Benchmarking and Baselines in SEM

Analysts and marketers love to ask benchmarking questions: "What's the average conversion rate for PPC search?" "Does Google's conversion percentage vary from Overture's?" "What about second-tier engines?" "What's the conversion rate for my industry?"

My ClickZ colleague Bryan Eisenberg has covered conversion benchmarking. It came up again this week at Search Engine Strategies on my first panel. I'm sure it came up in other sessions as well. Marketers seem to think they should know their competition's conversion rates, especially within their industry segment.

The Value of Industry Benchmarks

Knowing the average conversion rate within an industry or from a traffic source does have value. But that value is uncertain, as it would be in the brick-and-mortar environment. Do RadioShack, Best Buy, and Crutchfield have the same average conversion rate for customers walking into their brick-and-mortar stores? Probably not, due to very different product mixes, price points, and kinds of customers at each store. Even within one retailer, the conversion rate for walk-in customers likely varies dramatically between stores.

Benchmarking averages is more applicable when you benchmark against yourself, rather than against others whose businesses are substantially different from yours.

Add different media and marketing strategies to the inherent differences between businesses, and external benchmarks hold even less appeal. With search engine marketing (SEM), each marketer actually controls her own conversion percentages. By managing campaigns around keywords, engines, and creative executions that deliver specific return on investment (ROI), you can easily skew conversion percentages by adjusting the keyword mix.

For example, last week I wrote about the Westin St. Francis hotel's brand keywords. Westin emphasizes brand keywords with great conversion rates but runs a balanced campaign that includes keywords with lower conversion metrics and good ROI. If Westin's agency wanted to increase conversion percentage, it would just need to pull some keywords with average ROI and lower conversions. Similarly, the campaign strategy could be adapted for more aggressive ROI, adding keywords that convert more poorly.

Some SEM marketing strategies specifically extend beyond conversion rates, and even online conversion, to reflect the true customer buying process and nature of the marketer's business. For example, marketer A is an Internet pure-play, and marketer B is a multichannel retailer in the same industry. Marketer B may have call-to-action messaging on its site regarding store pickup or toll-free ordering. Obviously, the online-only conversion percentages for search traffic will likely differ.

Product price also affects conversion percentages, particularly in the e-commerce area, where products tend to be identified by model number. Many consumers will react to discounts or pricing, particularly if the price is from a trusted vendor. Some customers are more leery of unknown vendors, and price alone may not influence them to buy.

Using Internal Benchmarking

To a great extent, each marketer controls his own conversion rate. Instead, use internal benchmarking, or baselining. Beating your own baseline numbers is as satisfying as beating the competition because much of the lift comes from the competition. When you run smart, efficient campaigns, you take orders away from the competition while meeting ROI objectives.

To set up an internal baseline, measure your critical statistics. Use whatever statistics are part of your business or marketing objectives. Marketers typically include conversion percentage, order volume, and an ROI metric (e.g., cost per order, cost per lead, or revenue per dollar spent). Some also include a profit-related metric because they want to know the tradeoffs between profit (or ROI) and volume. This tradeoff is an economic or accounting concept relating to the change in one variable factor that is based on a forced change in the other factor. Such baseline testing is known as "elasticity testing," which I'll cover in a future column.

Once you have a baseline, begin measuring your performance against yourself. As you make changes to positions, bids, keyword mix, landing pages, product price, and ad creative, compare results against your baseline. Any improvements will show up in the baseline reports. Experiment to understand how campaign changes affect your overall marketing and business goals.

Remember, in auction-based paid search, results are influenced by not only your actions but also the competition's. A sudden product price war or crazy competitor can turn winning keyword settings into a horrid mess. Similarly, seasonality affects conversion and the pay-per-click (PPC) marketplace.

Expect to see a certain amount of volatility in your baseline caused by external factors. Baselines help gauge success by extending beyond measurement and tracking to provide a window on the market. In auction-based PPC search, the latest competition may be total lunatics, but you're stuck with them. To get the most out of PPC search, beat your own conversion baselines. Your competition will suffer while you succeed.

Want more search information? ClickZ SEM Archives contain all our search columns, organized by topic.