If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting!

Here are some of the most popular posts:

Open Ads - Open Source Ad Inventory Management
The Big List of Affiliate Marketing Events
Secret To Get On The First Page Of Google

Once you have a campaign running for a while, it’s time to optimize it and trim your non-performing keywords. The important thing is to not be emotionally attached to your keywords. You’ve got to treat it like investing in the stock market and make your decisions based on data, not emotions.

I don’t care how much you like a particular keyword, if it ain’t profitable, you’ve got to dump it.

Which brings us to tracking. Everyone knows that the ability to track your advertising is one of the beauties of the internet. Every single conversion can be tracked down to the keyword level. How you perform your tracking is up to you. There are third-party solutions out there but you can also do your tracking on your own just by populating an Excel sheet with your links and giving each one a unique subid (that’s going to be another post for another day).

I have just starting looking into Prosper202 as another option for tracking. Just from my initial observations, I don’t like the fact that you have to manually enter the CPC for the keyword rather than using an API to access that info. Other than that, it looks like a promising system. Aside from Prosper, I’m also using a private tracker made available to members of PPC Coach.

It’s best to encrypt your subid’s rather than use the actual keywords from your ad campaign. Why? Some people are paranoid about the CPA networks seeing the actual keywords that bring in the conversions and go on to run their own PPC campaigns. I’m not saying that it happens but it’s possible. You’ve got a business and you have to protect it.

So… back to analyzing the performance of your keywords. When do you decide to dump a particular keyword? Do you wait for a certain number of clicks or ad spend before you decide? Some people like to wait for 100 clicks and see if they get any conversions. But if you’re bidding a in the $2-$5 range per click, this can become an expensive experiment.

Why wait for so many clicks before deciding? It boils down to the average conversion rate. If the average conversion rate is 1%, that means that you’re going to get 1 conversion in 100 clicks or 2 in 200. In the latter case, those 2 conversions could happen in click #151 and #199. So if your decision threshold is 100 clicks, you could think that your conversion rate is 0% when in fact it is 1% if you had waited long enough. The point I’m trying to make is that you have to wait for enough clicks to make a valid decision.

Others consider the ad spend vs. the CPA and use a multiple of the CPA as the threshold for eliminating the keyword. For example, you might choose 2x or 3x of the CPA as the point to dump a keyword. So if you’re getting $10 per conversion and your spend is $30, then you’ve reached the 3x CPA point and it’s time to delete that keyword.

Why is it important to trim your non-performing keywords? Anecdotal evidence for YSM shows that the QS there is primarily based on your CTR. So if your ad group CTR is high, then you’re going get a high-ranking ad while spending less per click. CTR is also important for AdWords but there it is just one of many factors deciding your QS.

High QS = lower CPC = higher profit margins

Whatever metric you use, you’ve got to have enough data to make an informed decision. For campaigns where the clicks are cheap(er) you can make your decisions after running the campaign for a few days. For more competitive markets where clicks are in the few $$ range, you’ll have to wait longer.

A campaign I have where my bids are in the $2-3 range, I review it after about a month. That campaign is profitable but in order to optimize it fully, I’ve got to get rid of keywords that are just sucking up the ad dollars but not converting. In some campaigns, you’ll see quite a bit of fluctuation in your daily ROI but you’ve got to look at the long-term trends rather than the short-term changes.

Technorati Tags:

Popularity: 5% [?]

Other Posts Of Interest

    Looking For A Few Good (Wo)Men

    Time constraints aren't letting me update the Marketing-Tools-Review.com site very often. I'm looking for a couple of people to help me review products and also serve as contributing bloggers here.You'll need to:1. be able to show your personality through your writing. 2. love to write.3. love internet marketing and/or direct

    Don’t Forget To Test Other PPC Networks

    There is one campaign I was running on AdWords that was hardly making any profit. But once I moved it to Yahoo, I started seeing an average ROI of 40%. The moral of the story is that you have to test all your traffic sources. Overall, the volume of impressions

    Looking For A Few Good (Wo)Men

    Time constraints aren't letting me update the Marketing-Tools-Review.com site very often. I'm looking for a couple of people to help me review products and also serve as contributing bloggers here.You'll need to:1. be able to show your personality through your writing. 2. love to write.3. love internet marketing and/or direct

    I Love Panama - Yahoo’s Update

    In my view, Yahoo is finally taking the right steps to improve their search marketing (PPC) product (formerly Overture). Now approvals for ad campaigns and keywords are almost instantaneous much like Google AdWords. I hated having to wait 3-5 business days to get my ads and keywords approved. This is

    Holding Contests To Get Some Link Love

    Currently, Ms. Danielle's blog site is holding a contest to give away a 30GB Microsoft Zune. But who cares about the Zune? I'd rather have the iPhone. These contests are a common way of trying to get some link love and new traffic to a blog. Actually, the prize is