Nielsen Study in Ad Age: Confusion Abounds
Last week Ad Age published some research done by Nielsen suggesting that search played a small, indeed shrinking, role in online retail sales. According to Ad Age, the Nielsen study shows that less than 10% of retail sales came from search engines, and much of that was brand search.
That left many people scratching their heads, and some people fuming.
Scott Silverman, Executive Director of Shop.org asked me if I would interview Ken Cassar, VP of Industry Insights at Nielsen who authored the study, in an attempt to clear the air. The bulk of the interview can be found at the Shop.org blog.
I won’t repost the text of that part of the interview, but will summarize as follows:
The data is accurate for what it is. It represents data from the 200 largest retail sites, and shows the aggregated average percentages. In other words, the three largest “retailers” on the list have a hugely disproportionate impact on the final results. Those “retailers” are eBay, Amazon and Wal-Mart. Well, one could make a pretty good case that eBay isn’t a retailer at all, and that at this point Amazon is more of a “mall” than a retail site; indeed Google has started to consider Amazon a competitor.
Importantly, the numbers are not medians of the top 200 sites, and since most of the retailers who advertise through search aren’t among the top 200 even if these were medians of that contingent suffice it to say your numbers will look very different.
The point, according to Ken, is that brand matters. The biggest, most successful retailers in the world aren’t paying for most of the visitors to their sites; they’re generating those visits through the power of their brands.
There’s no question about this; the key to success for most profitable retailers is customer retention, not acquisition. Acquisition is expensive and generating repeat sales without the significant marketing outlay allows retailers to grow profitably.
However, all those navigational searches and direct visits to the site aren’t coming just from brand loyalty. In the case of Wal-Mart and other big retailers tons of marketing dollars are spent offline (print ads, circulars, TV and radio, catalogs, etc.) which drive much of the direct navigation as well.
Nielsen is not suggesting that those channels are more cost effective than search, only that they remain cost effective albeit harder to track. The question he says is where to put those offline dollars as newspaper and magazine circulations shrink. He thinks the answer is display ads online, and suggests that the current mechanisms for evaluating online display ads don’t capture all their value.
I recommend reading the first part of the interview over at Shop.org if you haven’t yet.
What follows are some follow up questions and comments between Ken and me:
George: It does seem to me that brand building for manufacturers is different from brand building for retailers selling commodities. It seems much harder for the latter to develop a unique selling proposition than it does for the former. To a certain extent “brand” is what you do every day to create a loyal customer base. The marketing surrounding that is hard. Arguably Amazon built it’s brand by doing, not by marketing.
If you’re not Wal-Mart, “Always low prices, always”, or Amazon selling everything with peerless fulfillment operations, how do you distinguish yourself?
Ken: There’s no question that Amazon and Wal-Mart have distinct selling propositions that lend themselves relatively easily to brand development. More important than any explicit marketing effort, exceeding customers’ expectations every time is going to establish loyalty, and the credibility of your brand. Without a returning customer base, a retailer is consigned to paying a search ‘tax’ with nearly every transaction. While there are a number of retailers that have differentiated themselves with service (Zappos, Lands End, LL Bean jump to mind), strong customer service is for all retailers a cost of entry. However, there are innumerable other ways that retailers can try to differentiate their brands, including; packaging, unique selection, local fulfillment, image presentation, product related content, return policies, etc…
George: Paid search is a particularly difficult channel for brand building we think. First, the ads are only shown to people who are looking for your products — like the yellow pages. The yellow page ads don’t convince someone to try yoga, they convince someone interested in yoga to try your studio instead of your competitor’s. Second, while it might be good brand building to have your search ads near the top of the search results page, regardless of the observed economics, for the sake of brand building, when you do the math the cost per thousand impressions of that branding is pretty steep, particularly if the real branding impression is a visit to the site where a $0.50 click translates to a $500 CPM.
Ken: This one is tough. I have seen studies that have argued that sponsored link advertising has a ‘bonus’ brand building value, and I suspect that there is something to be said for that. However, I think that there are more effective tools for brand development.
George: For an online pure-play retailer generating 50% of their sales through competitive search I gather you’re not saying they should do less search advertising. You are arguing for them to try to figure out other ways to get folks to come to the store (word-of-mouth, repeat purchases, what have you) by building brand awareness. In some sense, is generating that much of your traffic through search evidence of lack of brand awareness?
Ken: Over time, if a retailer is doing a good job in terms of delivering products that consumers value at a good price, with service that exceeds expectations, the bulk of its traffic should be organic. The only exception to this should be retailers that are growing rapidly.
A healthy retail business that has been around for awhile, I believe, should not be paying a ‘search tax’ on every transaction. There are two things that will allow them to do avoid this: positive past experiences reinforced with continued brand awareness.
I’m grateful to Ken for sharing his observations with me, and to Scott Silverman for giving me the chance to interview him.
Clearly, the folks at Nielsen aren’t anti-search, nor is their data inaccurate. Their data should not be used as a measuring stick for all businesses; the relative importance of search will depend on the size of the business, how long the business has been around, and how much offline advertising that company does to drive customers to their website.
Rather, the point is, as Ken puts it: “Brand matters.” These days social media allows “the word” to spread at Warp 9, meaning brands can be built and destroyed in a fraction of the time it used to take. Make sure “the word” on your brand is good, and think of creative ways to encourage its spread.