I've been reading just about everything I can get my hands on regarding Adwords and noticed there's some disagreement as to how many keywords one should start with..

Some people seem to suggest you should start with a modest amount, retaining as much control as possible, and working your way up, adding more gradually and testing as you go.

Others say get as many keywords as possible in there and gradually reduce your numbers based on performance -- one book actually suggests starting with 2000 to end up with 25 "good" ones..

I'd like to hear any views on this and supporting arguments.

G Connelly

Not withstanding much experience with AdWords. It would seem logical to me to say "it depends".

First make sure that you are ready for AdWords. Do you have a system in place that allows you to track the traffic and how it is converted into sales. At the end of the day you want the keywords and the ads that work to make you money not just bring you traffic.

With that in place, I would look at the natural searches that come into your website and convert to business. List their frequency and pick the onces that obviously build a cluster. You usually see a few build the bulk of your referrals from search engines.

These are the keywords that I'd start with and see how effective they are. Once you have the mechanics down to track your data and know what is really profitable, I'd go and set aside a budget as percentage of sales to try new keywords and expand the universe. What works stays and what does not work goes.

Just my two thoughts.

K<o>

__________________
Technical Support for non-technical users of OpenOffice.org,
Conficio, Blogging about Plan-B for OpenOffice.org

Hi
Like the above post It differs campaign to campaign. I usually start small to test if it will be profitable. Tightly niched keywords with few ads. I set a budget, if it converts well then I expand with lots of keywords and ads. The trick is to test & track the converting ones. A pattern will emerge after a while. This way I find you don't blow your budget as well

I also track the way the visitors are searching int he cpanel and keep tweaking the campaign.

__________________
Gita

Profit Pulling Websites. Design,SEO,Hosting
www.pixels.net.nz

I agree with previous replies that it depends. I would start with your "acceptable" budget and the timeframe that you want to learn.

If budget is no issue, and you have little to no idea about the each campaign, I'd start large and monitor quickly and adjust as needed. Yet, I don't know many people who say budget is not an issue.

Thus, in my experience, we've learned as much about the campaign/keywords/current search trend for our niche and started out small. Once we see campaigns and key words working we expand from there.

Unless you've money to burn, never start with many keywords, planning to winnow out the losers; if you do, you'll quite likely find that you've exhausted your budget well before you have sufficient data to separate the winners from the losers.

Start with a set of well selected, highly targeted relevant terms. (See above posts for advice re. determining such.)

And, do not use "broad match" terms; stick with "exact match." With "broad match" you've no way to easily determine which sub-phrases are best suited to your needs, no way to bid different amounts on different sub-phrases, and run the risk of having your funds rapidly depleted by irrelevant and/or unproductive sub-phrases. (As an example, suppose that you are seeking to attract those looking to buy tickets to concerts by the group named "Train." Bidding for "train" as a broad match, or even "train ticket" as an exact match, will get you clobbered. Best to bid on terms such as "train concert" and "train concert ticket" as exact matches.)

__________________
The Penn State Ticket Man http://www.pennstateticketman.com
http://www.happyvalleytickets.com http://www.hounddogtours.com

Start with lots of keywords vs. only a few keywords I would have to choose lots of keywords.

How much you pay for traffic from those keywords is another issue all together.

No matter how many keywords you choose you need to drive that traffic to your website and make that traffic take a desired action. No matter how many keywords you target those keywords will only be as successful as your website.

It is important to break down your large keyword lists into very small targeted groups. Once you have your keywords sorted into tightly bunched groups you need to decide where you are going to drive that traffic. Make sure the page you're sending that traffic reflects the ad you have created.

Someone searching for blue widgets does not want to see your entire selection of widgets when they click your ad. They only want to see your blue widgets.


Always start with a large list of keywords. You can find lots of good keywords using tools like inventory.overture.com and www.wordtracker.com as well as www.goodkeywords.com. Once you have a big list of keywords (hopefully thousands) start breaking them down into very small groups. Somewhere around 10-20 keywords in each group. Once you've done this you'll learn a LOT about what people are searching, therefore what they're looking for. If you feel you currently have a page or category on your website that caters to what this traffic is searching for put up an ad for that keyword group and drive traffic to that page/category. Then measure your return on investment. From there just keep on tweaking.

Once you've done this start looking to expand into other areas that you may not have thought about until you've done this keyword research.

Good luck gpconnelly!

__________________
He who has a hundred miles to walk should reckon ninety as half the journey.

What? This approach is both arbitrary & speculative.

1) It is not the case that even seemingly related search terms will require a similar number of keywords; there is no such correlation. For example, those searching for information re. a Madonna concert tour do in fact use far more keywords in their searches than do those seeking like information for concerts by Medeski, Martin & Wood, who use exceedingly few.

2) One's budget is an over-arching concern, not one of secondary import.

3) There can be no "ideal" no. of keywords per "group;" any such chosen number is purely arbitrary.

Better to begin with as few keywords as possible, do research on such using a broad match search, then select the appropriate keyword phrases to be used as exact match keywords in your ad campaign.

__________________
The Penn State Ticket Man http://www.pennstateticketman.com
http://www.happyvalleytickets.com http://www.hounddogtours.com

I was under the impression that we were to provide our opinions on gpconnelly's question. I have done this. However I think that you have misunderstood my previous reply. Let me clarify.

I would agree with this. I would not have a group of keywords related to Madonna concerts, then additional groups for Medeski, Martin, and Wood. I would start a campaign for each of these artists. Then each campaign would have a groups of keywords that were related.

Ex.

Keyword: madonna concert

I'm not in the ticket business but if I was I wouldn't try to sell someone tickets to a Madonna concert if they searched this keyword. I would direct them to a page on my website that listed more information about Madonna concerts. Perhaps the locations, dates, and other points of interest someone may want to know about a Madonna concert. Then I would provide some type of action mechanism asking the visitor to do something. Keywords like this one would be put into a group related to searches looking for information on Madonna concerts.

Keyword: buy madonna concert ticket

If I was purchasing traffic on this keyword I would send the visitor to a more direct sales page. I would ask the visitor to buy a ticket to the upcoming Madonna concert. I would group this keyword with other keywords related to searchers wanting to buy Madonna concert tickets. People looking to Purchase Madonna concert tickets would be in another keyword group.


Agreed. The budget is very important. But I must point out that the idea of a marketing budget is flawed. The whole purpose of buying a keyword is to turn that keyword into profit. If you purchase $100 of traffic and produce $1,000 profit or you purchase $20,000 of traffic and produce $100,000 profit you probably won't shy away from purchasing more traffic.

When you're starting out the tighter you group your keywords the more control you ultimately have over how much money you spend. How you decide to spend your advertising money is up to you.


To gain the most control I would say that you should limit your keywords to 20 per group. This would include broad, phrase, and exact match keywords.

If you're on a tight budget I would agree that you should stay away from broad match. Once you have built up some kind of return on investment and are looking to gain more traffic do some research on negative qualifiers (if you haven't already) before purchasing broad match traffic.

__________________
He who has a hundred miles to walk should reckon ninety as half the journey.

Obviousely start with few and track their performance with conversation code. Then pick up a second set of keywords and find the profitables & so on.

__________________
Get And Activate $100s Of Adwords Voucher Without Using Credit Card , Bank A/c

I didn't vote cause in Adwords, it's always on a case to case basis...

__________________
Webmaster Outsourcing

Deepsand, I appreciate your comments and concerns here, but I think I would encourage people to start with as many keywords as possible, rather than as few as possible, as computergeneration and others have suggested.

It goes without saying, as you rightly point out, that budget limitations will exist for most of us and that this is an over-arching concern... But, you can put limits on your daily expenditures and, so, it seems to me that the best practice is to get as many words up there as possible, generate as much feedback as possible (concerning impressions, CTR, etc), with a view to filtering down to the best ones and advancing.

In a sense it is precisely because you have a limited budget that you should do this as soon as possible because the sooner you find out a few things about which keywords are good and which are bad for you, the better...then you can go on and hopefully make more money and increase your budget some day.

I think it makes sense to do this in as short a time as possible and at the beginning of a project.

For me keywords are a bit like seeds -- some grow, some don't. The best way to find out is to plant them all at once and see... in future you'll know what seeds work for you and which don't and you can put all your time, energy, and resources into nurturing the right ones.

__________________
The saga continues...

cd duplication / dvd replication Still the best marketing tools available!

I don't agree here. If your interested in conversion tracking then starting small is better and easier, especially if your next o all of this.

__________________
KBK Marketing | Jaan Kanellis Search Marketing Blog | Be the best and a link is what you'll get!