Tag Archives: statistics

Mobile Design and What the Numbers DON’T Tell Us

Posted by

0

I, like so much of the Western Hemisphere, spent part of my weekend shopping for dolls, pajamas and the latest electronic gadget.  And like so many others, I turned to my mobile phone for support on more than one occasion.  I am far from unique in this regard.  What caught my eye was the number of children I saw with a cell phone in hand – not their own (although there were no doubt some pre-tween kids among the throngs who did indeed have the pleasure of owning a very expensive smart phone, though I hesitate to think they were the norm), but a parent’s phone, which they used to play games, watch YouTube and generally make the tedium of shopping less pronounced.  What struck me as relevant is that when asked in surveys, the data frequently comes back saying that parent’s rarely give over their phones to the kiddies while shopping.  Observations in the field would imply quite the opposite is true.  Mobile phone use in a shopping environment is not just about the owner of the phone, it is about the parent/child dynamic and the underlying practices that go beyond procurement of goods.

Why does it matter?  It matters because while we have plenty of data about what people say they do, we have precious few insights about what’s really going on.  We are still in the wild west mentality of mobile design and need to get a better understanding of the range of contexts as we design mobile experiences for shoppers.  One size does not fit all and the numbers, while compelling, mean little if we don’t address the bigger questions under the surface.  It matters because a good mobile design and a creative mobile strategy can mean the difference between a useful application and a waste of millions of dollars in development, reduced brand equity, etc.

As another example, look at the numbers around the Hispanic market.  The numbers show that Hispanics are generally younger and more technologically savvy. AOL’s Hispanic Cyberstudy reports 46% of Hispanics who are actively online are under the age of 35.  32% of Hispanics access the Internet through their smartphones, compared to 20% of the general market. With roughly $1.3 trillion in buying it’s no wonder that Hispanics are a significant target for marketers.  But the numbers don’t address the bigger questions.

First, which “Hispanics” are we talking about?  Do the numbers refer to 3rd generation Cuban Americans with money or 1st generation Guatemalan farm workers? While or bias would probably lead us to assume the latter isn’t using a smart phone to shop, there is neither qualitative nor quantitative evidence to support this; is simply a matter of our own prejudices and preconceived notions about immigrants.  Second, Hispanics skew younger as a whole when compared against the total US population, so of course they are accessing the web through their phones more frequently than the general market – the data show that this is the case for all younger shoppers, so the difference between Hispanics and everyone else is misleading.  Finally, simply being online via a mobile device doesn’t necessarily mean ecommerce is taking place or that it is even desired.  Is the phone being used to supplement computer-based online interaction or is it a surrogate?  Do the numbers even reflect  use or do they reflect self perceptions, desires, the search for status, etc.? The point is that while we can infer quite a bit from the numbers, we are filtering them through our own biases.  Until we rethink the questions a bit we are designing based on potentially false assumptions.

Regardless of the populations to which we want to cater, designing a good mobile experience should entail getting on the ground and spending time learning what’s really going on.  And sometimes that means getting inventive about how we gather insights.  For example, if you want to understand how a good mobile banking site should operate, it isn’t enough to know the numbers of people using financial apps.  You need to understand how they conceive of money, when they do their banking (e.g. work, at home once everyone is in bed, on the train, etc.), and how they view their bank (many people hate their bank, but the cost of switching doesn’t outweigh the pain).  If you want to understand what the ramification of something going wrong are, spend time looking at mobile transactions in a place like Afghanistan, where banking can be a deadly affair – whatever people are doing in a place where bad mobile design can get you killed will probably shed some light on what is and isn’t necessary the world over.

The point is simple.  Good mobile design means getting your hands dirty and learning how context shapes how people use a given app or mobile site.  If all you have are statistics, you’re bound to create a solution that addresses the wrong problem.

Insight vs. Absurdity

Posted by

4

Facts and insights are not the same thing and yet that is precisely what a great deal of market research seems to present. Insight is the understanding of a specific cause and effect in a specific context.  Insight is the act or result of understanding the inner nature of things, or what was in Greek term noesis.  From a business perspective an insight is a statement based on a deep understanding of your target consumers’ attitudes and beliefs, which connect at an emotional level and provoke a clear response which, when leveraged, has the power to change consumer behavior.  In other words, to get them to love you and buy your stuff.  Insights are not poorly constructed inferences based on statistics derived from unrelated questions.  And yet, this is precisely what defines “insights” to many, if not most business folks.  Perhaps that’s why there are so many mediocre businesses and so few innovative one.

I just listened to a webinar on the changing face of mobile shopping.  Of course, we were presented with the standard awe-inspiring numbers showing that internet and mobile sales have been increasing through time.  While that in itself is simple banality, it is the leaps of logic that are made that truly confound.  Case in point, because web holiday sales have increased roughly 3% – 4% a year for the last decade, it means that people are unwilling to “brave the crowds and bad weather” when shopping.  Let’s step back for a moment and think about the absurdity of this.

First of all, it doesn’t address what people are shopping for online or why they’re doing it.  Just because I buy, say, toys online (largely a commodity), it doesn’t mean I’m shopping less at the mall or the specialty store.  The difference is that the experience is simply that much more important in a brick and mortar setting – it is about finding the “perfect gift” not making sure I land the Barbie Dream House.  People are still more than happy to rise at 2:00 a.m. to make it to the Black Friday store opening not because of the huge savings, but because it is an event.  People still flock to the toy store, they just avoid the disheartening experience of a Toys-R-Us.

Add to this the fact that a good half of the US population lives in a geographic area where the weather and the crowds just aren’t that bad.  The point is that the inference made by the researcher tells you more about that researcher than it does the data. Numbers lead to conclusions that are simply without merit.  It’s rather like saying that if 25% of auto accidents involve people named Jones, then the word “Jones” must cause traffic accidents.  And this is precisely my problem with what so often passes for insights, whether derived from quantitative or qualitative methods.

Rather than attempting to use solid analytical tool and understand the complexity of a given system, we make leaps based on untested, under-analyzed information and present it as it we’ve found the holy grail of insight.  There is a science and an art to turning data, anecdotes, interviews, etc. into insights.  It means slowing down and questioning every data point.  Literally.  What does this mean?  What questions do we need to ask now?  What variable exist that influence what we’ve found?  How does it relate to other elements in the system that will lead to new ideas, products, service, etc.? What biases do we have that are clouding our judgment? What information about the human condition can we derive from our facts?  And are they facts at all?

An insight is not an observation of behavior pulled from research. It isn’t a collection of data.  It is the analysis and interpretation of data that induces meaning and furthers understanding of a situation or issue that has the potential of benefiting the business or design process. It is the understanding that re-directs thinking about that situation or issue.  It requires hard work to get there, not half-baked ideas rooted in supposition.  Until we start to demand more from the people who sell “insights,” the real thing will remain sadly elusive.

 

 

Stats Are B.S. 9 out of 10 Times.

Posted by

0

Nowadays, every company and organization within a company has an agenda of some sort and has taken to throwing out statistics, all in an effort to convince naive people into embracing that agenda. Rarely are the statistics meaningful in a holistic sense. Rather they are meant to bolster an opinion that has no real depth or substance.  Indeed, it more dangerous than the “gut feel” information ethnographers are accused of practicing (by people lacking real knowledge of ethnography).  Why? Because they are absolutes grounded in a singular worldview – they lack any collaborative checks and balances.

The truth be told, statistics are absolutely worthless when it comes to proving anything at all. Statistics are only as good as the method used to come by them, and the way they are presented.  And they are only as good as the minds that construct and apply them.  When numbers are involved, arrogance becomes more dangerous because it holds a weapon the uninformed tend to fear.  In other words, stats are often more about the person talking than anything meaningful.

So the next time someone throws statistics at you, keep an open mind. You do not want to become a victim of misinformation by statistics.

While Buying Beer Today…

Posted by

1

I was shopping for beer today after several hours of planting and couldn’t help watching and listening to the people around me. An woman of around 60 was buying a bottle of wine and I noticed she had a tattoo on her all-too-grandmotherly wrist.  A small, pink star.  And then there was the stereotypical suburban dad buying a six pack of Warsteiner, a six pack of Alpirsbacher and a six pack of Hofbrau.  Overhearing his phone call it turned out he was a finance guy from 8:00 to 5:00 five days a week, but on the weekend he became the beer aficionado. Based on the kinds of information derived from traditional segmentation, these folks didn’t fit. And yet, they are the postmodern shopper.

The days of mass marketing may be coming to an end in many respects.  The advent of social media, incredibly rapid modes of communication and a postmodern view of socio-cultural ties that allow us to largely construct  our identity from moment to moment have changed the way we group, think and act. These days we’re the tribal people, not the demographic probabilities of a region or zip code. Over the past few years especially, what we do and why we do it is becoming of increasing interest to business. And while that may have always been true in a macro sense, the interest has now shifted to the multitude of grouping and sub-grouping to which we flock. It’s humans rather than numbers that they observing as we go about our daily lives and adapting their messages, products and services to fit the moment as well as the person.

For instance, how iron workers with advanced degrees in English Literature  react to a mountain of texts on their phones; why retired women are using their iPads to search for shamans to help heal their ills; how we do the grocery shopping, how we pamper the cat. And it is this sort of thing that is perhaps the most relevant to both the buyer and the seller.

Why does it matter? It’s in the math. 70% of purchase decisions happen in store. 68% of in-store purchases are impulse buys. 59% of purchases are unplanned. Looking at those numbers it speaks not only to the need to develop experiences that draws people in, but ones that keep them coming back again and again. This doesn’t happen when the only choices are shades of vanilla. It happens when you start to think of marketing as an ever-shift process that speaks to the mercurial nature of the human condition.

By Gavin