I am looking for a tool or just a search method to accomplish the following:
1. I have a URL address of a company, say www.vocalocity.com
2. I want to get a list of competitors in the same market as www.vocalocity.com
3. I want to do this with as little work as possible. This means I do
NOT want to have to go to the company's website, find keywords for their
market, go to a search engine, enter the keywords, then sift through
all the search results finding real competitors from all the other
stuff.
Do you have any ideas or approaches that would help me here? If there
is a tool that might help, I would be willing to buy something for
under a $200.
I think Hoovers has a subscription service that provides competitors
in an industry, however, looking at it, I think they only have in
their database the largest and most well-know companies. I will be
analyzing small companies also. |
Request for Question Clarification by
webadept-ga
on
12 Feb 2004 07:13 PST
Just checking around on this, with a few quick searches, I'm not
finding anything which does this off the shelf. Even for more than
$200.00. As you say, the Hoovers focuses on the top levels of
buisness, mainly because they are the easiest to keep track of, and
those businesses report to Hoovers, not the other way around. I know
this, because I use to send the reports.
I believe I could create a Perl program which could do this to the
level of my needs for accuracy, but not sure about the level of yours
;-) or what format you would need the responces in? Just names of
companies? Just websites? Scraping out addresses and phone numbers is
not something I'm willing to do for less than 200, but another
researcher here might.
How many queries would you want to run a day, week, month? hour? This
is important, because many of the more accurate resources for these
types of searches, limit the number of queries.
webadept-ga
|
Clarification of Question by
mags311-ga
on
12 Feb 2004 16:37 PST
For a custom tool to be effective for us, it would need to return the
company name and web URL. But it needs to be highly accurate, it can't
have junk mixed in with the results. The number of queries per day
would only be around 10 to 50.
I would be skeptical of a custom tool providing high accuracy. But if
you have an idea on how you could build one that is, I would like to
hear it. I might be willing to pay up much more than $200 for a tool
like that.
|
Request for Question Clarification by
webadept-ga
on
13 Feb 2004 07:43 PST
Hi,
Sorry for not getting back to you sooner on this.. Can I? Probably,
but there are also several other researchers in here, which can do so
as well, and with next few days, being what they are for me, if one of
them steps up, and helps you, great.
The Google API, with a few other reference sites for cross checking
and evaluation of the results is a great tool for this. The API allows
1000 queries a day (last I heard), and I've used it in the past for
similar tools.
://www.google.com/apis/
So we are well within your needs there. With that tool the program can
gather in a seed data file, and begin comparisons on what it found.
There are many ways of doing this, which I won't go into here (gory
details), but as I said, I've done this kind of thing before, and so
have many others.
A tool developed in this manner would work best as a web page, for
you, but would run faster as a command line script. As a command line
script you would need Perl installed on the machine. (very easy
install, and very easy to write instructions for). From that point,
you would type something like this at the command line.
C:\> perl findcomps.pl www.vocalocity.com
From there it would run, probably something like 20 to 30 minutes,
perhaps a bit longer. You could put in your 20 to 30 websites in a
text file, and do something like
C:\> perl findcomps.pl thisfile.txt
and it would go through them all, which would be better for you,
because you could just start the program and let it run at night. The
programs I make generally output a excel file or a html document, so
you have hyper links to follow when it is done.
For me, 200 would be a bit much, 100 would do. But as I said, my days
are pretty full right now and will be into next week. If another
researcher doesn't get in here and answer this question with an "off
the shelf" answer, you might consider asking a new question, with the
bid around that range, describing the program you want, and refer to
this question.
I will check back here to see how you are fairing and I wish you the
best of luck. It does sound like a fun project, and I really wish I
could jump into it.
thanks,
webadept-ga
|
Request for Question Clarification by
webadept-ga
on
13 Feb 2004 07:49 PST
By the way, have you already trid the "related" function on Google?
such as searching for
related:www.perlmonks.org
on the websites you are interested in? Check that out to see if this
would do the trick for you. It generally works well, in my experience.
webadept-ga
|
Request for Question Clarification by
webadept-ga
on
13 Feb 2004 07:57 PST
Well, while I'm at it, check dmoz for a list as well.
On Dmoz and searched for : Voice Web Solutions : and got this page
http://search.dmoz.org/cgi-bin/search?search=Voice+Web+Solutions
using a "related" search on Google
related:www.vocalocity.net
gave me this listing
://www.google.com.ni/search?sourceid=mozclient&ie=utf-8&oe=utf-8&q=related%3Awww.vocalocity.net
Check those two, after all, why reinvent the google? :-)
webadept-ga
|
