22 SCS:DATA SYNCHRONISATION
benefits for a supplier is being able to code their data
once into a data pool of their choice and then publish
that to as many trading partners as they need to,” says
Besford. “Because the data is standardised and the
recipient accepts that as the type of data they want to
receive, you are not in a position, as a supplier, of having
to fill out different types of information with different
retailers… It reduces the amount of overheads you have
in terms of inputting data.”
Perhaps the greatest challenge facing the initiative is
how to agree common standards. “That goes to the heart
of what GS1 is about,” says Besford. “We are a standards
body. In the UK we have a group called the Data
Synchronisation Group and that has members who are
large suppliers – the blue chips – and it has all the major
UK retailers on-board. The group agrees that there needs
to be a standard around collecting ‘x, y or z’, that the
group will then review and make a decision on whether it
has a valid business requirement. If it has, and it is not
being met by an existing standard, it will then be put
forward by GS1 UK into the Global Standards
Management Process, called the GSMP. This comprises
all GS1 member organisations globally – 150
organisations. Most of the major retailers and suppliers
in the world are members.
Standard
“The proposed standard will be submitted to one of the
groups where it will be reviewed, agreed and put out for
everyone to comment on – eventually it will become a
global standard.” Besford agrees that this sounds a
lengthy process but, “we can bring in these standards
quite quickly; we brought some in last year in about
three months.”
Most of the data flowing around the GDSN is presently
related to food and beverage products. However, there
are other active sectors, such as electrical, and they are
taking a category-by-category approach.
Bryan Scott Larkin is director of marketing – retail,
consumer products and high-tech industries – at
GXS. Much of the data synchronisation
technology around the globe is powered by
GXS technology.
He believes that, “data
synchronisation had trouble getting
off the ground because there were
some pretty enormous
assumptions that just moving
from one form of electronics
communication – traditional EDI
or Edifax – to an XML-based
solution – the GDSN – format,
was going to solve all the
problems that were driving data
synchronisation, which were
data inaccuracies and conflicts
between buyers and sellers. It’s
been proven now that GDSN
doesn’t solve the data quality
problem. It does provide for a more
consistent and comprehensive means of
moving the data, but it still doesn’t address
data quality.”
GS1 UK has adopted a data quality solution
that GXS provides to help enforce and introduce data
VIEWPOINT
It’s not about the way
you move the data, it’s
about the way that you
manage the data.
BRYAN SCOTT LARKIN
GXS
NOVEMBER 2008 SUPPLY CHAIN STANDARD
www.supplychainstandard.com
quality into the equation. Many leading companies are
now looking at Product Information Management (PIM)
tool sets, both on the buy side and sell side, to help
manage data, often making this part of a master data
management programme.
PIM tools act as a combination of validation,
integration and workflow tools that allow you to assign
ownership of product attributes to one or more roles, or
people, within a business. It helps you set rules and
enforce those rules to ensure that you accumulate,
aggregate and validate your data. On the supply side you
would do this before you send information out to a
retailer, on the retail side you would use this to receive
data from a supplier validated against your own rules
and then also add whatever enhancements you might
have, such as your own retailer SKU.
Manage
“It’s not about the way you move the data, it’s about the
way that you manage the data,” says Larkin. His
company has a product information management
solution called GXS PDQ that helps create clean and
accurate data. “Data quality is essential to making data
synchronisation work”.
One might think that ensuring data quality is a highly
manual operation, but according to Larkin, there is a lot
that can be automated. “You can develop a whole set of
business rules and then submit data and be able to identify
the data that does not match, provide suggested fixes to
that data and identify those that can’t get fixes. You would
be surprised at how quickly attribute issues can be caught
and resolved [an example would be the use of ‘&’ or ‘and’]
if you have a good validation engine.” He adds, “manual
processes got us into these problems in the first place.”
It’s widely reported that the average keystroke error
rate is between two and five per cent. Manually entering
product data at various points across an enterprise just
compounds the number of errors. However, using a PIM
You can develop a whole set
of business rules and
then submit data and
be able to identify the
data that does not
match, provide
suggested fixes to
that data and
identify those that
can’t get fixes. You
would be surprised
at how quickly
attribute issues can
be caught and
resolved [an example
would be the use of ‘&’ or
‘and’] if you have a good
validation engine.
To view a Ceros magazine you need to have JavaScript enabled in your browser and have the flash player 8 or higher installed. To install the latest flash player, click the icon above, install the flash plugin and then click here to refresh this page.