According to recent research from Forrester, conducted over two years with Shop.org, social networks like Facebook have not generated any significant revenue or return on investment for online retailers. Forrester retail analyst Sucharita Mulpuru's conclusion is that people use social networks for -- well -- socializing, not shopping.
Social networks would seem to be an ideal medium for product discovery and social sharing of product preferences and reviews, but so far the evidence for this is non-existent. It may be true that the activities we typically associate with social networking will never themselves drive transactions for online retailers. However, that does not preclude Facebook from someday having a bright future in sales.
For now, though, that future is cloudy. The Forrester research bolsters earlier reports with similar results. A holiday shopping survey from last season showed very strong engagement with online marketing tactics among holiday shoppers. More than 80 percent of holiday shoppers surveyed engaged with one or more interactive marketing tactics before making a purchase, and about half of shoppers engaged with more than one tactic.
Yet fewer than 2 percent of orders had a social touch point, putting social marketing dead last among online marketing tactics, regardless of whether it was considered alone or paired with another tactic. In other words, there was scant evidence that social drove purchases either directly or indirectly.
On Cyber Monday, soft goods (basically, apparel, and accessories) retailers saw a 45 percent lift in orders from social links. This appeared to correlate with the many special, one-day offers available that day, as people shared links to hot deals. Even then, social networks only accounted for 4 percent of such word-of-mouth sharing, versus 9 percent of respondents who purchased after receiving an email from a friend.
Sharing offers seems to be the sweet spot for social when it comes to brands as well. A Razorfish study in 2009 found that among even a very online-savvy set of respondents -- 100 percent shop online, 84 percent share links and bookmarks with friends, 74 percent visit social network sites, 56 percent have a smartphone -- only 25 percent follow brands on Twitter. Of those who do, 44 percent say it is primarily to get offers. Of the 40 percent that have "liked" a brand page, 37 percent say it was to get exclusive offers -- the top reason cited, even above "because I'm a customer."
Other studies show that even fewer consumers (13 percent) say they visit retailer social network pages, and fewer still (7 percent) say they research products on social networks. According to Mulpuru, two-thirds of retailers say that if Facebook disappeared, it wouldn't affect sales at all, and only 7 percent of them say social is an effective customer acquisition channel.
Buysight's study of our own e-retailer client data revealed that only 1.21 percent of traffic originated from Facebook during the study period, in line with Forrester's numbers.
Is social media a total dead end for retail marketing? Mulpuru's certainly a skeptic, and her feeling that people use social networks to share social and life experiences, not commercial ones, rings true to me. Marketing guru Seth Godin used to talk about different consumer modes and the need to understand what mode someone was in to market to them effectively. Shopping is often a social activity, but people typically decide to go shopping with friends; they don't usually spontaneously shop in the middle of socializing.
Consumers do talk about products they like, want to buy, and are researching, but general social networks like Facebook aren't an ideal forum for that activity. Facebook is optimized for creating a broad web of relatively shallow connections between people with something, usually another person or an organizational affiliation, in common. It is optimized for real time, not research. Conversations bob up and down your wall, and comment threads are designed for short quips, not meaningful analysis and discussion.
Social shopping sites have taken the same model and tried to optimize it for shopping, encouraging members to share products they've bought and ask others for help finding products they're interested in. It's probably too soon to tell whether they'll succeed, but at least they're explicitly shifting consumers into a shopping mode and not competing with the sharing of hysterical pictures of children and pets and expressions of undying loyalty to their local baseball team, despite their pathetic performance in last night's blowout.
Those sites need to make money, and many of them are counting on affiliate marketing income. That means long, meaningful conversations about products are less productive than quick hits with product links that drive purchases. It remains to be seen whether this is more effective at driving e-commerce than a comparison shopping engine (CSE). CSEs, and general search engines for that matter, are highly optimized for surfacing product choices, without a lot of content, to consumers who generally know what type of product they are searching for. These engines deliver consumers with specific purchase intent directly to the products they want to buy.
Even Amazon is a more efficient social shopping engine than Facebook. Nearly anything you can imagine is sold through Amazon, and not only can you get price comparisons, but there is also a rich set of reviews from fellow consumers and even third-party purchasing choices through the display and text ads running on the page.
Visit an enthusiast website if you want to see real engagement around specific products. Bikeforums.net has, as of this writing, 8,120 threads with 108,553 posts just on folding bikes. AVSForum.com has 2,313 threads with 107,129 posts just about ultra high-end home theater gear that costs more than $20,000. These sites are entirely focused on bringing together people with common interests in specific activities and the product categories that enable that activity.
Such sites are optimized for connecting people that have no connection other than a common interest, and they have toolsets that help encourage education, long discussions, product research, analysis, and discovery. Further, they foster their own sense of community, which leads to loyalty, higher visit frequency, and regular use of the forum to drive purchase decisions. They usually restrict explicit retail activity to well-defined areas and advertising, which makes retailer participation in the product discussions even more effective and authentic.
Despite the slow start, however, I think social networks do have a future in retail marketing. For one thing, the sheer size and level of participation on Facebook should make it a good source of purchase-intent data. That data is the key to making display advertising work for retail marketers because it delivers the kind of intent and response dynamic that has been so successful for retailers in search engine marketing. When you know, explicitly or implicitly, what a consumer intends to buy, and you can deliver a product-specific ad that responds to that intent, you're going to get a positive return on investment.
In our industry these days, we're all paying a lot of attention to data-driven targeted display because it gives us new options for delivering advertising to audiences and individuals -- not just locations on a page. We can separate the data we use to target an ad from the decision about what ad to serve from the location where we serve it.
This level of control and optimization is generating higher performance -- higher click-through rates, higher conversions on e-commerce sites, and lower costs of acquisition. It allows retailers to move from thinking of display advertising as an expense to a cost of goods sold and -- eventually -- to a revenue driver. It makes display advertising a cost-effective direct response medium for retailers, which is why it is starting to finally capture some of that search-heavy retail ad spend that accounts for about 30 percent of all online advertising dollars.
Facebook is just not there yet, but I expect it will be. At this point, you can't think of Facebook as a publisher or even as a specific advertising tactic. It's not even a medium -- it's an online ecosystem, and it will represent a significant advertising opportunity across a variety of marketing tactics: community engagement, loyalty programs, search marketing, branded entertainment, video, promotions, and of course, display advertising.
For one thing, we're only beginning to see what can be done with the Facebook platform. You can now rent movies on Facebook, for example. As new applications emerge within Facebook, we'll see more diversity of modes among Facebook visitors, and Facebook will continue to develop ways to get consumers to shift from social mode to entertainment mode to shopping mode and back again, creating opportunities for marketers.
Finally, there's reach and frequency. According to comScore, Facebook now accounts for nearly one-third of all online display ads in the U.S. Those ads can be targeted using Facebook's social graph, which isn't particularly effective for retailer advertising, but I believe it's only a matter of time before marketers can deliver ads to product-specific audiences based on purchase intent signals from within and outside of Facebook.
If Facebook allows marketers to receive those signals and process them into insight about purchase intent, and gives us very automated, granular, and dynamic control over ad content and delivery, it's going to play a major role in retail advertising.
Jeff Weitzman is the chief marketing officer of Buysight.
On Twitter? Follow iMedia at @iMediaTweet.