Hypotheses about tasks (in online shopping)

18 Aug 2006 - 1:28am
8 years ago
4 replies
780 reads
Mike Baxter
2004

Hi Robert

I have been struggling with similar thoughts specifically concerning
interactions on e-commerce sites. I too felt that I needed to de-construct
the magnitude/complexity/motivational issues in ways that could be
operationalised and tested more effectively. So I came up with a graphical
model of the interplay of factors going on inside the customer's head, as
they progressed through their shopping journey.

http://www.flickr.com/photos/71489601@N00/218208922/

[This is published in my Online Retail User Experience Benchmarks 2006
http://www.e-consultancy.com/publications/online-retail-user-experience-benc
hmarks/ - available to subscribers only but contact me offlist if you'd like
the section explaining how I see the model working and being applied]

So, how does this help me generate and test hypotheses about customer tasks?
Firstly, it teases apart predispositional factors - at its simplest level
this could be 'I love/hate shopping online' or, more subtely 'Amazon messed
me around on my last order (which in fact they have!) so I'd really rather
not put myself through all that again'. These, I suggest, resolve into a
single, and at least to some extent measureable, variable, purchase inertia
(the customer's inherent task-resistence). Then purchase impetus kicks in -
what makes them want to embark upon the task. I suggest this is determined
by the perceived need (I want a digital camera), the trigger (my mate's
asked me to take photo's at his wedding next week), and the shopping tactics
(I'll buy it at my local camera store but will have a quick look online
first). If the purchase impetus exceeds the purchase inertia, purchase
momentum results and the task commences. If the task is to be completed,
this purchase momentum must be sufficient to overcome all the hassles us
interaction designers have put in the way of task completion (okay, only
joking - well, sort of!).

Now we get into the specifics of task analysis. Consumer psychology seems
pretty well sorted on the interplay between constructed preferences (what am
I looking for and why) and the consideration set (what products am I
selecting for consideration and screening using my constructed preferences).
The only issue here is what comes out at the end of this decision-making
process. Obviously, if successful it results in product choice. But choice
is not the same as purchase and I suggest that purchase motivation (how keen
am I on my chosen product) and purchase justification play a key intervening
role in turning choice into action. The issue of purchase justification is
an interesting one. It is obviously an internal construction within the
customer's mind but research suggests that it is strongly influenced by
information presented to them. I've done a couple of jobs for clients
looking at how to customise purchase justifications depending upon the
journey taken by the customer and their interests/needs that can be inferred
from this. An interesting example of a 'purchase' justification engine was
the online manifesto published by the UK Labour Party during the 2005
General Election - they asked about your circumstances (family, level of
income, gender) and customised their manifesto promises accordingly (see
http://www.cx-i.com/online_politics/persuadePers.aspx).

As may be evident, this model is work-in-progress and there is still some
way to go before it is fully resolved (e.g. what exactly is the difference
between purchase momentum and purchase motivation - intuitively they seem to
be somewhat different things but I can't see how to operationalise that
difference). The model is also designed to be specific to online shopping
but I feel there are a number of aspects of it that may be generalisable to
other areas of interaction design.

My conclusion, however, in relation to your original post is that digging
deeper than simply number and complexity of tasks can lead to a richer and
more subtle set of hypotheses.

Mike

Comments

18 Aug 2006 - 10:45am
Robert Hoekman, Jr.
2005

> My conclusion, however, in relation to your original post is that digging
> deeper than simply number and complexity of tasks can lead to a richer and
> more subtle set of hypotheses.

Indeed. But the deeper I dig, the harder it gets to keep something in the
hypothesis that can be measured. Yes, I could probably measure the
frustration level of users by surveying them at the end of every step or
something, but it still seems like the results would come out as a guideline
rather than an empirical fact.

So, lacking the ability to measure difficulty in a way that stands to become
a hard and fast law, I'm focusing on the time factor. Enough can be inferred
about what to do with the information that it doesn't need to describe the
user's frustration level.

Thanks for your input. Great discussion.

-r-

18 Aug 2006 - 12:16pm
Mark Schraad
2006

Hey Mike - here is some input that I think you might find valuable.

The model (Extended Decision Making) best used for consumer behavior
is as follows: ( I use offerings as a generic group for products and
services, and attributes to encompass features, benefits and outcomes)

1) Problem Recognition
In this stage the person realizes a need. That need may expand to
group associated needs (that are later matched against product
attributes.)

2) Search for information
This is were families, friends, observations, inquiries (research or
shopping). This is often outsourced or partially outsourced. As an
example: some people buy what Walter Mossberg recommends... others
listen to what Walter says, look up specs, test drive at a store, ask
a current user, etc. This is the critical stage for most eCommerce
sites. The earlier in the process, the better the experience,
credibility ease of use, etc. the better chance of getting the sale.

3) Evaluation of Alternatives
Again, often outsourced to information leaders (reviewers,
salespeople, consultants or a buddy who obsesses), this is where the
final decision is often made.

4) Purchase (decision)
Pretty obvious.

5) Post purchase behavior
This is a critical step that many ignore. Follow up service calls,
installation, assembly or configuration documentation or help, tech
support, promise fulfillment of both product and vendor, etc. All of
this can result in buyer's remorse, or glowing referrals to friends,
families or complete strangers in the parking lot.

Many people unfamiliar with the behavioral side or consumer behavior
tend to over simplify this into what they call a "buying funnel." I
believe that the funnel inaccurately focusses on the convergent
process - of making a purchase selection from a large list. But, in
fact there is typically more divergent activity prior to the
convergence. The divergence is in direct correlation to the perceived
importance of the purchase by the decision maker (not always the same
person as the one paying.) So, if you want to use the funnel analogy,
it needs to be a dual funnel - one divergent and a second and final
stage that is convergent.

There have been a ton of research on consumer behavior in both psych
and marketing. That we are working with this process online only
slightly varies the considerations and certainly increases the
complexity for vendors while empowering the consumer. There is not a
need for IX or UE folks to reinvent all of this... we can take
advantage of the previous work done.

I would additionally recommend "How Customers Think" by Gerald
Zaltman, one of the most powerful works on the subject in many years.
His basic premise (and I am rounding from recall) that nearly all of
cognitive activity is subconscious, but more importantly, he proposes
that most of behavioral activity is also subconscious. The
implications of this render market and testing research such as
surveys or direct questioning nearly useless. I tend to agree. I
think observation and acute researcher interpretation of behavior
(especially in an equally incentived situation - as close to a real
world situation as possible) is the key. Most of the surveys and
usability research done is worth very little. Quantitative can give
you some great indications... but rarely proves anything when it
comes to human behavior.

Sorry, this went much longer than I had hoped... but could go on for
much more.

Mark

On Aug 18, 2006, at 2:28 AM, Mike Baxter wrote:

> [Please voluntarily trim replies to include only relevant quoted
> material.]
>
> Hi Robert
>
> I have been struggling with similar thoughts specifically concerning
> interactions on e-commerce sites. I too felt that I needed to de-
> construct
> the magnitude/complexity/motivational issues in ways that could be
> operationalised and tested more effectively. So I came up with a
> graphical
> model of the interplay of factors going on inside the customer's
> head, as
> they progressed through their shopping journey.
>
> http://www.flickr.com/photos/71489601@N00/218208922/
>

18 Aug 2006 - 2:00pm
Scott Bower
2006

And do no forget about the open experimentation on etsy.com. I would be
great if the 4 people that built and designed that system could come on
this list and share analysis from their site and how successful two
different paradigms in ecommerce work when they are presented in the
same system. The fact that they have introduced "distraction" and
"play" that relates to a shopping mall experience is fascinating. Of
course, customers/users are in an environment similar to an antique
store and not so much a hi-tech widget price and user feedback
situation.

I am sure they are capturing analytics through the interface since
scripting has become much easier.

Scott

On Aug 18, 2006, at 12:16 PM, Mark Schraad wrote:

> [Please voluntarily trim replies to include only relevant quoted
> material.]
>
> Hey Mike - here is some input that I think you might find valuable.
>
> The model (Extended Decision Making) best used for consumer behavior
> is as follows: ( I use offerings as a generic group for products and
> services, and attributes to encompass features, benefits and outcomes)
>
> 1) Problem Recognition
> In this stage the person realizes a need. That need may expand to
> group associated needs (that are later matched against product
> attributes.)
>
> 2) Search for information
> This is were families, friends, observations, inquiries (research or
> shopping). This is often outsourced or partially outsourced. As an
> example: some people buy what Walter Mossberg recommends... others
> listen to what Walter says, look up specs, test drive at a store, ask
> a current user, etc. This is the critical stage for most eCommerce
> sites. The earlier in the process, the better the experience,
> credibility ease of use, etc. the better chance of getting the sale.
>
> 3) Evaluation of Alternatives
> Again, often outsourced to information leaders (reviewers,
> salespeople, consultants or a buddy who obsesses), this is where the
> final decision is often made.
>
> 4) Purchase (decision)
> Pretty obvious.
>
> 5) Post purchase behavior
> This is a critical step that many ignore. Follow up service calls,
> installation, assembly or configuration documentation or help, tech
> support, promise fulfillment of both product and vendor, etc. All of
> this can result in buyer's remorse, or glowing referrals to friends,
> families or complete strangers in the parking lot.
>
> Many people unfamiliar with the behavioral side or consumer behavior
> tend to over simplify this into what they call a "buying funnel." I
> believe that the funnel inaccurately focusses on the convergent
> process - of making a purchase selection from a large list. But, in
> fact there is typically more divergent activity prior to the
> convergence. The divergence is in direct correlation to the perceived
> importance of the purchase by the decision maker (not always the same
> person as the one paying.) So, if you want to use the funnel analogy,
> it needs to be a dual funnel - one divergent and a second and final
> stage that is convergent.
>
> There have been a ton of research on consumer behavior in both psych
> and marketing. That we are working with this process online only
> slightly varies the considerations and certainly increases the
> complexity for vendors while empowering the consumer. There is not a
> need for IX or UE folks to reinvent all of this... we can take
> advantage of the previous work done.
>
> I would additionally recommend "How Customers Think" by Gerald
> Zaltman, one of the most powerful works on the subject in many years.
> His basic premise (and I am rounding from recall) that nearly all of
> cognitive activity is subconscious, but more importantly, he proposes
> that most of behavioral activity is also subconscious. The
> implications of this render market and testing research such as
> surveys or direct questioning nearly useless. I tend to agree. I
> think observation and acute researcher interpretation of behavior
> (especially in an equally incentived situation - as close to a real
> world situation as possible) is the key. Most of the surveys and
> usability research done is worth very little. Quantitative can give
> you some great indications... but rarely proves anything when it
> comes to human behavior.
>
> Sorry, this went much longer than I had hoped... but could go on for
> much more.
>
> Mark
>
>
>
> On Aug 18, 2006, at 2:28 AM, Mike Baxter wrote:
>
>> [Please voluntarily trim replies to include only relevant quoted
>> material.]
>>
>> Hi Robert
>>
>> I have been struggling with similar thoughts specifically concerning
>> interactions on e-commerce sites. I too felt that I needed to de-
>> construct
>> the magnitude/complexity/motivational issues in ways that could be
>> operationalised and tested more effectively. So I came up with a
>> graphical
>> model of the interplay of factors going on inside the customer's
>> head, as
>> they progressed through their shopping journey.
>>
>> http://www.flickr.com/photos/71489601@N00/218208922/
>>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>

18 Aug 2006 - 2:14pm
Mark Schraad
2006

Scott brings up a great point... and one that we debated recently on
a project (which I lost) - is the button labeled "shopping" or
"research"? Is the user interested in shopping or doing research? Is
this a peasurability experience or just another bothersome task that
the user needs to work through. How do we customize the experience to
accommodate these extremes and everything in between?

What I do know is that my significant other and I are compatible in
most every area, except shopping. She loves to browse and wonder off
topic. Searching for a new sofa last week, she spent a large amount
of time looking at quilts (not on the list). I tend to "attack" the
research, come to, or let a decision come to me, and buy... done
deal. Very efficient. While very cool - I would never use etsy.

Mark

On Aug 18, 2006, at 3:00 PM, Scott Bower wrote:

> [Please voluntarily trim replies to include only relevant quoted
> material.]
>
> And do no forget about the open experimentation on etsy.com. I
> would be
> great if the 4 people that built and designed that system could
> come on
> this list and share analysis from their site and how successful two
> different paradigms in ecommerce work when they are presented in the
> same system. The fact that they have introduced "distraction" and
> "play" that relates to a shopping mall experience is fascinating. Of
> course, customers/users are in an environment similar to an antique
> store and not so much a hi-tech widget price and user feedback
> situation.
>
> I am sure they are capturing analytics through the interface since
> scripting has become much easier.
>
> Scott
>
> On Aug 18, 2006, at 12:16 PM, Mark Schraad wrote:
>
>> [Please voluntarily trim replies to include only relevant quoted
>> material.]
>>
>> Hey Mike - here is some input that I think you might find valuable.
>>
>> The model (Extended Decision Making) best used for consumer behavior
>> is as follows: ( I use offerings as a generic group for products and
>> services, and attributes to encompass features, benefits and
>> outcomes)
>>
>> 1) Problem Recognition
>> In this stage the person realizes a need. That need may expand to
>> group associated needs (that are later matched against product
>> attributes.)
>>
>> 2) Search for information
>> This is were families, friends, observations, inquiries (research or
>> shopping). This is often outsourced or partially outsourced. As an
>> example: some people buy what Walter Mossberg recommends... others
>> listen to what Walter says, look up specs, test drive at a store, ask
>> a current user, etc. This is the critical stage for most eCommerce
>> sites. The earlier in the process, the better the experience,
>> credibility ease of use, etc. the better chance of getting the sale.
>>
>> 3) Evaluation of Alternatives
>> Again, often outsourced to information leaders (reviewers,
>> salespeople, consultants or a buddy who obsesses), this is where the
>> final decision is often made.
>>
>> 4) Purchase (decision)
>> Pretty obvious.
>>
>> 5) Post purchase behavior
>> This is a critical step that many ignore. Follow up service calls,
>> installation, assembly or configuration documentation or help, tech
>> support, promise fulfillment of both product and vendor, etc. All of
>> this can result in buyer's remorse, or glowing referrals to friends,
>> families or complete strangers in the parking lot.
>>
>> Many people unfamiliar with the behavioral side or consumer behavior
>> tend to over simplify this into what they call a "buying funnel." I
>> believe that the funnel inaccurately focusses on the convergent
>> process - of making a purchase selection from a large list. But, in
>> fact there is typically more divergent activity prior to the
>> convergence. The divergence is in direct correlation to the perceived
>> importance of the purchase by the decision maker (not always the same
>> person as the one paying.) So, if you want to use the funnel analogy,
>> it needs to be a dual funnel - one divergent and a second and final
>> stage that is convergent.
>>
>> There have been a ton of research on consumer behavior in both psych
>> and marketing. That we are working with this process online only
>> slightly varies the considerations and certainly increases the
>> complexity for vendors while empowering the consumer. There is not a
>> need for IX or UE folks to reinvent all of this... we can take
>> advantage of the previous work done.
>>
>> I would additionally recommend "How Customers Think" by Gerald
>> Zaltman, one of the most powerful works on the subject in many years.
>> His basic premise (and I am rounding from recall) that nearly all of
>> cognitive activity is subconscious, but more importantly, he proposes
>> that most of behavioral activity is also subconscious. The
>> implications of this render market and testing research such as
>> surveys or direct questioning nearly useless. I tend to agree. I
>> think observation and acute researcher interpretation of behavior
>> (especially in an equally incentived situation - as close to a real
>> world situation as possible) is the key. Most of the surveys and
>> usability research done is worth very little. Quantitative can give
>> you some great indications... but rarely proves anything when it
>> comes to human behavior.
>>
>> Sorry, this went much longer than I had hoped... but could go on for
>> much more.
>>
>> Mark
>>
>>
>>
>> On Aug 18, 2006, at 2:28 AM, Mike Baxter wrote:
>>
>>> [Please voluntarily trim replies to include only relevant quoted
>>> material.]
>>>
>>> Hi Robert
>>>
>>> I have been struggling with similar thoughts specifically concerning
>>> interactions on e-commerce sites. I too felt that I needed to de-
>>> construct
>>> the magnitude/complexity/motivational issues in ways that could be
>>> operationalised and tested more effectively. So I came up with a
>>> graphical
>>> model of the interplay of factors going on inside the customer's
>>> head, as
>>> they progressed through their shopping journey.
>>>
>>> http://www.flickr.com/photos/71489601@N00/218208922/
>>>
>>
>> ________________________________________________________________
>> Welcome to the Interaction Design Association (IxDA)!
>> To post to this list ....... discuss at ixda.org
>> List Guidelines ............ http://listguide.ixda.org/
>> List Help .................. http://listhelp.ixda.org/
>> (Un)Subscription Options ... http://subscription-options.ixda.org/
>> Announcements List ......... http://subscribe-announce.ixda.org/
>> Questions .................. lists at ixda.org
>> Home ....................... http://ixda.org/
>> Resource Library ........... http://resources.ixda.org
>>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org

Syndicate content Get the feed