Text linking - Controlling / Selling them yourselves

Replies to This Discussion

Permalink Reply by Douglas Hebbard on January 5, 2009 at 9:55am

Having a "library" inside the existing CMS seems like an attractive solution, and one that would be provided by the existing CMS partner. The library would need to have some of the same properties as the ad banner manager: that is, start and expiration dates, weighting, and possibly audience targeting, as well.

I would assume that currently most companies are using an outside firm for text mining -- though I have had no experience with this to date.

While the vast majority of our advertisers are nowhere near ready to pursue this type of promotion, the market leaders are already there. But of course, it all depends on the markets you serve, doesn't it?

▶ Reply to This

Permalink Reply by OlegR on January 21, 2009 at 10:03am

Wanted to add to Michael's question (I'm coming from a company that both provides CMS and text-mining/linking technologies) since I think they are related.
How important would it be for publisher's bottom line to automate this kind of process (either working with a 3d party or with CMS-built-in dictionary)?

How much saving does the publisher see in a CMS system that analyzes textual content, automatically suggests keywords for SEO, links relevant articles (related article linking on the front end of your site) and removes all of the tagging activity from the editorial side of the business?

What about your top-line if this got automated and consistent?

▶ Reply to This

Permalink Reply by Michael Zingalis on January 21, 2009 at 10:26am

I'm with you, but my question is...what part do the robots need to do? (specifically around the taxonomy / advertising suggestion)

Scenario 1: A company wants to buy keywords on your site. They want to buy the same keywords on your site as they have seen successful in using to bring people to their site. They already know this information / list.

Doesn't this relegate the responsibility of the system down to simply linking the proper text from the library provided to the ad mechanism per page load? Where are suggested keywords required at that point?

Scenario 2: A company wants to advertise on your site, and has no idea what keywords currently get to their site, but they had webstats currently.

Can't the primary list of keywords used and the numbers simply be recieved from something like Google Analytics to provide the library (since it doesn't need to change on a daily basis during their period of the advertising)? If a company can know through a service like google analytics or whatever web software that they use...'what' keywords they should buy, then again, what is the purpose of the content relating robot? You already have sold the client the best possible keywords, which people currently use to get to that client's site.

Scenario 3: A client wants an ad, doesn't know their keywords, doesn't have previous analytics on the site to know

This one I think I get, because the company is not even in existance yet, and needs something.

I'm just trying to figure out 'why' it matters that there is

I think that what would help best is a reminder to content posters in their console (maybe a refreshing window) of what words have been sold, so they know to write the adwords around the content, and not the other way around..lol.

Maybe even a forced thesauraus that takes one word that is close enough to the other paid word to change within all records on the site to get intial numbers before a sale. Kind of like buying time on google to make a client advertising requirement...but within your site...so you can control the variable without constantly shelling cash to show your site.

▶ Reply to This