Optimise Pay-Per-Click Advertising

Using Web Analytics

Pay-Per-Click (PPC) advertising is often an integral part of an organisation's multi-channel marketing strategy and can prove costly if done badly. Therefore it is essential to regularly review and optimise your PPC ad campaigns to ensure you are getting maximum return on your investment. In order to be successful, you should regularly research your keywords, have a clear and concise advert with consistent and mirrored messages on the landing page and a clear call to action. Web Analytics provides an intuitive way to optimise your PPC campaigns, allowing you to test and monitor the changes you are making to your PPC ads.

Business Goals

To maximise your desired outcomes by allocating a restricted PPC budget optimally among PPC vendors, PPC types (search based vs. content network), campaign segments, ad variations, and paid keywords.

Sample PPC advertising vendors:

• Google Adwords based on search or based on Google's AdSense content network
• Yahoo! Search Marketing ads based on search or content network

Background information and common terminology

Pay-Per-Click (PPC), (also called cost-per-click or CPC), has become the standard model for advertising with search engines. PPC vendors allow you to bid a maximum price you are willing to pay for a click on your "ad" (or listing) which is to appear when a visitor serarches for the paid keywords on which you bid. If you allow broad matching rather than exact matching, your ads may appear even if visitors search for derivatives of your paid keywords. In other words, your paid keyword can be different to the actual keyword that a visitor employed when they were served your paid listing.

Your ads can also appear on any search engine that contracts with the PPC vendors' content networks. The latter are affiliated web sites that provide ad space which the PPC vendors fill automatically based on their indexing of each page's content. The rank in which your ad appears, and the actual CPC, also depend on the maximum bids that competitors place for the same keywords.

Finally, within the PPC vendors' advertising consoles, you can group your paid keywords into ad groups (Google) or categories (Yahoo!). You can also supply one or multiple ad variations that will be served. The PPC vendors will automatically serve your ad variations that maximise your clickthroughs. However, they are unable to automatically optimise for visitor conversions, which is of course priority.

Sample Affinium NetInsight Trend Report

Most common challenges

• While the actual search keyword that a visitor entered into a search engine can be automatically identified from the visit's referring URL, there is no information in the referring URL on the PPC vendor, paid keyword, ad group or ad variation which triggered your ad.
• The exact cost of a visit is nearly impossible to determine because PPC vendors only report the average cost-per-click per day (where the day is typically measured by the US Pacific time zone).
  

Required Information Sources

• Destination URLs which contain additional URL parameters revealing the PPC vendor, paid-keyword, ad-group and/or ad variation that triggered your ad, e.g. http://mysite.com/?ppcvendor=Yahoo&paidkeyword=Blue+Shoes&advariation=3.
• Ideally, data obtained from each PPC vendor (via their report exports) approximating the average CPC for each unique destination URL for each day.
  

Solutions for measuring your success with PPC advertising

Affinium NetInsight's Keyword Summary shows your visits by the actual search keywords or phrases that visitors entered into search engines. If, however, you have stuffed your destination URLs for your PPC ad as described, you can employ Affinium NetInsight's Campaign summaries and URL Parameter summaries to measure traffic attributable to each of your PPC vendors, ad groups, or paid keywords. Using the configurable Key Performance Indicators, you can create conversion metrics for significant business events, such as registrations or purchases. Furthermore, if you enter your ad-variation texts into a Microsoft Access database or similar, you can employ the Affinium NetInsight data conduits to automatically replace URL parameters such as "ad=3" with your actual PPC ad texts, to view with their conversion rates.

Investigating clues for optimising your PPC advertising

If you have enabled "broad matching" with your PPC vendors, you should cross-tabulate your paid-keywords with the actual search keywords that visitors employ, triggering your ads. This will give you further critical insight, that is unavailable from PPC vendors' reports, namely whether part of your costs go to mismatched search queries, for example, "history of color copiers" in the report below for a copy machine vendor's paid keyword "color copier". With this insight the vendor can exclude certain keywords, such as "history", from broad matching to ensure PPC dollars go only toward prospective clients.

Furthermore, with the help of ad-hoc analysis you can investigate what each segment of visitors coming from different paid keywords is looking for on your website. For example, create an ad-hoc Path-From Summary from your landing page for a specific paid keyword, to see how the visitors who arrive there then navigate your website. Next, click the affinium push pin icon to make the filter criteria that you entered persistent, and select the Content Group and Exit Page Summaries for further information on the behaviour of this visitor segment.

Action items

• Experiment with reducing bids and rank while measuring how low you can go without affecting conversions.
• Increase spending on paid keywords depending on resulting conversions.