Google AdWords Workshop – Part 9:
Campaign Performance Analysis

The following article is part 9 in our series on running your own pay-per-click advertising campaign with the Google AdWords program. If you are new to AdWords, read the introduction here first.

In this post we’re going to look at how to analyse and act on your AdWords campaign performance reports.

I’ll use the example of a company making direct response sales. (The approach for a lead generation campaign will be different to the extent that there’s generally no immediate sales transactions that can be reconciled back against campaign costs. The work around in that case is to use the revenue value that you assigned to conversions in your budget).

When you take it together with the work involved in tracking and measurement, performance analysis can be a real drag on profitability if you go mad with it. So just have a think about what’s appropriate for the size and scale of your campaign.

It’s obviously important to know whether your campaigns are collectively making money or running at a loss. So as a rule, I always start by establishing the profitability of the account overall on a traditional accounting basis.

To measure this accurately, you’ll need to have a system in place that identifies and tags sales invoices raised as a result of campaign related conversions. Ask your accountant to set up AdWords as a profit center in your Chart of Accounts, and instruct your bookkeeping staff on it’s use. Profit and loss at the aggregate level should then become a routine accounting exercise.

Ascertain the total gross profit made on goods sold from clicks that converted, subtract the ad spend for the period, then subtract an allowance for overhead to run the campaign.

Obviously, you’ll want to compare that result against your budget and in particular against the components of the budget objective.

Larger organisations will probably also want to establish the profitability of individual campaigns and reconcile them back against the overall total. Depending on the campaign structure, that may mean you can track profitablity by product line as well as in the aggregate.

Next, you’ll want address basic questions like which of your ads are being searched, but not clicked and which ads are being clicked, but are not converting to sales.

Bear in mind that analysis right down to keyword level won’t be possible for content network campaigns and for some business models, it won’t be appropriate anyway.

For example, business-to-business vendors selling high value products with a long sales cycle often only have a few conversions during the average month, so analysis at the keyword level would be overkill. Analysis at the AdGroup and campaign levels is usually more meaningful in those cases.

Likewise, that small number of transactions means the reporting period appropriate to a B2B advertiser is likely to be longer than for the average business-to-consumer merchant.

B2B = Business-to-Business
B2C = Business-to-Consumer

Daily reporting could be necessary for a busy B2C merchant who completes hundreds and sometimes even thousands of transactions every month . . . but monthly reporting may be the right option for a B2B organization.

Those qualifications aside, you can generally learn a lot by examining the key performance indicators at all levels of the profit center.

Reports downloaded in spreadsheet format are invaluable in this respect because they allow you to sort and segment results very quickly. For example, it’s quite common to find the old 80/20 rule at work . . . 80% of your clicks will probably come from just 20% of your keywords. So sort your results by click-thru-rate to highlight the under-achievers.

A common approach with problem keywords is to quarantine them in their own AdGroup and then apply the optimization techniques that you can find summarized in checklist format on our web site. You can take the same approach with keywords that have low conversion rates, quarantine them for better focus on the factors that contribute toward conversion.

Campaign structures that group keywords in themes that correlate to the buying cycle can help in this respect. For example, you would expect early cycle keywords to have lower conversion rate, so grouping them together can highlight end-of-cycle terms that under perform.

What about return on investment . . . where does that fit in?

And how is it calculated anyway?

Answers to both those questions are coming up in the next post in this series.

Gary Elley

Leave a Reply

You must be logged in to post a comment.

The 7 Secrets of
Websites That Sell
Simply enter your name & email address to get INSTANT ACCESS!
We will not spam, loan, or rent your information!