Stuart McKelvey, president and CEO of TMP Directional Marketing, took some time today at ILM:07 to dissect some of the data that came out of the consumer survey, released in August and co-produced by comScore.

TMP executes the largest placement of Yellow Pages advertising in the U.S., and this survey was done to discern the value of online and offline media that constitute the growing local advertising mix.

This was done via online panel so the results should be weighed accordingly. In that respect it is telling that 33 percent of these respondents, who McKelvey characterized as “power users,” use print as a primary source for local search.

“We were frankly surprised by how print is still valued to this degree by 33 percent of these heavy online users,” he said.

What About the Rest?

The remainder of the media pie that constitutes local search is well known to be fragmented, so the key goal for marketers and CMRs, according to McKelvey, is to allocate budget and decide how to best piece together this media mix.

A great deal of this comes down to looking at the data, which suggest that there is a greater opportunity to modify ad spends based on what is learned from tracking behavior further down the purchase funnel.

“Move beyond clicks to see how media leads to conversions,” he said. “Utilize certain tools to better track post-search behavior, given that most of this activity is by phone and walk-in.”

This post-search behavior was shown by the data to be mostly identical across IYP, local (vertical) search and general search. In all these cases, about 10 percent of searches led to offline transactions. Offline purchasing meanwhile constitutes 90 percent of overall retail activity, according to comScore.

But How?

This has been discussed for some time in local search circles as a key challenge. But how do you track this post-search activity in practice (beyond survey data)? A number of examples were given including coupons, call tracking, landing pages that are specific to traffic sources and in-store pickup features.

“You can also make assumptions based on changing sales volume as it is tied to testing different campaigns,” says McKelvey. “In-store surveys can also be easy to integrate but sometimes aren’t precise.”

All these methods have been discussed a great deal in the local space but haven’t happened yet in large part, with the exception of call tracking. There is still growth to be had in coupons, and in-store inventory data/pickups, given that these methods bring you closer to the conversion.

A lot needs to be developed for the former, including business models in the mobile space around coupons. This development is being held back by a number of things including relatively low mobile local search adoption, but give these factors time to develop in the wake of some inflection points in the world of mobile product development.

The latter point (in-store pickup) is meanwhile gaining steam and shown to be very successful by Circuit City and Wal-Mart. Krillion has also jumped into this and is interestingly moving toward more of a transactional or CPA model, for which there is a big opportunity in certain categories (flat screen televisions in Krillon’s case). A few other local search plays such as Citysearch are meanwhile developing more ad models that are transactional.

As this develops, CMRs and search engine marketing firms will be able to better measure the value of the disparate parts of the local ad mix and customize campaigns in different categories and (hopefully) in individual campaigns. There is still a long way to go.