Infographic | Where the Wealthy Shop Online

US Wealthy Shoppers Online Source: Signature9

Where the Wealthy Shop Online Feb 09-10 | Source: Signature9

NEW YORK, United States — As we emerge from the ruins of a deep recession, where are wealthy consumers shopping most? An analysis of online traffic statistics for February 2009 to February 2010, gathered from Google Ad Planner, reveals some interesting traffic trends, although these are only based on estimates.

The Numbers

Macys.com attracts more visitors with an annual income of over $100,000 than the websites of upmarket department stores Nordstrom and Neiman Marcus combined. In fact, mid-range stores like Kohl’s and JCPenney are reaching more affluent shoppers online than their luxury counterparts like Neiman Marcus, Bloomingdale’s and Saks.

What’s more, amongst the top 20 US apparel sites attracting the largest number of wealthy online shoppers, 30 percent are flash sale sites like Gilt Groupe and Rue La La. If you include discounter Bluefly.com, sites dedicated to price reductions represent a full 35 percent of the top 20. Indeed, Gilt Groupe attracts more visitors earning $100,000 or more than any other online-only apparel site except Zappos, the 11-year-old company acquired by Amazon last year in a deal valued at $1.2 billion. Not bad for a company that’s barely 3 years old.

The Insight

Even as the Dow edges over 11,000, a psychological mark that some market observers have been looking to as a sure sign of economic recovery, there is evidence that affluent shoppers are still looking for real value as they do the cost versus benefit calculation. At moderately priced stores, wealthy consumers are comfortable shopping at full price, but for luxury goods, these same high earners prefer to wait for seasonal sales or visit flash sale sites.

Make no mistake, fashion-focused consumers still lust after luxury labels, but getting the best deal has become far more important than being the first to buy. Given a choice between securing the right size or colour immediately, or securing the best price among reduced options later, even those who can afford to spend now have decided to take their chances at discounters or wait for trends to trickle down to mid-range retailers. Indeed, for recession-hardened shoppers who are accustomed to seeing heavy winter clothes marked down when it’s cold, and lighter summer clothes on sale when the weather is still warm, this “wait and save” mindset looks like it may be here for some time yet.

The Opportunity

One thing is clear: for upmarket shopping sites, it would be foolish to disregard the success mid-range department stores and flash sale sites are having in attracting wealthy visitors. Indeed, it appears as though luxury retailers have ignored an opportunity that’s been sitting under their noses all along: their outlets.

Saks, Neiman Marcus and Barney’s all have outlet locations, but these have not been heavily promoted. Retail sites selling in-season stock have large zooomable photos, eye-catching banners, shipping promotions and affiliate programs that help them make their way into social shopping engines and online magazines promoting shopable content. But in sharp contrast, outlets like Off 5th, Nordstrom Rack and the Barney’s Outlet have nothing more than spartan webpages listing store locations. Last Call (Neiman Marcus) has its own website, but it’s a relatively uninteresting and shallow experience that also offers little more than a list of addresses.

With flash sale sites and discounters like Bluefly and Net-a-Porter’s The Outnet replicating the outlet model online to overwhelming success, one has to ask: why aren’t more high-end retailers promoting their outlet brands on the internet in a bigger way? A look at the numbers suggests this is an opportunity worth seizing.

YM Ousley is Publisher and Founding Editor of Signature9 and an internet marketing consultant.

US Wealthy Shoppers Online Source: Signature9

Where the Wealthy Shop Online Feb 09-10 | Source: Signature9

The original “Where the Wealthy Shop” Infographic appears here.

Post a Comment

15 comments

  1. This is wrong “reveals some interesting facts”
    These are not facts! This data is comprised of predicted figures from various aggregated sources all estimated.

    Quantcast, Google Ad Planner, Alexa etc are all estimated figures with VERY high error rates, especially when it comes to demographic prediction.

    There is no excuse for this, it’s plain old sloppy journalism. This should be prepended by a disclaimer stating all these figures are estimated only.

    How Google Ad Planner works:
    http://www.google.com/support/adplanner/bin/answer.py?hl=en&answer=175532

  2. You’re right Alistair, we should have mentioned that the figures from Google Ad planner are estimates. This change has now been made. The results, however, are interesting nonetheless, even if they are only directionally accurate. And, the fact that high-end department stores are neglecting their own sales sites, also holds true.

    Imran Amed, Editor from France
  3. I’m afraid the whole argument is weak and misleading, these department stores might not have strong sales/discount sites but any correlation (represented as fact here) is drawn from unreliable statistics.

    Understanding complex multi-agent systems like this requires a vast knowledge of macro-economics and even the new field of behavioral economics to understand the implicit psychology of shopping and bargain hunting. Even then you’ll find different groups of shoppers (multiple use-cases) with different thought patterns, possibly throwing in a little psychophysics in to the equation. Macroeconomics teaches us that over-pricing a product can sometimes lead to more sales griffen vs veblen goods.

    I really don’t think “insights” can be drawn from such vague statistics without proper investigation, especially when I see no credible accreditation from the author.

  4. Hi Alistair, I’m used to working with various online tools, and to some degree they all have flaws. That said, if you consider the wide variety of data Google has access to – toolbar data, analytics, advertising tracking codes (don’t forget, DART publishers are included in that group now) on both the advertiser and publisher side, as well as their own search data, the information is likely fairly accurate in terms of trends.

    Is it down to the single user accurate? No, but there’s not any publicly facing service that is. Quantcast is fairly accurate for publishers who install the tracking pixel directly, for those who don’t it can be massively inaccurate. Since Alexa relies primarily on toolbar data, it’s fair to assume it skews to the demographic that might have it installed. Even services like comScore and Hitwise rely on a combination of panel and factual ISP data, and especially for smaller sites they can be inaccurate as well.

    Whether Macy’s only gets 5.5 million visitors earning over $100k+, or Neiman’s actually gets 1.6 million in that bracket, the trend and the gap is still there.

    While Neiman’s positioning may be exactly what those 1.x million visitors are looking for, it would be a smart move to offer something to the 3.x-5.x million who are going with Zappos or Macy’s instead.

    YM Ousley from Paris, Île-de-France, France
  5. “I’m afraid the whole argument is weak and misleading, these department stores might not have strong sales/discount sites but any correlation (represented as fact here) is drawn from unreliable statistics.”

    With no offense intended, what statistics do you have to offer that would suggest the ones from Google adPlanner are unreliable?

    Visitor count, I’ll give you, could be off from 10-50,000 people, but even up to 100,000 missed people in either direction that’s not enough to change the obvious pattern.

    “Macroeconomics teaches us that over-pricing a product can sometimes lead to more sales griffen [sic] vs veblen goods.”

    Giffen goods are in most examples related to food, and normally those at the lower end, but let’s apply the theory to clothes. If Macy’s average price point for clothing rises from $150 to $170 while Saks’ holds steady at $500, the person shopping at Macy’s may be spending more, but are they doing so at the expense of the $500 item they may have included earlier? I’m not saying this definitively answers the question, but look at sales and earnings reports from the companies mentioned and I think you’ll find the online trend reflected in dollars and cents.

    And if you follow the theory, it would support the position that sales at the $170 price point are increasing to the detriment of sales at the $500 price point.

    If you argue for luxury apparel as a Veblen good – one of conspicuous consumption, I wouldn’t go against that, and in fact would say that’s where many brands found themselves in trouble. Now there are still over 1 million people making $100k or more visiting the Neiman Marcus site, so I’m not taking the position that the people who support that model have disappeared.

    But there is a clear gap in the number of high income shoppers visiting the top end sites, vs. the value sites. They’re aware that Neiman Marcus and Saks and Bloomingdale’s have luxury goods at full price, and they’re visiting Macy’s, JC Penney, and Kohl’s instead.

    It doesn’t take an economist to see a clear trend.

    “I really don’t think “insights” can be drawn from such vague statistics without proper investigation, especially when I see no credible accreditation from the author.”

    I’m not sure what you would consider proper investigation – in my original posting, I mentioned that there may very well be other factors that work in favor of luxury retailers (higher visitor values, higher average sales, etc), but the data is clear. A margin of error in this case might mean that Rue La La gets 10,000 more $100k visitors than Gilt, or that Bloomingdale’s and Neiman Marcus are even so far as that demographic.

    Unless you can point to specific data that refutes the overall trend that $100k+ online shoppers are visiting mid-range and discounter sites in numbers equal to or greater than high-end, I don’t think the fact that the data is estimated changes anything.

    YM Ousley from Paris, Île-de-France, France
  6. Allistair – i agree this article is vague and far from insightful and wanted to comment on it but what in the world are you talking about? You’re last comment is the most random thing ever.

    “Understanding complex multi-agent systems like this requires a vast knowledge of macro-economics and even the new field of behavioral economics to understand the implicit psychology of shopping and bargain hunting. Even then you’ll find different groups of shoppers (multiple use-cases) with different thought patterns, possibly throwing in a little psychophysics in to the equation. Macroeconomics teaches us that over-pricing a product can sometimes lead to more sales griffen vs veblen goods.”

    how is that relevant to anything? Behavioral economics predates Neo-classical economics, see adam smith – how in the world can that be considered new? How in the world can you even get Veblen and GIFFEN goods in a comment on this post. Truly amazing. Giffen goods have nothing to do with luxury apparel. Maybe luxury versus normal goods but there is no way to bring Giffen goods into the discussion

    my sweet gucci loafas from New York, NY, United States
  7. I am not saying you are wrong I am just saying you do not have barely enough evidence to prove you are right and you are vastly oversimplifying the situation.

    Regarding those stats, speaking from experience, I have used Google ad planner and tested against a number of large domains for which I have access to real data (including a massive luxury retailer domain). Visitor data is somewhat reliable I have seen mis-caclulation on Google ad Planner traffic by more than 50% on some larger sites (under-reporting visitor counts). Demographic data is even worse.

    We currently employ Quantcast tracking on one of the sites I work on and could see the distinction between estimated and real data when we turned on tracking (demographics still inferred even with tracking).

    There are a number of flaws in this principle of online discount shopping, and it’s a lot more involved than simply trying to correlate a single demographic to visitor count.

    You’re extrapolating shopping behaviour from visitor count and salary (not even sales and product price point).

    You have no data correlating any of this to sales.

    Auto-stock management, logisitcs and fulfillment means less overstocking in eCommerce, which could mean less dead stock and no large need for department stores to require a discount outlet.

    Companies are making less profit, less need for tax write-offs to balance out profits, so it might not be in their benefit to overstock.

    There are probably many different agents and reasons why and how people shop, and also the reason why online department stores don’t do discount.

    Comparing discount shops to current season department stores. Discount stores have different business models to the larger department stores.

    Breaking it down into a simplistic system from various uncontrolled environments like you have is not how complex economic systems should be analysed even on a basic level.

    These are merely correlations based on unreliable data, no cause and effect pattern can be deduced from this as there are many unidentified and unknown agents acting upon the system.

    Your insights expressed here are based on no reliable tangible evidence.

    The only meaningful thing I can take away is that some Department stores have seemingly abandoned their discount sections/sites. Reason unknown. Why not just ask them why they have done so?

    ps. Giffen goods are not just related to food stuffs, demand curves can be applied to many different sectors (not necessarily on entire product groups though) pick up a few microeconomics books, the subject is fascinating.

  8. ps. Giffen goods are not just related to food stuffs, demand curves can be applied to many different sectors (not necessarily on entire product groups though) pick up a few microeconomics books, the subject is fascinating.

    your logic is so misguided its hard to even argue.

    Demand curves can be applied to everything. Giffen goods, however, by definition violate the law of demand. Giffen goods are inferior goods where as price rises quantity demanded rises. Additionally Qd is only a function of P. nothing else.

    YM”s attempt at clarifying this for you with a luxury apparel analogy was a good one but still doesn’t even really work which i think YM would agree with.

    Veblen yes, Giffen no

    my sweet gucci loafas from New York, NY, United States
  9. “Visitor data is somewhat reliable I have seen mis-caclulation on Google ad Planner traffic by more than 50% on some larger sites (under-reporting visitor counts). Demographic data is even worse.”

    I’m a little confused by this statement. I’ve worked with various types of analytics programs, including raw server logs, and I’ve agreed with you that no publicly facing numbers are going to be 100% accurate. A 50% difference on a site over 500,000 visitors per month is highly unusual though. For smaller sites? Absolutely possible, but once you cross the half million mark, the vast majority of panel based measurements become closer to accurate. Also, the one thing server logs can’t tell you is who your audience is composed of.

    It is a misconception that everything can be measured online, but you can certainly get more data than you can from reader surveys that take 1-2000 responses and extrapolate them over an audience 10 to 20 times the size. The TV industry still relies on panel based data from a relatively small sample set to determine who’s watching what.

    “There are a number of flaws in this principle of online discount shopping, and it’s a lot more involved than simply trying to correlate a single demographic to visitor count.”

    I’m certainly not taking the position that $100k+ shoppers are the only ones keeping Gilt Groupe or Neiman Marcus’s online business afloat, and I don’t think that’s implied. I do think it’s interesting that the data – even if estimated, shows more people in that demographic (one that’s likely a luxury brand’s target market) are visiting mid-range retailers and specialized sale sites than high end sites.

    If you sell gourmet dog food, and find that the majority of people visiting your store own cats, it doesn’t mean no one’s buying your product. It is an indication that you should consider different strategies, because the one you think is resonating with your audience may be different than the one that does.

    The original article this was expanded from is here
    Where the Wealthy Shop Online [Infographic] and has some context that wasn’t included here.

    “You’re extrapolating shopping behaviour from visitor count and salary (not even sales and product price point).You have no data correlating any of this to sales.”

    I would love to compare this with conversion rates (particularly within the specific demographic), average cart value and lots of other data that would make this more than a snapshot. As I’m sure you know, that’s not likely to happen. Talk to someone in marketing for the various sites, and you may get an idea of conversion rate, frequent shopper profile and other interesting information, but it’s not freely available.

    And while it wasn’t included in this piece, I have made the point that those are factors which could make a difference in sales. If Macy’s visitor value and average sale is 5x lower than Neiman Marcus’, it doesn’t matter that Neiman’s gets 5x less traffic. But there’s no two ways about it: someone earning $100k+ would certainly fall into the target demographic of Neiman’s and the brands they carry (because in theory, that’s where there’s the type of disposable income that allows people to buy a $2000 dress instead of a $200 one).

    That said, I don’t know many people who go to e-commerce sites with the intention of not shopping. Do they browse, come back, browse more, decide on favorites, then come back a few days later to buy? Sure, but there’s an obvious intent whether you’re going to Zappos or Saks, and it’s usually commercial.

    Comparing discount shops to current season department stores. Discount stores have different business models to the larger department stores.

    This is true, and I’m not suggesting high end retailers jump on the bandwagon by trying to become value stores. I am suggesting it would be smart to put more effort into the discount stores they already have, and establish them as their own brands. Net-a-Porter was smart about this in establishing the Outnet. Net-a-Porter retains their premium branding, doesn’t alienate shoppers who come to them for that. At the same time they appeal to consumers who shop on value rather than timeliness, and strengthen their overall position with the $100k+ demographic.

    “These are merely correlations based on unreliable data, no cause and effect pattern can be deduced from this as there are many unidentified and unknown agents acting upon the system.”

    I was asked to provide additional insight on my original piece which provided more basic assessments of the information, and focused more on audience composition. While I don’t disagree that the data may not be pin-point accurate, considering the source, I don’t think it’s unreliable to the point that you’re arguing. If I understand your position, it’s that Google’s demographics are missing $100k+ visitors to the upscale sites in such large numbers that the resulting conclusions – that wealthy people visiting online e-commerce stores are visiting mid-range etailers more often than high-end etailers – aren’t true.

    Again, if you have counter data to support that, I’d certainly be interested in seeing it, but I think Google has a large enough user panel to provide trend information on a relatively accurate level.

    Whether those $100k shoppers are collectively spending more money with Macy’s than they are with Neiman’s is something else entirely, but that they’re visiting 1 e-commerce site more often than another is something supported by multiple sites. Quantcast included. I wouldn’t put much stock in the visitor numbers, but even there, they put 33% of Macy’s 6.4 million visitor audience at $100k plus, and 44% of Neiman’s 960,000 interest in the same HHI. No matter how you slice it, that equals more wealthy visitors going to shop at Macy’s site than Neiman’s. Check ComScore, Hitwise or anyone else and I’d put money on the same trend there. The numbers may be different, the resulting pattern would be the same.

    The fact that wealthy online shoppers are visiting mid-range and sale sites in numbers equal to or greater than high-end sites is not in and of itself complex, related to stock management, logistics or anything else.

    YM Ousley from Paris, Île-de-France, France
  10. YM”s attempt at clarifying this for you with a luxury apparel analogy was a good one but still doesn’t even really work which i think YM would agree with.

    Veblen yes, Giffen no

    It was the closest analogy I could make using clothes as an example, but I agree, it’s not the same. A better example would be that the price of a Big Mac goes up, and so does demand, while filet mignon sales suffer because more people are eating Big Macs where as they used to eat Big Macs and filet mignon, and can’t afford both due to the higher Big Mac price.

    But I wouldn’t equate Macy’s or Gilt with a Big Mac (no offense to Big Mac fans), and I don’t think there an inherent correlation that $100k+ shoppers all gave up shopping at the high-end because prices at the mid-range went up.

  11. I am wondering as to what the conversion rates are for these sites. Have you looked into that stat Imran? Number of unique visitors is only the start. My guess is that even during the recession the unique visitors count probably din’t get affected. What probably changed significantly is the conversion rate.

    Rana M from Forest Hills, NY, United States
  12. I am wondering as to what the conversion rates are for these sites. Have you looked into that stat Imran? Number of unique visitors is only the start. My guess is that even during the recession the unique visitors count probably din’t get affected. What probably changed significantly is the conversion rate.
    vibram five fingers

  13. Who would have guessed that Macy’s

    tristaliu from Beijing, Beijing, China
  14. This is an awesome idea! I would have guessed Macy’s to be number one! If any of you would like to take the time to visit my website, please do. It is a good cause!! Thanks!