| 8:32 pm on Jul 7, 2004 (gmt 0)|
No you are completly wrong. You need to read some more on this website. First off the Overture tool means nothing. I have found **** to be more helpful. Even WT has it's limitations. I optimize thousands of keywords. If you have thousands of kw's that get one hit a day then that is better than one word that gets 1000 hits a day. The long kw's convert better. Way better. None of those tools can tell you exactly what to expect. I have terms that get me 20 visitors a day that don't even show up on the overture tool. Just because a term is typed in a bunch of times on se's does not mean they click on you. The big words are very general. Many specific 3 or 4 word phrases is often much better than some real general 2 word phrase.
| 8:58 pm on Jul 7, 2004 (gmt 0)|
I understand that more specific = better conversion. I also understand that to optimize for something general like 'movies' is a waste of time.
My concern is some of our campaigns only have 50-100 keywords, not the thousands you seem to work with. If half of them have very low clicks then how effective can it be overall? I'm just wondering if it could be made more effective with a balance of specifics and more generalized terms.
| 9:12 pm on Jul 7, 2004 (gmt 0)|
|My concern is some of our campaigns only have 50-100 keywords, not the thousands you seem to work with. |
It's not the size of your keyword list that matters, it's how you optimize what you have. ;)
|If half of them have very low clicks then how effective can it be overall? |
If you have a keyword that only gets 20 clicks, but 15 of those convert into paying customers, I'd call that very effective.
| 9:30 pm on Jul 7, 2004 (gmt 0)|
If you can't come up with thousands of keywords for any subject you are not very good at SEO. I would be supirised if you had a subject that really was limited to a few hundred keywords.
| 9:50 pm on Jul 7, 2004 (gmt 0)|
I could come up with thousands, as could a grade schooler, so I don't think that defines the quality of an SEO... Clients limit keywords because we up the charge as the number of keywords grows. Working within the limit of how many keywords they choose, I am trying to find the best balance between visibility and high conversion. Being ranked higher than the competition on common keywords is important to many of the clients, sometimes most important!
| 3:59 am on Jul 8, 2004 (gmt 0)|
That method of charging is just going to make you look bad. It is just as easy to do 1000 as it is 100. Their competiter may hire somebody for the same price and make a ton more money. If you are going to have thousands of kw's you don't write a page for each one. I spent 2 hours today putting up 2500 pages going after 2500 different phrases. I guess your not in the big league. I will have another 2500 by Monday. Each page is at least 15% different by the way. I do work for legit ecommerce companies. They are quite happy with the traffic and sales. The site looks very professional and only somebody that new what they were looking for would know that there are that many pages. (not cloaking)
You have to go after the small and big traffic because the big traffic is always built on sand. That way you don't look so bad when you lose one of them. If you put all your value on a small list of words a couple G changes and you get very angry calls because all of there traffic is gone. This way they only lose a small portion.
| 5:02 am on Jul 8, 2004 (gmt 0)|
|you are not very good at SEO |
|I guess your not in the big league. |
Illah -- just want to emphasize that ogletree isn't on the official WebmasterWorld welcoming committee. :) But he is trying to help, in his own way.
I'm not on that committee either, for that matter, but welcome to WebmasterWorld.
| 5:07 am on Jul 8, 2004 (gmt 0)|
Sorry it's late and I tend to be a little blunt.
| 5:36 am on Jul 8, 2004 (gmt 0)|
|Sorry it's late and I tend to be a little blunt. |
Maybe, but more to the point, experienced and informative. Care to give some insight into how you get those 2500 pages indexed? What does your link structure look like?
| 6:08 am on Jul 8, 2004 (gmt 0)|
The point is to have an old site with lots of good natural backlinks or one good backlink from a site like that. You can do whatever you want. Even if you don't have that if you have 2500 kw's you don't need much PR. I have found it important to have each page link to several other pages internally. When I make the pages just have the current page link to the 10 previous pages I just created. I have a front page with about 50 links that links to a bunch of other link pages that link to the entry pages. I don't expect people to come to my front page it is just a PR passer. If you have that many words they tend to be very similar. If you have a page about red widgets and blue widgets and so on have them all link to each other. At the bottom of each page have 10 or so links or more if you want (the more the better) to the other similar pages. I found that my sitmap page were really ranking well because they were just a simple html page with 80 links to pages that all shared a common kw. I then added a title that it was already ranking for and a little conten and they ranked even better. Make sure you never have more than 100 links per page.
There is more to it than that but I can't give out all my secrets.
| 6:47 pm on Jul 8, 2004 (gmt 0)|
Ogletree - I understand what you're saying. Now please, stop saying it :) All I'm asking is how to work best within my limits. I don't own my own SEO company, I work within another company so I have to work within the limits given to me. I'm trying to find the best solution I can given my limitations.
At my last company we sold computer games, and I did a major SEO with MANY keywords - some specific for each game, as well as more general ones. We have hundreds of top 10 rankings, and with our affiliate and link program if we're not #1, one of our affiliates is. With that said, I don't think I'm a 'bad SEO'.
Anyways, I just got off a call that confirmed my original question. Ranking #1 for a 10-search-a-day term just wasn't that impressive to the client. They don't sell hundreds of items where I can do a more traditional SEO, it's more of a branding and 'beat the competition' type of game. In fact some clients don't sell anything at all - it's all about being #1 and being 'found' before their competition.
| 8:23 pm on Jul 8, 2004 (gmt 0)|
Illah - you keep returning to "clicks" as a metric. Ogletree is speaking to "conversions" as a metric. It sounds like you understand the difference, but it is not clear that your client does.
Now, I count myself as a beginner in all of this stuff, but I do know that when I started going after multiple search terms I saw my traffic increase by several hundred percent. And, over 60% of the SE traffic comes from terms that are searched for less than 2 times per day.
Your original question was, "How many searches per month makes a good keyword?" Well, if 60% of my traffic comes from 1500 search terms that average 2 or less searches a day, then I would have to say that all of those terms are "good" for me.
| 9:15 pm on Jul 8, 2004 (gmt 0)|
Yes, conversions would be a better metric for SEO, but for some clients they're not selling anything on their site(s)! Because of that clicks and SE rankings are the only metrics that the client can look at. A specific search deep-linking into the belly of their site may not be worth as much as bringing in the masses to a landing page of some sort.
Anyways, I think over the course of this topic I kind of answered my own question. I just need to find those perfect keywords that aren't too specific and not too broad either.
| 7:34 am on Jul 9, 2004 (gmt 0)|
|Because of that clicks and SE rankings are the only metrics that the client can look at. |
Every good site has conversion events, should have them or has them and just hasn't yet realized it. Many fall into the third category. Even some FORTUNE 500 companies.
Lot's of companies don't sell anything online. They collect leads, they provide whitepapers for download, they have brochures that can be downloaded, they have store locators, they have newsletters people can register for.
Behind every website there needs to be a purpose, and that purpose for a commercial site is usually to ultimately increase sales. Nobody goes out blindly and purchases anything, there has to be some degree of learning about the item or service, even if its just looking at the package in the store or the description on a website.
At minimum, a company needs to look at site penetration from search campaigns. It is in most cases a complete waste of time, money and effort to drive people to a site if they hit the homepage or other landing page and bounce right out.
You'll help your company, Illah, and your clients if you can show them what conversion events they have on their sites. Think of how people buy whatever it is your clients sell, what steps do they take before buying? What information do they need to make a purchase decision? What actions can or do people take on the site that indicate they are getting closer to making a purchase? It can be a long process of education and some never get it but identifying the conversion events and tracking them will help you show the value you provide for your clients.
| 3:38 pm on Jul 13, 2004 (gmt 0)|
>>> I spent 2 hours today putting up 2500 pages going after 2500 different phrases.
How did you do that? Dynamically through database?
That will take me, not 2 hours but 2 years for pure html pages done manually at a neck-breaking speed. :(
| 6:50 am on Jul 15, 2004 (gmt 0)|
I can only speculate, but there are HTML generators that work off of templates. Give it a feed, and it pumps out pages. I've seen them used to genereate sites out of datafeeds from affiliate programs, but I suppose they could be used in this fashion as well.
| 2:11 pm on Jul 15, 2004 (gmt 0)|