The Expensive, Delicate Ship
In a year of loud noise and distraction at PeopleSoft, somehow, somebody was doing some work.
PeopleSoft Leadership
At the recent PeopleSoft Leadership Summit, I
was reminded of some lines from a poem by Auden.
And the expensive delicate ship
That must have seen something amazing
A boy [Icarus] falling out of the sky
Had somewhere to get to and sailed calmly on.
Even in the most dramatic
circumstances, says Auden, the people surrounding the events still have their own concerns,
still go to work every day. For the past year, PeopleSoft has been the center of some pretty dramatic events (at least for our industry). At the Summit, I was reminded that still, somebody was going to work every day.
The Total Ownership Experience
The big news for me at the Summit was the progress
that has been made on the Total Ownership Experience (TOE) program.
The TOE is a retrofit and repair project focused primarily
on PeopleSoft (Enterprise). (It was announced before the merger.)
The overall aim of TOE is to improve the usability and manageability
of the PeopleSoft (Enterprise) product.
Much is included under the TOE rubric; even one of
my really long
Short Takes wouldn't give you the full idea. But let me give you a few
examples of what's going on.
- Usability. A team
has been looking at a hundred or so core
processes and trying to improve them.
They've used
specially developed tools to track the keystrokes of actual
users, talk to them about their experience while they're working,
and determine what can be done to improve their work.
Sometimes, they redesign the screens; sometimes, they make
access to information or other tasks easier. They measure
such things as total time spent on the task, and PeopleSoft
marketing reports substantial (30-40%) reductions.
- Patch Management. Big application companies
routinely burden their customer base with
bundles of product fixes (called "patches") put out about once
every two weeks. A new PeopleSoft tool relieves the burden
somewhat. The tool checks through your current installation, lists
the patches that have not been installed, and allows the user to install
selected patches (or patch groups) the same way you install a
Norton AntiVirus update.
- Implementation Speedup. Newly added to the implementation
toolkit are setup wizards,
pre-loaded integrations with selected Oracle and SAP processes,
and improved Excel-based data loading. All should make implementations
go faster.
None of this sounds earthshaking (except maybe that 30-40% number), and
it isn't. Reducing keystroke
time, making tasks easier to complete, or improving
the information content of screens: good things, you might say,
but so what?
The so what depends on the breadth of the project. If you
improve a few screens or make life easier for a few low-level clerks, no one
will care. But if you make hundreds or even thousands of changes, the impact
ends up being substantial. The life of everybody who touches the product
is affected for the better.
This project apparently has that kind of breadth. According to
PeopleSoft marketing, $800 million is being spent on it; the spreadsheet
they use to list the improvements goes to hundreds of lines.
I try to do spot checks on big projects like this, and certainly I do have
some criticisms. Much of what I saw suffers from a problem endemic in
the industry; the people who did it chose to do what was easy, rather than what
was important.
Take the patch management tool, for instance.
It will be a help, no question. But the real burden imposed by these
patch policies lies in the fact that customers have to figure out
what the patches do, assess their impact, and then test the blazes out
of them. In relieving that burden, the help is limited.
Or, take that 30% decrease in task time. Good, of course.
But for many users, time actually spent on PeopleSoft tasks isn't a huge part of the
day. They're spending their time on answering phone calls, chasing down discrepancies,
etc. So a 30% improvement in task time doesn't translate into a 30% reduction in the workforce (though it might translate into less overtime at peak
periods).
That being said, I want to say that my overall reaction is still overwhelmingly
positive. It isn't just that an already good product is being made substantially
better. It is that PeopleSoft is actually providing leadership for the industry.
What other software company has embarked on a project that even approaches this in scale that is claimed for it. $800 million is
2/5 of the amount spent on JD Edwards. According to my rough calculations (15% of $2 billion
is $300 million) it is almost 3 years worth of
R&D for PeopleSoft Enterprise.
The conventional wisdom is that retrofit and repair
are never worthwhile. Why spend money improving something that you've
already sold? All the money you put into it is lost, because you can't sell any of the work you do.
If this wisdom is right, and I think it is, the decision to spend the money is essentially an innovative and courageous bet on the future direction of the industry.
Why do it, if you're getting zero direct return? Wel, you're betting on indirect return.
Improve the product, after all, and you
should improve customer references. Improve it, and you should
reduce the cost of support. Improve
it, and you should make customers
slightly more comfortable about buying more stuff from you. Improve it, and
you should reduce
the chances that existing installations will be replaced. You're betting essentially that in the future, these things will matter a lot, where in the past, they have mattered very little.
In a way, this project is Rick Bergquist's (or Craig Conway's)
reply to SAP's Netweaver project.
Both take a substantial amount of development resources (again, assuming
that the scale is roughly what I've been told it is). Both are emphatically
investments in the future of the product. But look at the difference.
SAP is saying that their current technology base needs to be
replaced if the product is going to keep up. Bergquist and Conway are saying
something completely different. "As a software company, all we really
need to do is improve the value proposition that our software presents,
and people will buy more of it."
Demand-Driven Manufacturing
Whenever you have major product announcements, the move from the sublime
to something else comes quickly.
The other big statement from PeopleSoft marketing was
that PeopleSoft is now a leader in "demand-driven
manufacturing."
If you are a member of that very small group of people
who have been around manufacturing and manufacturing
software for a long time, you will probably have a reaction
to that statement rougly akin to one's reaction to the
"Never needs sharpening" commercials on late night TV.
It's a shame, in a way, that PeopleSoft chose this
way to present what they're doing. Because, at least
in this fight, they are on the side of the angels.
Let me try to disentangle what's going on.
Most members of that same small group believe that there is
a "good" way of organizing your manufacturing effort and a "bad" way.
The good way has something (more or less) to do with the
ideas of Eli Goldrath, as expressed in his book, "The Goal." Mixed in with those ideas,
of course, are a lot of other terms and techniques, like "Kanban" and "flow," that
don't derive from Goldrath, but let's not be purists about it.
The ideas behind good manufacturing have been around for a long time, but so far, they've never really taken hold.
They are right (I believe), and they have had
many passionate and vocal adherents. But the American manufacturing community
just hasn't really bought into them. If I walk into a factory, I expect
to see one that is organized pretty much the way factories were organized
when my grandfather worked in one or one that at most uses Kanban
for some replenishment.
One reason for this is that adopting good manufacturing techniques
requires that you
reorganize your manufacturing processes.
Technology
is "Necessary, but Not Sufficient," to quote the title of another Goldrath book,
co-authored by an old manufacturing hand named Carol Ptak.
Now, Carol Ptak is at PeopleSoft. And what is evidently meant
by the confusing term, "demand-driven manufacturing," is that PeopleSoft
is now going to support, espouse, and provide the software for
"good"
manufacturing. ("Demand-driven" is confusing because most people
who do manufacturing the bad way also think
that their manufacturing is demand driven.)
Good. I'm glad when anyone
wants to lead customers down the right path. But this doesn't in any
way amount to leadership. There's some software
(the old Numetrix stuff, some demand management licensed from Demantra,
the old JCIT flow manufacturing software). But none of the software
they have is particularly innovative; there are many other companies that
have pretty good "flow" and Kanban products. And there's Ptak, who is
a leader. But a software company isn't even the right sort of entity
to provide leadership in this space. If you believe, Ptak herself,
what
you really need to is a manufacturing consulting organization.
Now, there are some companies that might want what PeopleSoft is
offering. There are companies today that have PeopleSoft
(Enterprise) HR and Financials and might prefer to use PeopleSoft (Enterprise)
as their manufacturing platform, if they could. A small subset of those companies
might even want to change their manufacturing organization, using new software
as the hammer. But even to do this, PeopleSoft
seems undergunned at this point. Carol
Ptak could send me down any old dark alley in any big city in America, and
I'd just go.
But the software itself is at best pieces and parts,
and the number
of people that have Ptak's stature and fervor is obviously limited.
The JD Edwards Acquisition
The JD Edwards customer base was welcomed to the
Summit, and many took advantage of the invitation. It is obviously
in PeopleSoft's interest to keep these customers, to get them to upgrade,
and, in some cases, to move World customers onto the OneWorld (Enterprise One) platform.
The customers I talked to said that PeopleSoft marketing has been
able to make PeopleSoft's interests
very clear to them. They were a little less certain about whether those
interests coincided with theirs.
This was a small sample.
And even in this small sample, there wasn't a scintilla
of hostility, just a feeling that perhaps
PeopleSoft was doing more telling than listening. The issues, also,
were various.
A World customer talked to me at length about the new upgrade
policy (according to him, it's, Upgrade To 8.11 (?) Soon!)
He doesn't have the personnel or budget
to manage an upgrade. So he's leaning today toward going off maintenance.
A construction industry customer (also World)
told me he doesn't want to move
to Enterprise One, but is in the minority. Most of the
construction industry customers think it's an
excellent idea. This customer understands the theoretical
benefits that would be conferred by the current Enterprise One
architecture and the promised Enterprise One enhancements,
but he's not sure when they're coming or whether they would be worth the disruption.
I don't follow the construction industry closely enough
to be able to answer his questions about whether those enhancements
are a) really coming and b) what was promised. But
I do have the same questions he did about other parts of the old
JD Edwards road map.
I have reviewed the many promises
made at last year's Quest conference and at the September analyst day (when the acquisition
was complete). It appears to me that development has fallen somewhat (but
not terribly) short of what was promised.
But a person who is intimately familiar
with the JD Edwards development organization says
that the road map is pretty much right on schedule, despite some
cuts. "The big difference
now is that we work PeopleSoft hours. The old JD Edwards was a life-style
company; everybody left at five. Now the parking lots are full on Saturday. We've
also saved a lot of money by outsourcing product testing and other development
infrastructure to India."
If so, this is good news for a customer base that still appears to be wary
and a good opportunity for PeopleSoft as the stream of usable new products
comes into the marketplace. As I say, the area of the product road map is not
one where PeopleSoft marketing is given to strong, clear, reliable statements,
so it is hard to assess what the impact will be.
The Big Kahuna
Was it coincidence or the beginnings of a trend that
several name-brand customers I talked to were looking at whether
it makes sense to replace their
PeopleSoft (Enterprise) modules with the SAP "equivalents."
For these customers, SAP is the primary, corporate platform,
but PeopleSoft is the incumbent provider of HR across the enterprise
or (in one case) of HR/Financials for
a division.
I think this kind of replacement rarely makes sense,
and in my quite detailed conversations with these customers, that feeling
was confirmed. But the very fact that it came up shows why the Total
Ownership Experience is so important. SAP is now capable of challenging
PeopleSoft in these customer sites, so PeopleSoft better
be providing some clear differentiation.
These days, SAP leads with Netweaver in any differentiation
discussion. So I was curious about whether PeopleSoft had a riposte.
I think the potential riposte is very strong (something along the lines of,
"The best product is the one that works the best.") but I think PeopleSoft marketing
may need to work on their articulation. Ram Gupta
talked about TOE, but his rapid, forced recitation of statistics didn't convey
anything meaningful to me.
Rick Bergquist also addressed the issue of Netweaver quite directly.
("His
talk could have been called, "Netweaver, what
Netweaver?") He said that PeopleSoft also has an enterprise services
architecture and that all the key components of Netweaver (e.g.,
integration capabilities, portal, process management, etc.) are already
provided by Peoplesoft. He's right, but again, I'm not sure he seized on
a compelling formulation.
The problem is that SAP has managed to persuade people that
eventually, their toolset (Netweaver) will be the foundation
for an enterprise information hub, whose heart is the
SAP command interpreter. Even if PeopleSoft has all the components
and they work as well as SAP's, what PeopleSoft is offering is not
equivalent, because there's no hub being offered. To be compelling,
they need to argue that there's no need for a hub or else that the hub technology
(if needed) will be provided in the open market. They also need to point out
that the technology development being done at SAP could have a cost for customers.
Money being put into Netweaver is not being put into making the applications more
effective, a point that PeopleSoft marketing ought to be making.
"Saying Doesn't Make It So"
You may have noticed that throughout this piece,
what PeopleSoft marketing
says about what they're doing is as confusing as it is helpful.
When PeopleSoft made strong, clear statements, they seemed unlikely or implausible,
and when strong, clear statements were needed, PeopleSoft was fuzzy.
Software marketers are like this, I admit. Like politicians, they like to spin things.
But it seems to me that at some point, software people, like politicians, need to lead as well as spin.
If claiming that you spend
$800 million on the TOE is just spin (I'm not sure, but PeopleSoft representatives backed off the number as soon as I questioned it), what good does it do you? You just leave your customers and investors wondering about how you are spending your development dollars.
Or take "demand-driven manufacturing." As spin, it's not very good, because it's confusing and implausible, even if it's on the side of the angels.
Isn't it time, I wondered after the Summit, for a little straight talk?
Oddly enough, we did get some good straight talk at the Summit from some out-of-work politicians (Bob Dole and Al Gore). And we also got some pretty direct criticism of other software vendors for not talking straight. In his keynote, Craig Conway criticized Henning Kagermann for claiming that his
software is flexible. "Saying doesn't make it so," he said. Don't, in other words,
try to put something over on us when everyone knows
it's just not true.
Sounds like good advice.
To see other recent Short
Takes, click here for a listing.
|
