Pruning your PPC campaigns

[Jon12345]

I would like to solicit your views on the process you go through when pruning a PPC campaign. Let us assume we have 1,000 keywords, with 1 keyword per adgroup.

When dealing with a large number of keywords, it can get time consuming to tweak bids and test ads. So what process do you go through to prune your keywords? When do you decide that you will delete the keyword from the adgroup, or pause it, or go to work on it.

This is an area that I have the most conceptual problems with, since there are many conflicts that I can see. But before I comment on those, I would like to see what others do.

Thanks,

Jon

[Dave Ward]

Originally Posted by Jon12345

I would like to solicit your views on the process you go through when pruning a PPC campaign. Let us assume we have 1,000 keywords, with 1 keyword per adgroup.

When dealing with a large number of keywords, it can get time consuming to tweak bids and test ads. So what process do you go through to prune your keywords? When do you decide that you will delete the keyword from the adgroup, or pause it, or go to work on it.

This is an area that I have the most conceptual problems with, since there are many conflicts that I can see. But before I comment on those, I would like to see what others do.

Thanks,

Jon

Jon, it depends on so many variables, every business is different, but in general what it comes down to is whether the keyword is bringing you the desired action ? It could be a optin to your list, or a sale. If the keyword does not bring results, you you either pause it, or lower the bid/ increase the bid, but as I say it is all dependent on what you expect. The answer is different for a business who makes £10 per sale, to one that makes £1000 per sale. or a lead is worth £1 or £25.

I am not sure why you would have a 1000 keywords in 1000 ad groups though.

[Jon12345]

For me, it is easier to track having each keyword in a separate adgroup. My questions are relating to promoting CPA affiliate products, normally from $15 to $40 per lead.

I see some dilemmas with pruning.

E.g.

1. 80:20 rule - only keep the top 20% of keywords, but...keep all keywords so you benefit from volume on the long tail

2. Delete adgroups with a low CTR since this drags down your quality score, but...keep adgroups if they are profitable since they that is the goal.

So, I want to know what to prune, when to prune it!

[Dave Ward]

Jon,
Here is one general rule. If your ad for a keyword has had 200 impressions and has less than a 1% CTR. Kill the ad and write a new one. If you still can't get above a 1% CTR after a few attempts, then kill the keyword.

If a keyword has had 200 clicks and is not converting, kill the keyword. Now again if your clicks are £5 you many not want to wait for 200 clicks, again I don't know your case. If your promoting a $15 lead, you know your keyword bid price and your tracking at keyword level, its easy enough to decide when you want to cut that keyword.

If any keyword in your list is not bringing you the ROI you want,prune away !

[Jon12345]

But what about the two dilemmas I mentioned? You might get 200 impressions and a 2% CTR but it might be one of the 80% rather than 20%. So straight away, there is a dilemma. Keep oir delete?

Also, you may get a 0.9% CTR but it is a profitable adgroiup. So, keep or delete?

[Dave Ward]

Originally Posted by Jon12345

But what about the two dilemmas I mentioned? You might get 200 impressions and a 2% CTR but it might be one of the 80% rather than 20%. So straight away, there is a dilemma. Keep oir delete?

Also, you may get a 0.9% CTR but it is a profitable adgroiup. So, keep or delete?

You are making it more complicated than it needs to be. At the end of the day it comes down to if the keyword is profitable for you or not. If on 0.9 %CTR its profitable, then keep the word, and try and get the CTR up by A/B testing the ads on the keyword.

It is about constant testing and improving, but the benchmark is always whether the keyword gives you the results you are happy with.