Make sure your metrics give credit where it's due
When measuring the return on investment of search engine marketing campaigns, retailers may be giving branded terms far more credit for conversions than those keywords deserve. New research conducted by 360i and SearchIgnite assessed the relationship between branded and non-branded paid search keywords by reviewing aggregate data of traffic and conversion rates for retailers over the course of the first quarter of 2006.
The research looks at paid search campaigns from the retailer's perspective. Roughly 20 percent of searchers clicked on a retailer's ads more than once, and these consumers accounted for approximately 25 percent of transactions. Consumers who completed a transaction clicked on an average of 15 percent more of the retailer's ads than consumers who did not convert. These data illustrate that for searchers who click a retailer's ads more than once, the conversion rate rises significantly. As an extreme example, those who click a retailer's ads 10 times are three times more likely to convert than users who click an ad just once.
Here are four takeaways from the data:
Search is a process. Only about 75 percent of a typical marketer's sales from paid search are generated on the first ad click. The rest come from searchers who conduct more research, after which they're more likely to enter branded terms. So when a branded term is entered, the searcher is likely to be closer to purchase. Once visitors arrive at your site through a branded term, it's especially important that there are no obstacles to conversion.
Retail branded terms might not be as profitable as you think. Generally, only the last term a searcher enters gets credit for the conversion, and branded terms are especially prevalent toward the end of the search process. Non-branded terms earlier on in the process tend not to show up in the results, so they don't get any credit for leading to the conversion that the branded term locks in. Retailers that want to conduct internal reviews of the profitability of various terms should consider the following metrics: unique users clicking on branded and non-branded terms; the number of ads consumers click on before converting; whether the first and last ads clicked are branded or non-branded terms; and conversion rates. To fully optimize the campaign, retailers need to review findings at the keyword level and adjust spending accordingly.
Brand message and brand equity carry through the search process. The landing page for non-branded terms, even when going to specific product and category pages, should carry a strong branding message, not just the price and product. Searchers are far more likely to convert if their final click on a paid search ad comes from a search for a branded term, even if earlier searches are for non-branded terms.
Branded and non-branded terms must be managed together. The interactions between branded and non-branded terms have a significant impact on the overall success of a paid search campaign. While it's important to tag branded and non-branded terms separately for tracking purposes, the same internal or external resources (staff members, technologies, or third-party firms) should be responsible for both groups to ensure that the interactions between branded and non-branded terms are properly quantified and leveraged to the advantage of the entire campaign. Tracking and reporting systems must be sufficiently advanced to allow you to measure and quantify these effects.
Retailers that study how non-branded and branded terms interact with each other will develop a more authentic picture of how consumers search and buy. This can help marketers tailor the process to the consumer, which should improve return on investment, conversions, and campaign efficiency. If you're not doing this analysis, you are likely misallocating search dollars.
Now, how are you evaluating the true returns from your non-branded terms?