| A global $1
billion business services company was facing
consistently declining income in a line of business
that accounted for more than 60 percent of U.S.
revenue. There was no obvious reason for the
drop-off that had begun with the new fiscal
year. Among the company's senior executives,
there was a range of opinions as to the reason
for the decline.
While the economy was an obvious scapegoat,
the sales force blamed the product group for
introducing new lower priced products that
were cannibalizing the revenue from high-priced
older products; the product group blamed the
sales force for not acquiring enough new contracts;
and both blamed the technology group for failing
to develop appealing product features already
adopted by the competition.
The company had the information needed to
find the answer for solving the problem. But
getting to the right data was not a straightforward
matter. As in most large organizations, a
significant amount of valuable information
resided in separate systems. Transaction level
details were stored in several discrete legacy
databases, one for telesales, others for web-based
sales, third-party sales, and the like, as
well as the accounting general ledger. Information
accumulated at the rate of 4.5GB of raw data
every month. Staff tried to aggregate this
transaction-level information in order to
explain the revenue drop-off. However, given
the complexity, traditional data collection
tools proved inadequate. The presence of irrelevant
data and exceptions further complicated the
effort. No clear viewpoint emerged; in fact,
the limited evidence was contradictory. |
