redtide1969
BANNED
- Feb 15, 2009
- 1,899
- 1,575
Hey all,
I had typed out a long email to a friend tonight explaining how I turn my PPC campaigns consistently profitable and figured I'd paste it here for any of the newer members to soak in if they so wish...
This email doesn't cover niche and offer selection so if you can't do that I suggest quietly stepping away from PPC in general.
Hey man, no problem.
Some of this you may be aware of but because I don't know where you're at with PPC knowledge I'll try to hit on things that I think will help the most.
First thing is that PPC networks are all setup with default campaign and adgroup settings that are made to feed their pockets and not yours. I'll try to dig up a session I did on messenger that explains all of the default settings I change for every campaign. In the meantime read this post I had made that explains how I manage my campaigns http://www.blackhatworld.com/blackhat-seo/general-ppc-discussion/226618-making-killing-ppc.html
To dig deeper into what I was explaining there, I run two PPC accounts at all times with any of my campaigns that are in their initial 3 phases of optimization, and the other account is only used for my already optimized campaigns that have high CTR's, quality scores, etc.
This secondary account will eventually allow you to be paying pennies on the dollar for clicks on your keywords that may have initially been costing you $3-$5+ a click in the original account.
Obviously you're stowing away the golden KW's to only be run in the clean secondary account and continuing to optimize down the original campaign that will always inevitably have a weak CTR (click through rate), low QS (quality score), and lack of relevancy from having so many KW's, in so many adgroups, all carrying high impressions. This means you are paying up the ass for the long term if you keep your campaign in that accout.
A huge key part is what I call the triangle of relevancy and it's really important to optimizing down a campaign that was initially losing money, then breaking even, and then within a few weeks of losing money starts turning solid, long term profit.
The triangle is the relevancy between your KW's - ad title/adcopy - and LP copy/keywords and is a main factor in how much you'll be paying per click. You probably know how important everyone sais this relevancy triangle is with PPC, and it's totally true. My main priority is getting my winning or at least "promising" keywords into that secondary account as soon as possible where I can keep my adgroups tight and concise with just 2-3 closely related KW's per adgroup - with two already optimized ads per group that are carrying the exact keyword as the title.
Let me do a timeline:
Initial testing phase...
Start campaign with at least 100-300 KW's or more depending on your budget going in.
Start it with as many adgroups as needed but always trying to keep each adgroup with no more than 50 KW's per adgroup for QS and relevancy.
Sort all KW's into groups that are as relevant to each other as possible.
I start all campaigns running exact, phrase and broad match (one thing I've learned is you never know which match type will turn into a golden KW.)
Two ads per adgroup to split test CTR's and CPC's.
2-4 weeks later...
2nd phase
Decide which KW's are worth raising your bids on and keeping (based on which are getting impressions/clicks and have converted more than once or twice). Delete or pause non-performers but remember, lowering or raising your bid for different ad positions can and will turn a non converting KW into a golden KW so it's somthing to play with before getting rid of a good, targeted KW or its match types.
Add negative KW's in the PPC account based on broad and phrase match KW's that trigger your ads for non-converting terms. For me, I try not to send more than 10 clicks at a KW without a conversion before I play with my ad position. If a KW hasn't converted within another 10 clicks after playing with bid price, it's gone.
Continue split testing ads obviously to keep your CTR rising.
If I have had to bid really high to get the top spots I DON'T lower my bids in most cases. To dominate your highest converting ad positions it's best to let the PPC network's QS lower you actual CPC's rather than trying to keep lowering your bid in an attempt to get cheaper clicks.. The actual clicks costs will come way down to pennies on the dollar but it just takes some time. Once done, a competitor would have to bid outragously high to ever beat you for that ad position (for example position #1 while bidding $6 but actually paying $0.78 per click based on relevance) so having the balls to initially bid that high basically "locks in" your converting position so it's too costly for competitors to try and outbid you.
Last step of phase 2 (there's more but I'm trying to run through this) is I pause or delete the original campaign and recreate it with only my remaining "promising" KW's and optimized ads (at this point I'm still in the first PPC account).
Keep your new version of the campaign and adgroups with no more than 15 KW's per adgroup at this point. The bottom line is, the smaller the adgroup, the higher the relevancy and QS and the lower you pay per click (when initially starting with 1000's of KW's like I tend to do, it's not realistic to be able to create and manage that many adgroups).
2-6 weeks later...
Phase 3 = repeat phase 2 and keep the campaign in the initial PPC account.
2-6 weeks later...
Move the remaining "golden nugget KW's" to the secondary PPC account and either create a seperate adgroup for each keyword (gives you the most relevance possible and is the way I do it,) or put no more than 3-5 KW's per adgroup if the KW"s are really close variations of each other.
1 month later...
Break the campaign down one last time by pausing or deleting the old version of the campaign and recreating it this last time with seperate adgroups for all of your various KW "match types". So create a seperate adgroup for the KW's exact match variaton, phrase match and broad match variations.
Last phase...
Make serious bank with low CPC's, high QS's and high ROI
You basically can just swap offers to split test conversion rates and the campaign can be set and forget, depending on how hard you want to work on building it out. You can tell that I kind of look at PPC like a miner panning for gold, it is like hunting gold in that it's a constant weeding down process of getting rid of the black sand (high click/high cost/low conversion KW's) until you're left with a handful of nuggets (high click volume/low cost per click/high conversion rate KW's).
You get there by slowly breaking down the campaign's KW/adgroup structure, making your adgroups more tightly focused each time, so that as time goes on you're suddenly paying $130 a day in adspend per campaign for $1200, $1600, $900 lead days consistently. Pick the right payout offers and you're set.
I also always start with a high budget that I'm prepared to burn through while collecting my stats (impressions vs clicks vs conversions) in P202 on my initial 100's or 1000's of KW's I start testing with. Once you're down to just a handful of low cost, high converting KW's...you're gold.
K man, hope some of this helps
I had typed out a long email to a friend tonight explaining how I turn my PPC campaigns consistently profitable and figured I'd paste it here for any of the newer members to soak in if they so wish...
This email doesn't cover niche and offer selection so if you can't do that I suggest quietly stepping away from PPC in general.
Hey man, no problem.
Some of this you may be aware of but because I don't know where you're at with PPC knowledge I'll try to hit on things that I think will help the most.
First thing is that PPC networks are all setup with default campaign and adgroup settings that are made to feed their pockets and not yours. I'll try to dig up a session I did on messenger that explains all of the default settings I change for every campaign. In the meantime read this post I had made that explains how I manage my campaigns http://www.blackhatworld.com/blackhat-seo/general-ppc-discussion/226618-making-killing-ppc.html
To dig deeper into what I was explaining there, I run two PPC accounts at all times with any of my campaigns that are in their initial 3 phases of optimization, and the other account is only used for my already optimized campaigns that have high CTR's, quality scores, etc.
This secondary account will eventually allow you to be paying pennies on the dollar for clicks on your keywords that may have initially been costing you $3-$5+ a click in the original account.
Obviously you're stowing away the golden KW's to only be run in the clean secondary account and continuing to optimize down the original campaign that will always inevitably have a weak CTR (click through rate), low QS (quality score), and lack of relevancy from having so many KW's, in so many adgroups, all carrying high impressions. This means you are paying up the ass for the long term if you keep your campaign in that accout.
A huge key part is what I call the triangle of relevancy and it's really important to optimizing down a campaign that was initially losing money, then breaking even, and then within a few weeks of losing money starts turning solid, long term profit.
The triangle is the relevancy between your KW's - ad title/adcopy - and LP copy/keywords and is a main factor in how much you'll be paying per click. You probably know how important everyone sais this relevancy triangle is with PPC, and it's totally true. My main priority is getting my winning or at least "promising" keywords into that secondary account as soon as possible where I can keep my adgroups tight and concise with just 2-3 closely related KW's per adgroup - with two already optimized ads per group that are carrying the exact keyword as the title.
Let me do a timeline:
Initial testing phase...
Start campaign with at least 100-300 KW's or more depending on your budget going in.
Start it with as many adgroups as needed but always trying to keep each adgroup with no more than 50 KW's per adgroup for QS and relevancy.
Sort all KW's into groups that are as relevant to each other as possible.
I start all campaigns running exact, phrase and broad match (one thing I've learned is you never know which match type will turn into a golden KW.)
Two ads per adgroup to split test CTR's and CPC's.
2-4 weeks later...
2nd phase
Decide which KW's are worth raising your bids on and keeping (based on which are getting impressions/clicks and have converted more than once or twice). Delete or pause non-performers but remember, lowering or raising your bid for different ad positions can and will turn a non converting KW into a golden KW so it's somthing to play with before getting rid of a good, targeted KW or its match types.
Add negative KW's in the PPC account based on broad and phrase match KW's that trigger your ads for non-converting terms. For me, I try not to send more than 10 clicks at a KW without a conversion before I play with my ad position. If a KW hasn't converted within another 10 clicks after playing with bid price, it's gone.
Continue split testing ads obviously to keep your CTR rising.
If I have had to bid really high to get the top spots I DON'T lower my bids in most cases. To dominate your highest converting ad positions it's best to let the PPC network's QS lower you actual CPC's rather than trying to keep lowering your bid in an attempt to get cheaper clicks.. The actual clicks costs will come way down to pennies on the dollar but it just takes some time. Once done, a competitor would have to bid outragously high to ever beat you for that ad position (for example position #1 while bidding $6 but actually paying $0.78 per click based on relevance) so having the balls to initially bid that high basically "locks in" your converting position so it's too costly for competitors to try and outbid you.
Last step of phase 2 (there's more but I'm trying to run through this) is I pause or delete the original campaign and recreate it with only my remaining "promising" KW's and optimized ads (at this point I'm still in the first PPC account).
Keep your new version of the campaign and adgroups with no more than 15 KW's per adgroup at this point. The bottom line is, the smaller the adgroup, the higher the relevancy and QS and the lower you pay per click (when initially starting with 1000's of KW's like I tend to do, it's not realistic to be able to create and manage that many adgroups).
2-6 weeks later...
Phase 3 = repeat phase 2 and keep the campaign in the initial PPC account.
2-6 weeks later...
Move the remaining "golden nugget KW's" to the secondary PPC account and either create a seperate adgroup for each keyword (gives you the most relevance possible and is the way I do it,) or put no more than 3-5 KW's per adgroup if the KW"s are really close variations of each other.
1 month later...
Break the campaign down one last time by pausing or deleting the old version of the campaign and recreating it this last time with seperate adgroups for all of your various KW "match types". So create a seperate adgroup for the KW's exact match variaton, phrase match and broad match variations.
Last phase...
Make serious bank with low CPC's, high QS's and high ROI
You basically can just swap offers to split test conversion rates and the campaign can be set and forget, depending on how hard you want to work on building it out. You can tell that I kind of look at PPC like a miner panning for gold, it is like hunting gold in that it's a constant weeding down process of getting rid of the black sand (high click/high cost/low conversion KW's) until you're left with a handful of nuggets (high click volume/low cost per click/high conversion rate KW's).
You get there by slowly breaking down the campaign's KW/adgroup structure, making your adgroups more tightly focused each time, so that as time goes on you're suddenly paying $130 a day in adspend per campaign for $1200, $1600, $900 lead days consistently. Pick the right payout offers and you're set.
I also always start with a high budget that I'm prepared to burn through while collecting my stats (impressions vs clicks vs conversions) in P202 on my initial 100's or 1000's of KW's I start testing with. Once you're down to just a handful of low cost, high converting KW's...you're gold.
K man, hope some of this helps
Last edited:
