In the beginning… by rt3463df


									The Revival of Observational

       Dr Hugh Phillips
              PFB Inc.
           866 513 9908
         In the beginning…

• Genesis of Market Research, three approaches:

   – Qualitative.

   – Quantitative.

   – Observational.

• Observational popular as “completely objective”.

• But fell out of favour.

• Until end of last centenary renaissance in UK & Europe.
      Why the renaissance?

• Oversimplified version…

• Consumers changed their decision making until in-
  store – 73% decisions now made in-store

• Major shifts in marketing from conventional media to
  in-store communications.

• In-store communications fastest growing media,
  overtaking the internet.

• Therefore we need research into shopping behaviour.
    The research challenge.

• Ah but…

• Modern stores are too big and complex to shop

• Therefore the great majority of shopping behaviour is
  unconscious or subconscious.

• Let‟s prove it with you.
Look at these please

  Iju   rty uio   asd   fgh
  jkl   zxc vbn   mnb   vcx
  zlk   jhg fds   apo   iuy
  tre   wqw qaz   wsx   edc
  rfv   tgb yhn   ujm   ujk
How many can you remember?
• Maximum of conscious attention span is seven

• A „chunk‟ is a word, phrase or random three letter set.

• In practise, maximum is usually 3.

• So, how do we cope with the 4,500 pieces of marketing
  communications in an average store?
            The Answer
• In the next display, when you find the red square
  please raise your hand
               How long?

• There were 30 shapes/colours in that slide.

• If you spent just 1/4 second checking each out, would
  require 7.5 seconds!

• If you completed it in less than this you are using an
  heuristic - a short cut.

• In fact you used, a subconscious stored routine,
  called a schemata in psychology .

• It is these schemata we use to cope with the
  complexity of modern stores.

• To scan, browse, select products and, in general,
  shop ~

 You have just contradicted
• You process visually complex communications

• But when you research shoppers – in visually complex

• You assume they act consciously!
    Back to classic research
•    Criteria for research questioning.

    1.   Does the respondent have the information?

    2.   Can they give it?

    3.   Will they give it?

•    By definition respondents are unaware of
     subconscious or unconscious behaviour.

•    Therefore most shopping behaviour falls outside
     this criteria.
• So if you cannot use survey or qualitative – what can
  you use?

• Observation captures behaviour whether it is
  conscious, subconscious or unconscious.
           CASE STUDY

Cameras vs. Interviews

• Hardware store – decorating accessories aisle.

• Cameras: observed behaviour in the aisle.

• Interviews: survey of shoppers as they exited the

                  Cameras   Interviews

Looked at paint
                    60%       33%
 brush display
• Nearly half of the respondents were unaware of their
  behaviour seconds after it occurred!

• It‟s not that they had forgotten.

• Just unaware of subconscious behaviour.
        It gets worse…
• Often the research issue is what didn‟t the shoppers
  see / do?

• Imagine a question:

   “What displays didn‟t you see in store and why?”

• Traditionally

   – Video cameras record behaviour in store.

   – Output, DVD, analysed manually & entered into

• New

   – Modern software captures images and tracks
     behaviour automatically.
Traditional technology
                The    law
• In Canada, we can observe shoppers on camera for
  research purposes.

• Unlike USA.
              Example       1
• Testing in-store communications.

• Product, stand, sign, packaging etc.
   Two ‘perfect’ executions

         Glancing - stand

           Fixating - stand

        Glancing - product

        „Fixating‟ - product


          Design B
                               0%   20%   40%   60%   80%     100%
          Design A

Design route B is just not communicating. It cannot convert
attention gained by stand to attention to product!
             Example 2

• Evaluating how customers shop a gondola
   And now a perfect display!

• In fact, a really dysfunctional display.

• The centre shelves are being ignored.

• Plus major variations in impact levels – from 7% to 100%!

• Loss in sales around 30%, loss of profitability even higher!
            Example 3
• Quantifying how customers shop a store
A new concept store


                      Invisible areas:

                       Estimated loss of
                      sales 27% - all easy
                             to fix
New technology
            Image capture

• Micro camera into a display – gondola etc.

• Attached to a computer.

• Modern software can isolate an image as a human

• Then identify if it is looking at the display.

• What it does is sort of like this.

• The software can also identify with reasonable

• The gender and broad age group.

• To test a pack, signage, display etc.

• In comparison with manual processing:

   – Lower cost.

   – Fast turnaround.

   – Legal in the US.
  Tracking shoppers around
          the store.
• It is possible to automatically track shoppers around
  a store.

• But tends not to be of value, as we inevitably want to
  know what they do!

• What they:
   –   Look at.
   –   Interact with.
   –   Buy.
   –   etc.
              Clip cameras

• Micro cameras attached to spectacles.

• Either the respondents‟ own or supplied.

• The visual data is then loaded into analysis
• Video of respondent using clip cam

• Total data capture of all in-store behaviour and

• Incredible level of detail available.
             And now…

• Launch today of a major new service.

• Part of an international project for in-store metrics.

• Six countries now including Canada.

• Standard output:

   – Store traffic, total and by area of store.

   – Visual impact of all in-store communications.

   – Interaction levels.

   – Buying.
         Shopper profiles

• Each shopper is also profiled by:
   – Age
   – Gender
   – Shopping mission
   – Categories to be shopped
   – Frequency of store visits
   – Basket size
   – Etc.

• Just a selection of the massive database.
What works!
 Interrogating the database.

• Key usage of observational research is to answer
  those questions you forgot to ask.

• As observational research = total behaviour capture.

• Can be post hoc analysed ad nauseam.

• We got so fed up with running these we now supply
  an interactive database!

• Why observational research?

   – Consumers changed to in-store decision making.

   – So research changed, to understand the

• Because ~

        It begins and ends with the consumer.

To top