for consultancy.. thats means that the client can call me any time they want, if they do not call me one month i still charge them £3500, if they lose ranking I might have to drive or fly with my own WebDev team to help there Webdev team sort out problems for them and that could cost me a lot more than the £3500.. ( of course my clients would pick up the planes and hotel bills anyway... but that isn't in the contract, they just realise that sometimes they need my team there ! and are willing to pay whatever it costs... )

Never let a client get away for not paying for a service... I have a one guy that keeps increase how much he pays me i have never asked for more money he just wants to be my biggest client, when i asked why he said that a 150% increase in profits each month and was well worth the money...

it's all about ROI, your clients site is working i guess , how much better would it work if you had a the site fully SEo'd from the floor to the roof...

DaveN

[Mikkel deMib Svendsen]

In All Fairness the website can make sales it's just not SEo'd, if a client asked for a SEo'd website then the orginal WebDev team should pick up the Bill if the client made the mistake they should expect to pay to get it fixed....

It's funny, I made that exact point in an article back in 99: When you ask a team to develop a website it is your responsibility to tell them what you expect from it. If you do not specifically ask for it to be optimized how can you expect it to be? On the other hand, if you did ask and the web agency said: "oh yes, we handle that", then they should pay.

I once did an analysis for a large company in the medical industry. The agency that hired me had build this vertical sales and info portal for the medical client and as (a small) part of the project they included a guarantee to get them listed and well ranked for 20 keywords. But you know the kind of keywords that industry have, right? :) And this was even the "adult medication" type of stuff, so to rank well for those 20 top-level keywords would be a VERY tuff job indeed. I think they allocated $700 for it and outsourced the job to some low level doowar company that, off course, did not have a bloody clue about what they where doing. You can guess the result... right, nothing, nada, zero!

So, I was hired to evaluate the damage, and how much it would cost this agency to actually deliver on this promise: To get them ranked well for those 20 keywords. I told them the price - so they realized they had to tell the client it could not be done LOL

In this case it was indeed the agency that failed and they DID indeed pay by damaging a good relationship to this medical client by not delivering. Thats only fair.

[rustybrick]

I can attest to Mikkel's advice.

The beauty of technologies such as php, asp, cold fusion, etc. is that you do not have to hand code each and every page. As Mikkel said, within a matter of days (if the database exists), we can create hundreds of thousands and potentially more, unique, useful, and relevant pages. Landing pages is a joke, just define the characteristics of the content of the page and presto a landing page, in a manner of seconds (after the technology is developed).

[DaveN]

hand built landing pages for databases is so 90's ;)

[rustybrick]

LOL, yea and so is the oldsmobile.