B-to-B SEM: How Are You Really Performing?
When I talk to catalogers these days, I often ask about their online acquisition performance. Inevitably, they respond by telling me about the percentage of their total orders they’re now taking online and how efficient that is. (If you’ve read my previous posts, you’ll know that I believe taking orders online, while efficient, has some downsides.) After I listen politely for a moment or two, I probe a little deeper by acknowledging their online order performance and restating my question. “Yes, but how many new customers are you finding online? New customers that weren’t driven there by a catalog mailing or some other offline marketing activity?”
From there the conversation usually goes to what the company is doing in the pay-per-click or online advertising areas and the results they’re getting, to which I respond, “Yes, that’s great, but what about organic search? How many new customers are you getting from your search engine marketing (SEM) efforts?” About then in the conversation there’s a pause, followed by an answer of some sort that usually includes the phrase, “We’re not really sure.”
Of course, measuring new customer acquisition resulting from SEM is complicated and difficult to do. But in today’s environment, gaining new customers through SEM is vital. Recent postal and paper increases coupled with recessionary prospect response rates have most catalogers facing higher customer acquisition costs in traditional mail activities. We’re forced to get better at SEM.
Gauging Good SEM
What is good performance in this area, you ask? The leading-edge catalog marketers who really understand the online space are acquiring more than a third of their total new customers during the year from SEM efforts. In addition, they can measure where they’re coming from (which search engine, which comparison shopping site and so forth). They’ve calculated their paybacks by source and near-term value (say, one- or two-year value) of new sites (for B-to-B) and names acquired through SEM. Can you do this?