| Expandable Software, Inc.
was founded in 1983 by three friends and serial
entrepreneurs - Marshall Guile, Jerry Lass
and David Kearney - who shared the idea that
small and mid-sized manufacturers could run
their businesses on a more cost-effective
software platform using networked PCs. |
| Manufacturing customers
- and their minicomputer suppliers - initially
scoffed at the notion, because in 1983 minicomputers
ruled the roost, and personal computers with
their 5.25 inch floppies, monochrome monitors
and inadequate memory seemed like laughable
toys. But Guile and friends had already built
NCA Corp., whose groundbreaking MRP product
ran on minicomputers, and they recognized
the fact that DEC and VAX machines were exceedingly
expensive and not very expandable. After taking
NCA public in 1981, they decided to apply
their MRP expertise to the PC platform and
Expandable Software, Inc. was born. |
| Expandable grew quietly
but steadily through the '80s, continuing
to enhance its powerful yet flexible manufacturing
software as the PC platform became more robust.
In the mid-'90s, the company redeveloped its
ERP suite for Windows and SQL Server, adding
new modules to address specific industry verticals
such as medical devices, pharmaceuticals,
telecommunications, and high-tech electronics.
|
| Through it all, Expandable's
senior management team has remained true to
a few fundamental principles. Their product
must be reliable, easy to learn and use, and
customers should be able to complete implementation
within four to six weeks. The software should
be robust and expandable so it can meet the
needs of emerging companies as they grow to
become large enterprises. And above all, Expandable
must deliver the best customer service and
support in the software industry. As anyone
who has talked with the company's fiercely
loyal customers can tell you, Expandable lives
up to these guiding principles every business
day. |
| Marshall Guile, Expandable's
CEO, and Jerry Lass, Vice President of Sales
and Marketing, were interviewed by Patrick
Porter, former editor of Software Magazine. |
| |
| Porter:
Expandable is your second ERP company. You
two and David Kearney were at NCA together
in the 1970s and '80s. |
| Lass: That's
right. I co-founded NCA Corporation in 1969.
We took the company public in '81. I left
the company in '82. A year later Marshall
and Dave left as well. That's when we got
together to form Expandable Software. |
| Porter:
Your first ERP product was text-based. How
did you develop the current suite? |
| Guile:
In '93 we realized that we had to come up
with a graphical suite that employed a relational
database. So, like everybody else at that
time, we started looking for object-oriented
development tools. We chose Delphi by Borland
because PowerBuilder was too new. We thought
it would take us two years to develop the
new suite. It ended up taking five. Developing
the software was a lot more complicated than
the old character-based systems. A lot of
companies were never able to make the change.
Ask Computer, which had acquired NCA in 1987,
couldn't make the change. That's why they
were sold to Computer Associates. There were
a lot of disaster stories. A lot of companies
did it piecemeal. They tacked on a front-end
but never took advantage of the relational
database. They couldn't compete when products
like Expandable came out. Another reason it
took us so long is because we changed the
original product so it would work with SQL
Server. We didn't want to leave our old customers
stranded, so we developed a migration path
which allowed them to use the old product
while they gradually adopted the new one.
Some of our customers ran mixed-mode for a
year or two. |
| Porter:
What are the advantages of Expandable over
ERP suites that are stitched together through
acquisition? |
| Guile:
We have consistency across the entire suite.
The functionality works the same in manufacturing
as it does in finance and sales. So everybody
uses it in the same way. The tables that users
interface with are different, but the way
you expose, input, and edit data is the same.
So if you know one side of the fence and move
to the other - if you are in manufacturing
and get transferred to sales - you will quickly
feel comfortable in working with the software.
ERP suites that are stitched together from
different companies don't have that level
of consistency. People seem to think that
finance goes and does their thing while manufacturing
and engineering are off doing theirs. But
there are a lot of cross relationships. Where
does the financial planner begin and end?
Where does the buyer begin and end? They all
want manufacturing, sales and financial data.
Consistency in look, feel and functionality
makes everyone more productive. |
| Porter:
Why do so many ERP implementations run into
trouble? |
| Guile:
If you are talking about Oracle, SAP and Peoplesoft,
their systems have to be highly tuned to the
customer. Consultants have to go in and set
parameters and condition the system to function
the way the customer wants. It's not like
putting in a standard package. They are selling
consulting. It can easily become a never-ending
implementation. |
| Porter:
Do your installs ever go awry? Or are they
done fairly crisply? |
| Guile:
In our case, the customer signs up for a certain
number of support hours which are dedicated
to installation, onsite training, and maybe
some data conversion. We have a pretty good
handle on how much time it takes to install
and train, so we can be fairly realistic with
them about the front-end cost. Their other
costs - hardware or database if they are not
on SQL Server - are all known items. As long
as their network has the right kind of computers
on it, they are in pretty good shape. There
are not a lot of line items in our implementation
budget. |
| Lass:
Customer support really has the implementation
process down. The typical implementation is
finished within four to eight weeks. |
| Porter:
That's the average implementation? |
| Lass:
Yes. From the time we get the order, the customer
goes live an average of 45 days later. We
get the order, customer support implements
it for a month, and then in the following
month the customer cuts over at the end of
the month and they are live. We quote 120
to 200 hours, depending on the amount of data
conversion and training. That's all they need
to go live. |
| Porter:
Does your internal staff go onsite with the
customer? |
| Lass:
Yes, our internal staff visits the customer.
They get the software loaded, train the people,
convert the legacy data if necessary, and
basically handle all the details needed to
go live, even helping the customer tie down
the month-end against the chart of accounts. |
| Porter:
The implementation comes out of your customer
support group? |
| Guile:
The implementation and training is all out
of customer support. The same people that
install the software are also on the help
desk. They rotate around. |
| Porter:
Isn't that unique? |
| Lass:
It's certainly unique in that the company
that manufactures the software does the installs.
We're usually competing against third-party
resellers who do the installs and training.
That's how they make their money. |
| Guile:
If the VAR customized anything, the help desk
wouldn't even know about it. At Expandable
we know about it. That's why our support people
can understand and resolve issues so quickly. |
| Porter:
These are very important distinctions. Oracle
and SAP sell to small and medium-sized business
through VARs because the cost of sale would
be too high for them to sell direct. |
| Lass:
Yes, that's their model. Unfortunately, the
channel partner stands between them and their
customer. |
| Porter:
But if the smaller ERP vendors are also using
the indirect channel, then even they are losing
touch with their customers. It's ironic because
close customer relationships should be their
biggest advantage over SAP, Oracle, and Microsoft.
Any enterprise software developer that uses
an indirect channel will tend to lose touch
with their customers. The developer will lose
sight of the implementation and user experience.
It's not like selling consumer software. |
| Guile:
If I were a customer and wanted a change in
the software, the company doing it would be
the VAR that sold it to me, not the software
development company. So I am totally dependent
upon my local reseller if I need something
changed. And if it's too big for them or buried
in the architecture of the system, it is not
going to get done. |
| Porter:
Maybe if you complained loudly enough it would
get escalated up to the vendor. |
| Lass:
The big vendor is only going to be interested
if you are a large company. They are not going
to be interested in a small customer. That's
why they have pushed it down to the channel.
They don't want to deal with those problems. |
| Porter:
I think any developer that is not involved
in the implementation and on-going support
will lose touch with its customers, and as
a result is going to end up with low customer
satisfaction ratings. It's inherent in the
process. If you are not close to your customers,
then you are not going to understand what
they value. |
| Guile:
Why do so many ERP vendors still not understand
what their customers want? I think we have
answered that question. Because they don't
touch their customers. |
| Porter:
I asked some of your customers what was important
to them. They said software that is reliable,
efficient, cost-effective, easy to use, with
great customer support from the vendor - all
the things respondents to the Yankee Group's
ERP Customer Satisfaction Survey said the
large ERP vendors were not delivering. |
| Lass:
When we are selling against competitors, we
try to get the customer to check references.
If we can get them to check references, we
are more than likely going to get the sale.
Even some of the VARs that are local to that
customer cannot find 10 reference accounts. |
| Guile:
I would clarify that as saying "support"
references, because a reference on the ERP
vendor is not about the third-party rep that's
selling the software. The customer wants a
reference on the VAR, because that is the
company that will implement and customize
the software, and provide at least some of
the ongoing support. The tier one vendor won't
have anything to do with it, except for providing
help desk. In our case, when customers contact
our references, they hear about the entire
customer experience, from the developer through
customer support. Our references can talk
about all the issues. |
| Lass:
They can talk about how the software works,
does Expandable provide good support, how
effective is their training, and do they have
customers with problems like mine? |
| Porter:
How do you decide which features to add to
Expandable? |
| Guile:
We have a program change request (PCR) system
which enables customers to log changes they
would like to see in the software. We look
at these things constantly when working on
a new release. If we see a trend that we can
address with new functionality, it will end
up in a future release. |
| Porter:
Some software companies are so sales-driven
that their sales people are always saying,
"if we just had one more feature, we
could get that sale." So they continually
add features, even though most customers don't
need them. |
| Guile:
It's the old 80-20 rule - 80 percent of the
requests come from 20 percent of the installed
base. If it's a feature that benefits a lot
of the customers who might have the same problem,
then it will be considered for a standard
upgrade, which won't cost anybody anything.
But if it's something unique to that customer,
and they are willing to pay for it, we can
put it in the software and support it just
like anything else in the standard product.
In effect, it will become part of the standard
product and downstream other people will get
the benefit. They may not use it, but it's
there. If it doesn't fall into that vein,
then we turn it over to professional services.
That is the arm of customer support that does
custom projects. |
| Porter:
Tell me about them. |
| Guile:
Professional services will, for a fee, write
a program that is unique to a particular customer.
They developed an electronic data interchange
solution for one customer. They developed
an interface for a customer who needed to
connect sales orders with the credit card
clearing houses. That actually became a standard
professional services product that has an
additional charge associated with it. It's
available on a license basis. |
| One
fairly common customization is the database
trigger. Some third-party applications have
alerts and use database triggers. A customer
may want a particular field updated in a certain
way in a table whenever a record is saved.
You can do that with a database trigger. Professional
services will write that trigger, which becomes
part of the customer's system, and then it
is executed whenever that transaction occurs.
It's like doing a custom program, only it's
a small database script. |
| Porter:
Is your approach to customization unique? |
| Lass:
That's how the resellers in tier three make
their money. They do custom work on a base
system. They get a fee for making the sale,
but they make most of their money doing the
installs and writing custom software for their
customers. |
| Porter:
What if the VAR goes under? It can be extremely
difficult to understand and support custom
code that was written by someone else, especially
if it isn't well-documented. |
| Guile:
That has been a big-time problem in the last
three years. A lot of big VARs went right
out of business. |
| Porter:
Because of the downturn? |
| Guile:
That's right. People weren't buying. But
the VARs didn't have the luxury, which a company
like Expandable does, of having that ongoing
maintenance revenue coming in. |
| Porter:
The vendors can survive in a harsh environment
a lot better than the VARs can. |
| Guile:
That's right. Or if their particular geographical
area has a downturn, then the customer is
stuck. Most VARs don't have much staying power. |
| Porter:
That's a hot button issue. If the VAR goes
under, who is going to step in? |
| Lass:
And what if the vendor gets upset with
the VAR and cuts them loose? Several vendors
keep changing their business models from direct
to VAR and back again. The VARs and their
customers are left holding the bag. |
| Porter:
I'm interested in the concept of "overshooting,"
wherein a software developer creates a package
that is far beyond the needs of the customer.
Expandable sells its ERP software in components,
so customers pay only for what they need.
Can somebody buy your financial application
but not your MRP? |
| Guile:
We have a core system that includes the manufacturing,
financial and sales modules. The modules have
controls in them, so if a customer does not
want to run our accounts payable, we can set
the system so purchasing ignores accounts
payable. But we don't see much of that. We
sell the base set of modules and then we have
optional modules - things like shop routing,
configuration control, purchase requisition,
consignment inventory, foreign currency, lot
control, engineering change management, a
multiple discounts module that is primarily
for people selling to retailers, and so on. |
| Porter:
So customers buy the core product and add
components as needed, paying only for the
ones they use. |
| Lass:
That's right. Another thing we do is add in
third-party products. If somebody wants a
sophisticated product data management application
like Agile, Arena or Omnify, they're available. |
| Porter:
Who are some of your third-party application
providers? |
| Guile:
Business Objects, which acquired Crystal Decisions
last year and now owns Crystal Reports, the
de facto standard reporting tool; FRx Software,
a Microsoft company that writes financial
analysis and reporting software; SalesLogix,
a division of Best Software that develops
CRM software for mid-market and small business;
Q4bis, which has some good decision support
and business intelligence tools, and others. |
| Porter:
That gives your customers plenty of choices.
Let's talk about your support organization.
All of your customers rave about Expandable's
customer support group. How do you do it? |
| Lass:
Part of it is philosophical. We really believe
in it. We have always had the attitude that
software isn't any good unless the customers
are running their businesses on it and are
happy with it. We support every one of our
customers directly. They and our support group
can draw on every person in the company for
help. Most of our support people are quite
senior and have very broad experience. We
have always believed it was best to rotate
our support people through installation, help
desk, and so on. We don't have any junior
people handling the first level of calls.
Since our help desk people are so experienced,
they know how to ask the right questions and
get to the bottom of things quickly. We have
developed very sound processes and technology
in place to help our support group. We have
a customer support tracking mechanism in which
everything is logged. When a support person
picks up a call the customer information is
immediately pulled from the database, so even
if they didn't take the last call they can
pick right up where somebody else left off.
|
| Porter:
Where do you find your support people? Do
you provide them special training? |
| Lass:
A lot of our support people have a manufacturing
background. They are not software people,
generally. They may be strong in finance or
production. They have worked in manufacturing
companies. Some of them have come from our
customers. Generally, they have experience
working in manufacturing facilities on the
finance, production or planning end. So they
have domain expertise. When they first come
in and start taking calls, if they don't have
the answer, they have all the other people
around to help them. And they all go through
special training before they start. |
| Porter:
Why do so many venture-backed and medical
technology companies like your software? |
| Lass:
Venture-backed startups begin with very small
staffs. They expect to grow rapidly, but they
certainly don't want to spend scarce resources
on a big internal IT support group to run
their software. So they put a high premium
on reliable, easy to install and use software.
We are fortunate to be based here where a
lot of venture-backed companies start. And
word of mouth travels pretty fast, so people
hear about us through their colleagues at
other startups who understand the importance
of headache-free software. Some startups are
so small they don't have enough expertise
to get their company set up properly. A lot
of them rely on us to help them get their
chart of accounts set up or solve some of
their business process issues to get them
going. It's not just buying the software. |
| Porter:
They will remember that kind of support. |
| Lass:
That's right. Entrepreneurs may work at a
startup for three or four years, then they
leave and go to another startup. If they were
happy with the software at their prior company,
that's the software they're going to want
at their next company. |
| Porter:
There are so many ERP software vendors selling
to small and middle market companies today.
How is Expandable different? Why should a
company run its business on your software? |
| Guile:
We designed and developed Expandable specifically
to meet the needs of small and mid-sized growth
companies. Our customers want fully-functional
ERP software that is highly reliable, easy
to learn and use, quick to install, and yet
can expand to meet their increasingly sophisticated
needs as they grow. We meet those needs. Above
all, our customers want superb, no-excuses
customer support. And we give them that. In
fact, if you ask them, I believe they will
tell you that we deliver the best customer
support in the software industry, and maybe
in any industry. |
Integrated Solutions
