Where O Where Do I Open
a New Consignment Store in Vancouver?
Problems
and Errors
For the most part my project went
smoothly. There were, of course, countless troubles with the
importation and manipulation of data, but these worries were easily
solved by Rob Fiedler and my fellow classmates. The more
important areas for discussion in this section are in data collection
and spatial analysis. In general, even the problems I had in
these areas I think could be solved by giving the project more time
(say, if I was being paid to do it instead of paying to do it).
Data Collection:
To begin with, the data
collection was not as accurate as I would have liked. I know for
a fact that there are a number of consignment stores that were not
included in my database of existing stores. To change this, though,
would have meant walking the streets where I knew missed stores were
and recording them. I used the Yellowpages and mytelus.com, and
included all the stores included in these databases. To improve
on this accuracy, to do the street-walking, would still not yield truly
accurate results because I would only be walking where I knew stores
existed. Ones I did not know about would still not be included.
As far as lease costs go, which was the most important factor in my
analysis, I only looked on one website on one day. This site did
give me valuable information, and points that had a fairly good
coverage of the city, but still not as good as I would have
liked. The biggest problem with this form of data collection was
that only those spaces currently up for lease were included.
Although this gives me an idea of where to open a store today, it may
be more useful to have information on where to open it anytime this
year. In that case I would want to get better average lease
values for all the commercial areas in Vancouver. I could then
use my analysis to refine my search for an available space once I
actually begin to look. The analysis I created is more of a snapshot of
where to look in Vancouver during November 2004.
Analysis:
My problems in my analysis were
only with how to deal with my existing store and demographic
information. I think that how I handled it was fine, visually assigning
values to neighbourhoods, but I would have preferred to utilized the
databases during this step. I also think it would have been beneficial
to analyze areas between neighbourhoods and DAs. DAs were so
small that ideal locations would have been so specific as to not be
beneficial in reality. On the opposite side, neighbourhoods offered
valuable information for large tracts, but was not quite specific
enough. An example is in the number of stores per
neighbourhood. Even though a neighbourhood can contain 7 stores
it is unlikely that I would shop at them all unless they were all
within 5 to 7 blocks of each other. It would have been very
beneficial to be able to group the stores together depending upon
whether they were on the same street, up side streets, within 5 blocks,
etc. With demographic information it is the same. If I were
to really open a store I'd like to know if there were a lot of women
within a 5 block radius, say, so I could assume many would shop on
their way to and from home.