142Re: [webanalytics] Re: Any Info on Pilot Software?
- Jul 28, 2004For 10 Points: http://www.amazon.com/exec/obidos/tg/detail/-/B00000281T
Aldo Nova (I don't remember, but Google does!)
I also prefer to pay only once per license (per machine/cpu/data source/etc.,)
This makes the most sense to me, and if I want to add hardward because
the traffic is too much then so be it - another license.
If I want to upgrade to the latest version - bam! another upgrade license.
This is much better for me, the customer, because the cost is only
once and not strung over the lifetime of the company.
I experienced the same problem when looking for survey technology and
went with a software solution. It was a bit more than the projected
cost per year, per respondent would have been for a survey ASP.
But, it was only one lump payment and then it was done - since then we
have been using it more than we thought we would, and actually
probably saved money even after the first year.
But software is software is software - and not a service. It requires
management, resources, and fine tuning. I would imagine that the ASP
model does too, but not to the same extent and you get those handy
"free upgrades" baked right in to the analysis.
I hear that this causes some problems though when one weeks analysis
is one way, and then the next it is improved by the new technology.
Makes it kind of hard to do an apples to apples comparison (even if
the apples are rotten).
----- Original Message -----
From: wexler <wexler@...>
Date: Thu, 29 Jul 2004 03:37:22 -0000
Subject: [webanalytics] Re: Any Info on Pilot Software?
--- In firstname.lastname@example.org, Jeff Seacrist
> But it certainly raises an interesting question -- What does this
> believe is the most "fair" way to scale the pricing of analytics
Hmm. Am I allowed to broach the assumption that the idea of "scaling
the price" is at all fair, if its based on use? Again, for an ASP,
the approach reflects growing costs: As I grow in PVs, I naturally
suck more resources from the ASP serving my web beacons. But when the
software is on my systems, in my humble opinion, you have no right to
address scale, beyond pricing for capability. I have to pay for the
increased hardware to run the solution; I shouldn't have to pay MORE
for the permission to use your software at that level of data.
I understand from Jeff's note how he sees it as an attempt at
fairness: Sites with smaller data pay less, yet get access to the same
powerful tool "the big guys" get. But it comes off as a progressive
tax: Those who have more data continue to pay more, even though
everyone is using the same tool. And yes, one can group it out so
that intial steps are smaller and higher steps are larger... but no
matter how you spin it, it still _feels_ wrong when compared to other
software pricing models.
You can offer a junior version which caps out and won't report past a
certain count of PVs... and a medium version with has no cap but
cannot handle certain features. MS SQL Server has this approach with
MSDE and the various version so MS-SQL 2K. But (arguing with
myself), that does require more code management and versioning, which
is distracting to the engineering staff.
But when I buy an enterprise product, I will pay for the license, and
I can run it on a little site or a big one. You have your license fee
and annual renewal fee for support; whether I use it on large data or
small is my business.
I expect Word to be priced per-user (simultaneous or max count), but
not per word typed into it. Should web analysis software (or ETL or
OLAP) be any different?
I am reminded again of the Oracle Power-Unit, Sun processor pricing,
and IBM's chip-speed pricing, and I have to ask myself, "Why accept
this and not the web-analytic software argument? How is it
different?" I also have to point out that many survey tools charge
either per survey shown or survey completed. But note that Oracle
doesn't charge me base on the data I put into it: When I get the big
version, I pay my money and I can put terabytes in it... or 4 tables
with 10 rows each. As long as the machine stays the same, I pay the
same (large) amount. And only the top of the line installable tools
(that is, 2 or so) can get away with volume-based pricing beyond
product cost; the rest cannot. (But that may be a market issue; not
that they shouldn't but that they don't have the momentum yet, so
perhaps bad example.)
So, I dunno, if its the same bytes and machine instructions doing the
same work, and you feel that I should be charged more because I put
more data into it... that just feels like it defeats the purpose.
I guess I would be happier with a tool which is enterprise priced but
has no artificial caps on capability (and I count an arbitrary fee
based on volume use of the tool as an artificial cap).
Again, just one man's opinion. I know everyone has to make a living,
and I continue to admire the clever functionality each product has
brought forward, so I mention all this as a thought starter.
Am I totally out of my gourd, living in a fantasy? Or is this a
pricing model which others also believe is untenable? And for 10
points, can anyone name the MTV darlings in the early 80s who also
lived in a fantasy ("Life is just a fantasy, Can you live this fantasy
Web Metrics Discussion Group
Moderated by Eric T. Peterson
Author, Web Analytics Demystified
Yahoo! Groups Sponsor
Yahoo! Groups Links
To visit your group on the web, go to:
To unsubscribe from this group, send an email to:
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
- << Previous post in topic Next post in topic >>