Steps to advance AdWords search campaign audit

If you are working in a niche market, search networks on AdWords and bing are bound to give you the most relevant traffic on your website. Here you have control over what search terms to target and optimize for long-tail keywords that have higher conversion rates.

This scenario might sound like an SEO problem but instead of trying to rank for some keywords you just throw in money to get the rankings. Trying to be on top of the search results for all relevant keywords all the time can be highly expensive. I have seen keywords for which we have paid 40$ as the CPC (Cost Per Click). Can you imagine 2800 INR for a click?

If you are bidding exorbitantly you will soon run out of money to even participate in the auction let alone getting the right CAC (cost of acquiring a customer).

The following report will give you a high-level analysis of the keywords you should focus on and think judiciously about your bidding strategy.

Create a report with all the search terms in the last year, along with their ad-group, campaigns, the keyword for which they triggered, along with impressions, clicks, conversions, cost, conversion rates and cost per conversions across these search terms (tools like Supermetrics should be handy in this scenario).

Going through the search terms you will immediately notice some irrelevant search terms because of your BMM and Broad match bidding. Next step is to add these search terms as negatives. Once this is done. You are one step closer to having highly competitive search campaigns.

Though your quality score and ads are important to get you a lower CPC and CAC for the purpose of this analysis, I have kept them out of the purview.

Now you categorise the search terms in categories namely A,B,C… or whatever you like on the basis of their cost per clicks and cost per conversions. A nested if statement will come in handy here.

So category A would mean a CPC of more than or equal to $12.5, B would mean a CPC of $12-$12.5, and so on. Similarly, category A1 would mean a CPA of more than or equal to $50 and B1 would mean a CPA of $48-$50 and so on.

In the above example, I have created categories A-Z on the basis of CPCs with an interval of 0.5$ and categories A1-Z1 on the basis of CPAs with an interval of 2$.

Now that you have categorisation done, it will be very easy to hone your efforts to the right category.

Now Plot CPA categories (A1-Z1) along with their CPC with the help of a pivot table and their sum of conversions across these categories. This will help you identify keywords that are having a low CPA (categories namely X1,Y1,Z1) and have a high sum of conversions. Pay particular attention to the CPCs.

Here you can see that categories X1 have a high sum of conversions, low CPC and low CPA and thus would be a good category to start with.

Note: you will find a lot of brand keywords in these categories as they tend to give a lot of conversions for renowned brands with low CPAs but the rest of the keywords can prove to be a goldmine.

You can then do a reverse analysis of plotting the CPC categories (A-Z) and have a look at their CPAs and sum of conversions to find the right set of keywords and search terms. Now you notice here category U seems to have a high sum of conversions and low CPA and CPC.

You may notice that the search terms in the best categories across CPA and CPC might overlap but in general, you would find some very interesting insights.

Now that you have finalised the set of keywords that have performed the best, you can look at the lost impressions report (keyword wise) and probably manually bid for these keywords to get better positions and a higher impression share.