How should web analytics tools be used to measure paid search?

Web analytics is essential for measuring and optimising paid search campaigns, and being able to extract the rind kind of data to improve your campaigns.

This post, which includes some extracts from our new Paid Search Marketing (PPC) Best Practice Guide, looks at examples of how to use analytics to track PPC performance.

By Graham Charlton March 12th 2013 19:01

Creating a PPC dashboard

A dashboard is the first port of call when you login, and allows for quick identification of key metrics and issues which may require further investigation.

To do this, a PPC dashboard should give users a snapshot view of the metrics that matter to them. This is an example Google Analytics dashboard, which you can add to your GA account here, but it’s likely you’ll need to customise it to suit your own needs.

Key metrics to look at include ad groups, keywords and conversion rates.

It’s also worth checking our Brian Clifton’s ‘Better Adwords’ custom report for a useful overview of PPC metrics.

Comparing paid search with other marketing channels (especially SEO)

Comparing channel performance is a great way to benchmark, uncover insight and discover themes. Users should exercise common sense when comparing channels because they will sit at different stages in the conversion funnel and may have differing goals.

For example, display traffic may convert poorly compared to paid search, due to differing user intent.

Within Google Analytics and Adobe Analytics, there are a variety of reports available to compare channel performance. A few of the most insightful have been outlined below.

Navigation within Google Analytics = Traffic Sources > Search > Overview
Navigation with Adobe Analytics = Traffic Sources > Search Engines – Paid / Search Engines – Natural

Use this for a top line, customisable view of paid against organic search. Visits, pages per visit, average visit duration, percentage of new visits and bounce rate are examples of metrics that may be useful in this report.

For example, you discover that paid search bounce rates are far greater than SEO, which may help you to determine that paid search campaigns lack deep linking to tailored landing page content, compared to SEO.

Evaluating landing pages

With PPC, sending visitors to your site is only half of the battle, and will be largely fruitless unless you have done a proper job of optimising the on-site experience.

Evaluating your landing pages is key here. According to Adobe’s Jonathan Beeston:

For example, judging landing pages on bounce rate alone might neglect the larger story. A landing page with a high bounce rate may actually be resulting in the greatest revenue per visitor, which may be more closely aligned with overall goals. Some specific landing page evaluation methods have been outlined below.

Navigation within Google Analytics = Content > Site Content > Landing Pages
Navigation with Adobe Analytics = Campaigns > Entry Page change Site content > Pages

This report shows you the overall performance of landing pages. In the context of paid search, there may be an on-going drive to increase average order value. Therefore, you may be interested in which landing pages lead to the highest average order value.

This report could show that a particular landing page has a very low average order value and high bounce rate, meaning further analysis should be undertaken or other landing pages should be tested.

Assessing advertising performance at ad level and keyword level

To fully optimise paid search campaigns, users often need to get very deep and granular with their analysis. Within analytics tools there are a host of reports that facilitate such analysis. Within Google Analytics many of these reports sit with the ‘Advertising’ tab.

Adobe Analytics users who are using Adobe AdLens or SearchCenter to manage their paid search campaigns can also sync data (in both directions) between them.

The most simple type of analysis may be determining the return on spend of keywords. In Google Analytics, go to Advertising > Keywords > and add relevant filters.

This report can inform decisions to optimise campaigns. Looking more deeply can help you to find the keywords that convert certain types of product, which can help you to tweak paid search ads or landing page copy.

Paid search performance can be significantly improved by discovering and acting upon keyword insight.

Evaluating conversion paths

According to Google:

Customers research, compare, and make purchase decisions at different times and in different places, so measuring return solely based on the last click gives an incomplete picture and potentially misses important insights about how you reach your most valuable customers.

With this in mind, evaluating conversion paths to determine the optimal mix of marketing channels is of high importance. Analytics tools, such as Google Analytics (via the ‘Multi-Channel Funnel’ tab), can help in this quest.

You will discover insights around how and when channels impact conversions, time lag to conversion and pre-conversion path lengths. Analysing this data will allow you to correctly value each channel and appropriately apportion budget to maximise future overall return.

Evaluating impact on soft goals like newsletter sign-up and content sharing

While core conversion metrics are important, analysing softer metrics (such as newsletter sign up) can also provide valuable insight.

Not every visitor that comes to your site via paid search will convert, but this doesn’t mean their visit was a waste. It may be that looking at ‘softer goals’ will show this.