When to Cut a Keyword Loose? : PPC Fool

Scissors Gagan over at Niche Affiliate Marketing, recently asked this question: “How long do you give keywords before you pause or delete it?” It’s a good question that every affiliate should have an answer for. If you never pause or delete non-performing keywords you could be losing lots of money.

Assuming your ad copy and lander are optimized, three things come to my mind when thinking about dropping keywords:

The Revenue Rule

A good rule of thumb that I follow is to wait until that keyword has accumulated about 2-3 times in cost what the payout would be. Example, if the payout is $30, then if a keyword racks up over $90 in clicks with no conversions. Time to cut it loose. Now if the keyword is in the top 3 positions or you really think the keyword has potential, you may want to lower the bid and wait it out. Dropping just a few positions can drastically cut the cost per click.

The Click Rule

I’ve read in other places waiting until you reach a certain number of clicks. Same concept as above. Waiting about 100-200 clicks will definitely get you some significant data. This is no exact science, but after 200 clicks most likely it’s not worth your time. This may also depend on what your target conversion is. If you’re assuming a 2% conversion, then 200 clicks is enough. But if you’re only counting on .5% conversions, then maybe not.

The Worthless Keyword Rule

Apart from above, you should be trimming your keyword lists every so often, especially in the beginning. 1-2 weeks after launching a campaign, go back over those keywords with 0 impressions and get rid of them. The same with getting lots of impressions with 0 clicks. Now that could be because your ad position is in 8,9,or 10 or maybe your ad copy is off. Raise your bids a little and double check your ad copy, but if still after 1 more week and no clicks my suggestion is to drop it.

I’m sure there are other ways of doing this, would love to hear your thoughts. To be honest, I always kind of do this on the fly and just eyeball it. It’s always good though to have some general rules

5 Comments Already

adobo republic Said,

September 10th, 2008 @5:32 am

good information. but how about click frauds? how many percent of those 200 clicks can be considered as click fraud? do you include it in the count?

Gagan Said,

September 11th, 2008 @9:34 am

Thanks for answering the question Dave.

Not to confuse people - recently I launched a new campaign. I started off with targeted & competitive keywords and had a landing page done from a professional. The offer was also targeted to the keywords and the landing page. However, once the campaign was running, the conversions didn’t stack up as I had hoped. I ran it for a week. I was thinking of cutting lot of keyword loose when later I switched the offer to another similar offer which did way better in conversions than my original offer I ran.

So, sometimes it could be the offer as well. When launching a campaign, also split test offers (2 or 3 at a time). Send huge amount of clicks in the beginning to test everything.

Gagan

b2b guy Said,

September 12th, 2008 @7:36 am

As you optimize the landing page, be sure to revisit those paused keywords.

Smaxor Said,

September 14th, 2008 @10:32 am

This is a question I see asked a lot over the years at Wickedfire. I think what needs even more attention then the keywords is the offer. Owning a network I see offers that just don’t convert. Plain and simple I don’t care if you have the best keywords and the best landers. As a network I try and delete these offers most of the time but it takes a while to flush them out sometimes. Honestly my style is start with words I know that should convert before I try and find a lot of long to medium tails. For example bid on “payday loans” exact match and see if the offer converts at all. Or the keyword “diet” or “weight loss”" to a green tea or acai offer.

Being on the inside and owning a network I’ve realized almost nothing qualifies someone to put an offer on a network other then they have one and have some money. A lot of people come to me and have never tested their offer themselves. Couldn’t tell you how it converts or any conversions numbers. They use us as affiliates as their testing board. We blow our money to an offer that just doesn’t work. So make sure before you start pruning words you’re sure the offer is worth working with at all.

Smaxor
Ads4dough.com

Dave Said,

September 14th, 2008 @10:55 pm

Nice, I’m sure I could learn a lot from someone who has been on both sides of the fence. It’s amazing someone would throw an offer at a network without ever testing it first.

That’s a lose-lose situation for both people. Who would want to promote something that doesn’t convert? I remember awhile back I was sending hundreds of clicks to this one offer I couldnt figure out why it wasnt converting.

Now, it is easy to see, their lander stunk and the offer was blah. That offer will never have any success. A word to you merchants out there