Sale Strategy

[Fairlight]

On Thu, Apr 05, 2018 at 02:32:53PM -0400, Nancy Palmquist via Filepro-list thus spoke:
> 
> But I have also seen the price tag for a runtime/dev system push
> away small users.  How can you make a little app that could help a
> small business when the cost of the runtime puts you out of
> competition.  How do you justify an upgrade from a system that works
> perfectly at 5.0 to a 5.8 version when the cost will be over $10,000
> and the customer will have to spend even more to get any development
> in place to use the new features if they even have such a need.  

These are two separate issues, though.

Cost of development is one issue.

Cost of licensing is an entirely separate one.  It's here that people get
blinders on and think long-term.  If it's $10k to purchase/upgrade the
licenses, let's look at the math behind what that -actually- means:

10000 / 12 = $833.333 per month, amortised over just one year.
833 / 30 = $27.766 per day.

I believe the cost of the licenses was around $300/seat?

10000 / 300 = 33.333 users for $10k.
27.766 / 33.333 = $0.832 = cost per user, per day, amortised -only- over
the first year.

We're looking at -only- the first year, here.  In reality, this gets
amortised over more like 1.5 decades or longer, based on the usual lifespan
of products written in filePro.  So let's look at, say...a decade, so you'd
divide by ten again, and it's $0.08 per user per day to have your users
using the application.

That is -ridiculously- low for a cost of doing business.  You pay more than
that in electricity, which itself is a necessity.  I -wish- my electric was
only $24.90/mo.  Try upwards of $150/mo in summer.

But actually -paying- for your software?  Oh, -hell- no!  Blasphemy,
heresy, and sacriliege!

If someone's margins are so small that they can't afford $27.76/day in
amortised expenditure, assuming the -worst- case scenario, then they're not
in a business where they can afford to use software, full stop.  That, or
they more likely simply -don't want to- afford it, which is imminently more
plausible.

Few things incite regret in me than doing a credit card integration
for a mere $5k, knowing that the business for which I'm doing it will
use it 24/7/365 in perpetuity, and I'm -not making a per-transaction
fee- off of each transaction.  They get to amortise the expense of the
integration and probably recoup it inside a month, maybe two (if even
that long), then it's pure normal profit as usual for the rest of time,
barring something like PCI specs requiring a TLS upgrade, as has recently
happened.  By-and-large, we write the software, it makes people money at
the heart of their enterprise on a non-stop basis, and we see -nothing-
after the initial work is done.  It's galling.  By all rights, we should be
making royalties like musicians do, -on top- of the initial outlay (which
musicians also get in the form of an advance, no matter what eventual sales
performance is like).

For someone to complain about that cost is utterly ridiculous, when you
actually sit down and look at the math.

Much as I begrudge Yet Another Subscription Fee, Adobe isn't actually
being horribly unreasonable in their Creative Cloud pricing.  I take
issue with the fact you can no longer purchase it outright at their usual
extortionistic prices, rather than being locked into leasing your tools for
as long as you want/need to use them.  However, it's less extortionistic
if you -need- the tools to do your job.  Then it's a necessity, not a
luxury.  At the point it's a necessity of doing business, such as being an
artist or illustrator on a professional level, it's more than reasonable as
a model.  It only hurts if you both can't justify it -and- don't have the
discretionary income.

In a -business- case, rather than for a hobby,it's -not- discretionary.
It's the cost of doing business, pure and simple.  Is the customer a
business, or are they not?

This is why I feel very little sympathy over licensing or development
costs.  They benefit perpetually, and we get a single go at a trough which
feeds them indefinitely.  Even if we try to make it really count, we're
-still- getting screwed.  That includes fP Tech, in this case.

People look at licensing costs entirely incorrectly, and only from their
perspective of not wanting to pay more, with very little regard for the
labour of those who developed the solutions upon which whole enterprises
rest for decades at a time.  There is very little regard for the long-term
success of said developers, no matter the industry they serve.

> Also the cost of actually doing the upgrade, since the system requires an
> entire index rebuild. In addition, in my experience filepro often breaks
> programming when you apply a new version to it.  This could disrupt the
> business while those issues are sorted out.

You can't make an omelette without breaking eggs.  You simply cannot get
from point A to point B without some disruption in some instances.  In
the case of upgrade vs migration, there's almost certainly going to be
disruption of some sort in -either- event, so it's practically a non-issue.

> The requirement they make for a system to move from Unix to Linux
> but the Linux has to be a current version so they are required to
> upgrade however many versions, often just buying a new filepro.  I
> understand the need for cash flow but when a customer does not see
> the benefit  for the upgrade it is a hard sell.  This pushes customers
> happy with filepro but unhappy with the OS to solving their
> programming with something other than filePro.

The customer can be -made- to see the benefit.  Offer the hypothetical:
"How much money will you lose per day if this software stopped working
entirely in ten minutes?"  I -guarantee- you that if the customer is not
overstating their needs, the number should -vastly- outweigh the potential
expenditure of a simple filePro upgrade, in most cases.

Likewise, When someone literally (the -real- literally, not the neologistic
figurative literally) tells me that they're losing 'tens of thousands' of
dollars -per hour- any time something stops working, I know damned well
that it's worth -far- more than anyone's emergency hourly rate to fix it.

If it would take ten minutes at a $300 1hr minimum for someone to fix on
an emergency hourly basis, and the customer will allegedly lose $10k/hr if
it's down, that means someone would be -perfectly- justified in switching
to a per-incident basis, charging them $5k/incident instead of $300/hr, and
the customer would -still- have saved over $3300 by not being down for the
rest of a single hour.  Examine the math behind the hypothetical, assuming
a single hour of $10k/hr loss if there's something down:

10000 / 60 = $166.666 cost per minute down

166 * 2 = $332 for hypothetical two minutes to fix at 1hr min / $300.

50 * 166.66 = 8333.00 amount saved for rest of hour spent up, if charged
hourly at that $300/hr 1hr minimum.

8333.00 - 5000 = 3333.00 amount -still- saved, even if you switched to a
per-indident charge of $5k instead of the hourly.

You could actually maximise it by doing it on a per-incident basis (like
Microsoft does), instead of hourly, and you are -still doing them a
favour-.  They won't see it, but they're not looking at the -actual- math
involved, apparently.  Somehow, they would find a way to begrudge you
bailing their arsses out and saving them $3300, simply because you didn't
do it as cheaply as they'd like.  This, despite the fact you just actually
-saved- them a substantial sum.

The same theory applies to software licensing sales, but multiplies out
to a degree bounded only by the lifetime of the product.  If that product
will last them two decades, that's a lot of money lost by not licensing at
a higher rate.  In mission-critical, enterprise-grade solutions, a higher
cost is -perfectly- justified.  Their business depends upon the software
working properly.

In short, software developers get the shaft, no matter how reasonable their
licensing and/or repair fees.

> I have customers that have been using filePro since 1982 and by 1993
> they had it working the way they wanted.  We did an upgrade for Y2K in
> 2000 and they have been working ever since with no issues, except
> small user errors.  How do you argue to a customer with that history
> that for a cost of upwards of $2000 they could get a new version
> with features they will never use?

Point out the fact that they're -still- making out like bandits, and they
have been for 36 years.  Have they been paying a lease on that software for
three and a half decades?  No?  Then they're making out like bandits.

Having gone through the math, I'm actually far more in favour of
subscription software and per-incident charges than I ever have been, from
the developer side of things.  From the consumer side of things, as someone
-not- making enterprise level income per hour, it still feels ridiculously
high.  However, when you have customers who demand enterprise grade levels
of service, and who -do- rely upon their software and infrastructure
to reliably deliver them enterprise levels of revenue, it is -totally-
realistic.  The numbers don't lie.

Now, is it something anyone would -willingly- incur?  In the case of a
failure, no.  But you'd still be saving them money in the per-incident
model, under circumstances as-outlined.  In the case of an upgrade, quite
possibly.  They're talking about migrating a whole system, and we all know
how -that- tends to pan out.  The filePro licensing costs of upgrading
are the -least- of their concerns, if they're even veering neer the
neighbourhood of -thinking- about migrating.

Why people don't see it as an investment in continued revenue and
efficiency is beyond me.

mark->
-- 
Audio panton, cogito singularis.