Never delete Keywords!

Status
Not open for further replies.

Jon12345

New member
Sep 14, 2006
187
0
0
Brighton, UK
Here goes...I think you should never delete any keywords. Why? Let me tell you about the frog that decided it would jump half way across a pond. So he did so. Then he decided to jump half way again. He kept on doing this to infinity and never reached the other side!

That is the problem with keywords. Your CPA offer pays out $35 and you have spent 50 cents per click on a keyword, got 70 clicks and spent $35. No sales. So what to do? I know, lets reduce the bid. If I delete the keyword, I may have lost out on an opportunity to make money long term on it, since more clicks could lead to a sale.

Consider this, even on a bad keyword it is likely to provide a sale if you left it running for years! So the question is not so much whether or not a keyword will be profitable, but rather at what bid will that keyword be profitable. It may take 1000 clicks for a less targeted keyword to produce a sale. But that is fine, so long as your bid is less than $35/1000=$0.035.

Therein lies my problem, I have been keeping all my keywords and not pausing any! Instead, I just reduce the bids.

Question to the experienced crowd: do you do this and if not, why not?

Thanks,

Jon
 


So that awesome long-tail that'll get 12 clicks a year you want to keep that one too? Why? You can't tell if it's ever going to be profitable no matter how cheap it is. Maybe you get lucky and maybe you don't, but one thing is for sure you should be focusing on getting your higher volume keywords to convert and leave that little shit on the floor and not worry about it.
 
Low bids = High average placement = Bad CTR = bad quality score and High Bids

See how this wont work?

Take your failures, learn from them, pause them, and then find something real.
 
So there are two issues here:

1. The time taken to monitor low volume stuff is an issue. Credit Rage9.

2. Lower CTR leads to poor Quality Score. Credit Drake.

Let's tackle #2 first. I read that Google has improved its algo to better take into consideration the CTR relative to its ad position. It would make sense for them to do this since it would mean that good ads go higher up and so they generate more revenue. So, does that mean #2 is not so important? Also, what if your keywords are still on the front page after reducing the bids? Do you still cut them loose?

Regarding #1, what is a good cutoff point? What about your lower volume keywords that could still turn a profit? Would you cut those too, since the volume could be low? If you don't cut them, the issue is more than just volume. So some combination of volume and ROI maybe?
 
The real question is...

Why would you waste your time to squeeze an extra 5% out of your existing campaign when you could be putting in the ground work for the meat and potatos of an entirely different niche?

You will make more money if you quit over-analyzing mundane details of your optimization. Including making threads, and coming back to summarize with bullet points.... all this for some throw away keywords.
 
80/20 man only worry about the keywords that will bring you in high volume don't waste your time on the small stuff. The time that your spending on kw's that have 4 clicks a day you could be making another campaign.
 
Any guidelines on:

a) when to cut keywords due to low volume. e.g. how many clicks per month as the threshold.

b) when to cut keywords due to poor roi e.g. after spending how much? or do you reduce the bid in increments and then see?
 
its true that all keywords have some value. it might be that you can get traffic for $0.001 and if you can get it to make $0.002 that's profit.

the question really is how much volume is there and where is your time best spent, as others have pointed out.
 
How much is a minute of your work worth?

If your going to make 100$ a year your budget for that longtail per day is 0,27$. Now from that you still have to subtract click costs.

Say you get 2 clicks a day at 1 cent each, leaving you with 0,25$ to allocate to working, you will be able to put in 1,2 minutes of work a day, assuming your earning 100$ a day, to break even.

just some math correct me if im wrong
 
You're going to accumulate a shit load of random ass keywords and you'll have to make sure that each campaign has an offer that's still running. How many offers run for years?

1) You're going to reduce your CTR.
2) You're going to have to make sure that the offers are still running for that one keyword.
3) You are going to have a lot of keywords to look at.

It's not worth my time, in my opinion. If it's not going to be profitable, pause the campaign or delete it.
 
Let's tackle #2 first. I read that Google has improved its algo to better take into consideration the CTR relative to its ad position. It would make sense for them to do this since it would mean that good ads go higher up and so they generate more revenue. So, does that mean #2 is not so important? Also, what if your keywords are still on the front page after reducing the bids? Do you still cut them loose?
That might be true. But reading something and seeing something is not the same thing. Google said they will treat underscore in urls for SEO the same as hyphens. Hasnt happened. They said they would index Flash for SEO. Hasnt happened. Both was over a year ago.

My experience says words that get impressions but dont get clicks get expensive. They weigh down a campaign. And they waste your time. Maybe a redesign of the site might make the CPC go down, but I pause them until that time.

Regarding #1, what is a good cutoff point? What about your lower volume keywords that could still turn a profit? Would you cut those too, since the volume could be low? If you don't cut them, the issue is more than just volume. So some combination of volume and ROI maybe?

Not trying to answer for Rage but I pause keywords that are less than Quality score 7 if I have made some effort to fix them. Either with ad copy, keyword selection, some high bidding, or landing page SEO. If I cant get it 7 or better I pause them.
 
Status
Not open for further replies.