Why Search May Not Click for Retailers
Consumers Going Directly to Retailers' Sites for Six Out of 10 Visits
NEW YORK (AdAge.com) -- As retailers get ready for the holiday season, their first instinct might be to throw as much money as they can afford into search. But recent traffic trends may point them otherwise.
|
Search Marketing: The State of Search Marketing: 2009
Google Dominates, but a New Microsoft/Yahoo Venture Could Mean a Stronger No. 2 How Online Retailers Generate Traffic
Complete Search Marketing Report 2009DataCenter Breaks Down the Share of Site Visits From Various Channels Explore the Full Version at Ad Age's DataCenter |
Nielsen found the majority of retailers' web traffic (61%, on average) comes from people going directly to a retail site -- consumers typing, say, Amazon.com into a browser address bar.
Nielsen's findings are based on second-quarter traffic for 200 online retailers. The analysis shows that comparison-shopping sites, such as NexTag or Scripps Networks Interactive's Shopzilla, collectively accounted for about 1% of traffic. Other referrals, such as affiliate programs or advertising (basically anything that wasn't direct navigation, comparison referrals or search), accounted for the remainder.
The 9.5% of traffic from search also likely included a good chunk of people conducting navigational searches -- typing Zappos into the search bar rather than searching for types of products (shoes) or product attributes (comfort footwear).
Brands matter
Ken Cassar, VP-industry insights at Nielsen's Online division, said a recent look at the top 50 search terms revealed only three that weren't branded -- and those were pornographic.
"When you take a step back and look at that together -- the fact that such a high percentage of people go directly to retail sites and even those that search generally have a pretty clear intent as to which website they'd like to go to -- it makes a compelling argument that brand and past experiences [with a marketer] matter an awful lot and will be far more significant determinants of success than any customer acquisition strategy that they're going to engage in," he said.
The findings, Mr. Cassar continued, "make an important case for the continued relevancy of display advertising. While search gets a lot of credit because it's quantifiable, there's a reason people are typing things like Expedia into the Google search engine."
When looking at the actual transactions and dollar volume that visits from various channels generate, direct navigation gains even more importance.
The challenges of tracking some less quantifiable channels (such as display or even offline channels) have caused some retailers to neglect them and instead focus on what's easy to measure.
Opportunity for display ads
Chris Paradysz, CEO and founder of internet marketing agency PM Digital, was surprised search only generated about 10% of retailer traffic but said the data shows "it's still the big push media channels that are still driving a lot of volume."
He suggested the research provides an opening for display sellers: If they can adopt some of the things that have made search attractive, such as the targeting and the pricing model, they will be in a good position to take some of the dollars marketers have been putting into search.
Still, search is a very important bottom-of-the-funnel marketing channel. It also helps retailers discern intent, in some cases helping them understand what consumer demand looks like.
"The primary reason you have to be there [in search] is it shows explicit intent and because it's so ROI-driven," Mr. Paradysz said. But beyond the issue of return on investment, he said, "there are very real branding implications that have impacts on downstream click activity."
Mr. Paradysz noted many people use search for product research, which influences their behavior down the line. He illustrated an example: You search for a red cashmere sweater, looking for who has the best deal. You see J. Crew has it for $119 and Neiman Marcus for $329. You don't buy it then but later head over to jcrew.com and make the purchase.
"You have to have the brand presence," Mr. Paradysz said, "because if you don't, you first don't benefit from the paid search, and second, you potentially lose some of that downstream activity."
RSS Feed
RSS (What is RSS?)
Or.. maybe I'm wrong and the sites are being looked at in depth and just aren't optimized.
Also, consider the fact that it is much harder to aggregate data on long-tail keywords. Many smart companies feast off these types of keywords for sales. Taken individually, they mean next to nothing, but altogether, they can make up a huge chunk of one's sales.
And with conversion (the idea of getting a person to your site and getting them to do a particular action [buy a product, download a PDF]), people may not purchase right away. The could come back to the site several times, or visit other sites.
It also doesn't tell us how many OTHER keywords were searched before they came to the keyword that landed the page hit.
Abbey - definitely check out Web Analytics in About an Hour a Day by Avinash Kaushik. He gives an easy to read, clear picture of all of this and goes into some detail on logical fallacies like the one above.
Also, regarding direct traffic -- it is more complicated than this article implies. For a good overview, check out a blog post titled "Direct Traffic is NOT What You Think It Is" (search for it!).
Nectar
All of this makes a case for digital display, yes, but it also makes a case for great customer experiences and word of mouth, as well as offline advertising. Our econometric models always showed that nearly a quarter of search engine brand queries for Travelocity were driven by TV and magazine and newspaper ads, at lower CPA than display often.
The above critics of the article fail to note that direct navigation (typing in the URL) as well as branded term users from search tend to have MUCH higher ecommerce conversion rates, from 2-10x over generic search clicks.
None of which is to minimize the importance of search (both SEO and SEM), which is certainly the fastest growing channel for everyone, along with search resellers like comparison sites.
Nice job Abbey!
J.Glueck
Look at the retailers thrown about in this article: Zappos.....Amazon. How many retailers are as known (top of mind) as much as these two? How many sites are known (top of mind) as your old company, Travelocity?
I've actually dove into analytics for retailers - from the inside. Fortune 100 companies with multi-million (billion?) dollar global marketing budgets may only get 10% of their traffic from search but what about everyone else? The data I've seen says search is a cornerstone of the average retailers business - not just a bullet in a marketing plan.
Focus Internet Services (Las Vegas SEO)
Macala
FashionablyMarketing.Me
I'd like to suggest a factor that hasn't been mentioned, which is view-through visitation and conversion. This is site visitation and/or interactions that occur as a result of exposure to a banner, but not a click (as you most likely know).
Now, I don't know about you, but I've seen PLENTY of Zappos banners, and in my experience running both paid search and banner campaigns for outdoor apparel companies, view-through drives a massive contribution in both visitation and conversion. But since the destination URL for nearly all banners is transparent to the viewer, what happens? That viewer, so profoundly affected by said banner as to visit the etailer's site, does so via direct browser visit. And, potentially, we can get figures like those quoted by this article. If this scenario is even remotely possible then view-through absolutely belongs in the conversation, and I submit that before this article's analysis and its damn statistics can pass muster, view-through contribution bears examination.
That doesn't really explain the 10% figure, though, does it? Another observation I can share from campaigns in which we've measured cross-channel exposure and contribution is that during a banner flight, site visitation and interactions attributable to paid search actually can decline. Overall site visitation and interaction increases (if we do our job right), but in a multi-channel media mix scenario it stands to reason that the total potential contribution of any single channel is constrained. There are only so many people to around.
This seems to falls well in line with other article comments about the importance of brand advertising. Search alone, year over year, typically yields a zero sum gain for top advertisers..shoving more money into the PPC kitty doesn't necessarily mean more [qualified] traffic. A multi-channel media mix designed to drive awareness and response is the logical solution. Meaning, any article pinning the absolute contribution of paid search to the chalkboard (white board? Wiki?) should acknowledge whether such a mix exists. (I'd also be interested in how contribution changes as a percentage as brand awareness changes, or as other market events such as corporate acquisition occur.)
As a quick follow-on to other comments regarding Web analytics and the final click, whether your analytics platform attributes a site interaction to the first or last click in a multi-keyword search path depends on your platform settings. The same is true for the length of time after which an interaction remains attributable to a click. Top-tier platforms allow users to customize these elements; all have default settings, a scenario we describe as we onboard clients onto new platform. I'm not commenting on the right or wrong way to do things, only that you have a choice.