Slashdot Mirror
Univ. of Wisconsin's 30-Year-Old Payroll System Needs a $40 Million Fix
jaroslav writes "The University of Wisconsin is attempting to update a payroll system they have had in place since 1975, but spent $28.4 million in a 2004 attempt with no results, and now is experiencing new overruns in cost and time after 'not hav[ing] the full picture of how complex this project would be.' The current estimate of the redesign is $12 million and years of further work on top of the money already spent."
I would totally sign up to do this job.
I just want to say how glad I am my tuition's going to a good cause.
Who am I kidding, right?
Their payroll system doesn't need a 40-million-dollar fix. That's just what they've ended up spending on it (hypothetically, once the $12 MM hot cash injection fixes all the problems).
The University should just scrap the system and go with a commercial payroll vendor. Bigger organizations have done the same, and there's no shame in it.
$40 MM is insane. That's over four years of tuition for 4500 students at UW-Madison.
No use throwing good money after bad.
"Trolls they were, but filled with the evil will of their master: a fell race..." -- J.R.R. Tolkien on Olog-hai
I'm sorry, but what a heap of crap.
It's a payroll system. Yeah, it's a biggee, and yeah, it's got a lot of old information in it most probably. It's written in an old language (Oh no! The end of the world! Soon we might not be able to understand our systems! Hold on... we just had three attempts and replacing it with something new and FAILED because we didn't know half the stuff it was running). But you're not telling me that MILLIONS of dollars and YEARS of work by supposedly professional IT companies isn't enough to get ANYTHING working well enough to say "We don't need to worry about that part any more". You can get an OS written for that sort of money, or kit out an entire borough of schools with an integrated network.
What's *more* disgusting is that by the looks of it, the IT people at the University are probably barely getting a look in - it's being project-managed by external companies. Come on, stop faffing about; seriously, this is just stupid. Get your *existing* IT team, hire a bunch of programmers directly (hey, you're a University... I wonder where you can get a crapload of cheap, intellectual labour nearby, trained in the art of programming properly and designing the systems from the start, supervised and educated by people who have spent years using their technical, professional and theoretical expertise in the subject?) and just write the damn thing from the ground up. It wouldn't cost anywhere near as much money/time as you have wasted on a single company out of those that tried to sell you crap. Oh, and you can make it do what YOU want any time and you'll have the programmer's hanging around for the next few years with an incentive to keep the system running properly ("What grade did I give you for that paper on your design of the new payroll system? I've revised it, it just crashed.").
If it's THAT damn big, you want to start breaking the thing up into pieces, anyway. Anything that you can't find out all that it does in that many YEARS, you really want to be breaking into smaller and smaller parts and replicating them one at a time. Don't pretend that you're the only place on Earth that has that amount of employees, that amount of computer data, and require mordernisation.
Get rid of the project managing companies, get rid of the "slice-off-50%-for-myself" companies, get rid of the stupid contracts that REWARD failure, and give the project to people who will give you a system that will not only last for ever but be documented and updated and revised and bug-fixed and converted for ever and a day.
This is a statewide system that needs to be deployed on all 26 UW campuses, administration and UW-Extension (which has an office in each of Wisconsin's 72 counties). It handles all types of employees from student LTEs to professors to staff to administration, all of their benefits through the state retirement fund and the state employees healthcare plan (which itself is fairly complex). It has to deal with union and non-union employees and their different pay structures, special deals for certain faculty, etc. It's a complex system that is specific to the State of Wisconsin, so no, there is no off the shelf solution.
On top of all that, much of the cost is in deployment and training of all the people who have to use the thing.
Do you even know anything about perl? -- AC Replying to Tom Christiansen post.
The real problem here is that best of breed software developers have too many great opportunities that are more inline with their passions to work on this backwater payroll system. This leaves the unmotivated drones managed by Dilbertesque managers to run with this ball.
If it is a typical complex, highly-customized business system, it will have:
So, I'd say your Phase 1 above is a vast under-estimation. And the idea that you can farm it out to an external organization, especially one without close personal contact with users, is pure fantasy.
At the end of your process you'll have a system that does maybe 20% of what the users actually need, and 50% of which is stuff no one needs any longer.
We see a lot of stories about this kind of thing on Slashdot. Often it is a politician showing that he or she is completely ignorant of technical issues, but wants his or her foolish opinions to be respected.
Maybe it wouldn't be sensible to attend a university that has such technically backward management.
The world will be a better place when all the managers retire who were raised without computers.
"I'll bugfix this thing with badgers gnawing on both my arms for that kind of pay."
That's nothing! I'll get the badgers to do the coding.
I've been involved in a few of these types of projects (unfortunately), and believe it or not, the money goes quickly. So does the time. It's not just coding -- that's actually a very small part of the money. It would take some time to burn through $40mm, but you'd be amazed how quickly these project eat up cash. I certainly was when I first got involved.
Here are some things to consider:
testing the new processes, and getting buy-in and approval on all that from all the stakeholders costs? You know there will always be 3 to 5 revision and feedback cycles for everything. That's an easy 6 to 18 months of work for a team of six to eight people probably.
(Ugh, thank God I'm out of that ERP systems business these days!)
Yes, a fair amount of the money is probably wasted. But these projects do cost big bucks. This isn't hacking up a new blogging tool from open source toolkits. I'm not saying it's right, or well managed (it almost certainly isn't), but to say "dude, I could hack up a payroll system in a couple of months, pay me the money!" just shows that while you may know how to sling code, you don't have a clue about delivering solutions to business problems.
I'd simply mod you up if I could, but I can't so, I'll comment instead.
Speaking from extensive experience in data integration and migration from legacy (no, I really mean ancient) systems, this really is just a simplified version of what really happens in successful projects of this scope. Having also seen the nightmare scenario that UW is going through, I can guarantee that the failure lies in a lack of project management. With a budget that large, it didn't even require good project management. All they needed to do is actually have documented specs. Something as simple as here's a list of everything our current system does that we need to keep, these are the additional features we want to add, and here's the process we have to use to ensure data integrity. A Post-It Note even?
I love it when you talk dirty like that! Gimme some more, and say it in a hoarse whisper!
Maybe...
just maybe...
we can call it "The Looney-versity of Wisconsin".
Would be appropriate.
There's nothing wrong with the current payroll system other than it's old and runs on old hardware. The guys who wrote it 30+ years ago did a pretty good job.
The problem is, those guys are long retired, and some are dead. The ones who are still living have some hard feelings. They got treated like crap and were told to give up their jobs to youngsters whose sole knowledge of COBOL was a CS professor saying how awful it was. Consequently, there hasn't been much in the way of maintenance or knowledge transfer; the young'uns simply weren't interested.
They brought an old guy in to deal with Y2K issues. They agreed to pay him well, but then got chintzy when it turned out that there really wasn't much that he needed to do. They eventually did pay him, but kicked him to the curb again afterwards.
Since none of the young'uns understand the system, and the old guy refuses to deal with them any more, they have no choice but to replace it entirely. The problem is, nobody really knows what went into the system except for the old guy, who has the irritating habit of wanting to be paid to have his knowledge tapped.
COBOL is not that horrible, except in the minds of the ignorant. If you could do BASIC or FORTRAN, you could do COBOL. The bulk of a COBOL program isn't code at all, but instead is structure and format definitions ("data division"). Don't expect to have recursion or local variables (those are all new-fangled extensions) or object-oriented semantics. Be grateful that the original self-modifying feature of COBOL got removed. Then just break it down. Each procedure is labeled, and unless the programmer was an idiot the variable names have some relationship to what they mean.
The only real PITA for COBOL is learning all the reserved words (there's a few hundred of them) and their semantics. Other than that, it's just drudgery.
Fire everyone, buy quickbooks is not an appropriate answer then?
Not to mention $40 million / 60k employees is $666 per employee - there's your problem. Its the payroll system of the Apocalypse (integer math only need apply).
Stanford had a very expensive conversion to PeopleSoft a few years ago. Stanford had a huge collection of in-house systems from the 1970s and 1980s, running on either DEC PDP-10 machines or IBM mainframes. They've finally phased out all the PDP-10 based stuff at Stanford proper, although SLAC is still running some PDP-10 code.
Who were fired?
Actually the University of Wisconsin-Madison (the flagship UW school) gets so little money from the state these days that it is within their rights to rename the school if they wanted. If I recall correctly, UW-Madison gets 10% of their budget funded from the state.
OMFG, you've figured it out! All of these years we've been supporting all of these complex systems, and all we had to do all of this time is avoid the complexities! You're a genius!
So, Kreskin, what do you do when one of the unions that represent a good chunk of your employees brings you to Federal court and wins a judgement requiring you to give workers employed between June 5, 1989 and December 31, 1994 who were on maternity leave a pension credit and healthcare refund equal to 8% of their average pension contribution during that period, paid in 104 bi-weekly portions?
Stuff like that happens all of the time. What are you going to do? Go to jail for contempt of court?
Conformity is the jailer of freedom and enemy of growth. -JFK
I've worked for government, small businesses and Fortune 50 corporations. In my experience, government is just as screwed up as a big corporation. The only difference is that most big corporations purge some people every year, and government tends to have more overhead of workers doing little/nothing.
It works out to be about the same. 15-20% of corporate people are busy sucking up to the boss and 15-20% of government people are making paper airplanes or whatever.
Government generally has professional staff who have some sort of clue, just like in the corporate world. The difference is that there is another layer(s) of management about the professional managers and directors -- political appointees. Usually the political types know they are dumb and stay out of the way, but sometimes they decide to flex their power -- resulting in many a dilbert moment.
Conformity is the jailer of freedom and enemy of growth. -JFK
Why?
Probably hard to find parts for whatever vintage 1975 mainframe it runs on. If you only know PC's, imagine trying to buy a brand new VESA bus SCSI card for your legacy 486 mission critical system.
What is it with these fans of Senator Bill Frist and them always wanting to sound off about him at the start of every slashdot thread?
And if it doesn't work, you give them all 'F's and start again with the next incoming class.
Not on that scale but still a hefty chunk of change nevertheless.
The cost and complexity of moving the entire payroll and finance system over to peoplesoft was so much that it lead to the resignation of the CFO of the university because he spent more without the authorization of the board - never mind that the board and the president pushed for this improvement knowing the budget will go over from $25 mil to $40 mil or so.
http://www.vermontnewsguy.com/tag/daniel-fogel/
Bottom line is that these type of projects are incredibly complex and no one really knows the long term costs when they get into it initially. But due diligence and oversight would be critical and helpful no doubt.
I am a software architect and ex comp sci. lecturer.
I guarantee you are being exceptionally naive.
$20 million+ rollouts for Peoplesoft systems on University campuses are the norm, not the exception. Their salesforce hooks in clueless upper management with tales of little customization and off-the-shelf savings, and then comes the roll-out consulting costs and news that any use of Peoplesoft for financials requires highly complex, site-specific customization at exorbitant consulting fees.
Data migration from the old mainframe systems always turns into a nightmare, cost overruns are legion, political pressure to meet deadlines causes internal staff to rack up huge overtime at huge cost, Oracle licensing runs well into 7 figure territory, etc, etc
This money was gone the second they selected a Peoplesoft "solution", management just didn't know it at the time
Who wants a job doing a payroll system using a third party tool set based on old languages and technology....doomed from the start.
I don't know about the rest of you guys but I have never looked at our payroll drone and wished I had his job.
Got Code?
First, it's not really $40m, it's only $40m b/c they're doing it 2x. It's closer to $20m.
And...it's not just upgrading, as pointed out. It's a complete new system. Any system as old as their previous system is probably in need of replacement rather than simply refactoring and basic updating.
I just did some back of the envelope calculations. So...outside firm bids. Let's say that we'll have 24 minions (basic programmers, project management, requirements, documentation, etc). We'll say they average $70k/year salary. We'll double that for overhead, so $140k/minion/year in salary & overhead ($3.36 mil). Let's say those 24 minions have 8 middle management/tech leads/etc on top of them. We'll say they make on average $100k/year, so $200k/year/manager ($1.6 mil). Finally, we'll say there are two top dogs (architect, partner) on the project. We'll be conservative and put them at $150k/year ($600 k). Now everyone is going to need computers. Let's say $2k/computer ($72 k). This new system isn't going to run on their old hardware, so we'll get them new fancy hardware for $10-$15 million. Finally, the good folks in the payroll dept at the university need new hardware as well to use the new system, so we'll put them down for 20 computers at $2k each as well ($40 k). Grand total is between $15.7-$20.7 million. Granted that this is still less than the $28.4 they actually spent, but clearly there was some overcharging and incompetence going on, so we'll put the extra $8+ million down to false-starts and other poor decisions made by the consulting firm. And, when you're all said and done, the simple project ends up being quite expensive. I have no way of knowing how my cost breakdown compares to how their money was spent, but it makes reasonable sense.
Has there ever been a Peoplesoft implementation that wasn't a very expensive fuckup? I've certainly never heard of one. The only thing that amazes me about this pathetic excuse for software is that the scam lasted as long as it did before Oracle mercifully put them out of their misery.
Often it is a politician showing that he or she is completely ignorant of technical issues
Payrolls are hardly technically challenging. By way of perspective, 30 years ago I worked at a computer bureau, which for those too young to remember such a thing, was a shop where businesses brought in their handwritten input data on paper forms, and our keypunch ops would encode it on to mag tape for us to process on our Burroughs B3700 computer.
We ran our in-house payroll package for everything from public services to market gardens, and there is no reason why it wouldn't work just as well today, other than that it was written in COBOL, which isn't so trendy any more.
The world will be a better place when all the managers retire who were raised without computers.
The managers who used our packages were ALL raised without computers. That did not make them incapable or stupid. The world will be a better place when kids stop belittling their elders for no factual reason.
I have to believe there are hundreds of universities that already have working payroll systems. Look at as many as you can, select the one that works best, purchase the basic product, pay the university in question for their mods and steal their configuration. Get a bunch of work study students and a decent consultant or two to migrate your data. Project complete!
These people are just pissing money away on junk!
I worked for a company (large company, thousands of employees, worldwide sales and marketing, multiple manufacturing operations in multiple countries) that had a horrible 5+ year SAP implementation with obscene amounts of customization and they spent less that that!
I bet a group of enthusiatic CS/IT students with programming skills and maybe one teacher with real life experience can build and/or fix this in 4 months. Give them the tools, have them prepare by giving them access to all personell doing payroll stuff and familiar with the process of payroll and pay them a good salary plus a bonus if they finish it before next winter-semester is over. Give them option to do their thesis or degree paper on the project. Add in a few law students if complicated German-style tax stuff is involved for some extra interdisciplinary flavour and results.
Voila! Top-of-the-line payroll system for something like 100 000$. ... And, sadly, I also bet that that won't happen, because then someone would have to admit that he burned 20+ Million on a project that was implemented start to finish with less than a tenth the money. Sometimes the sad and sorry state of our profession in some places makes me want to cry.
We suffer more in our imagination than in reality. - Seneca
What are you basing your optimism in?
I (and many other old timers on this thread) are telling you in no uncertain terms how the cookie crumbles, so what is your evidence that what you are saying could actually be done in the way you say?
What you are suggesting is stupid and naive (a word I have seen used several times on this thread, and rightly so), that you are moderated "Interesting" a the moment just comes to show how few people in /. are familiar with the complexities of such systems.
IANAL but write like a drunk one.
Sorry to dis-aillusion you, but take a look at NU standard of living data - we actually have quite good health care up here. There is a lot of scare mongering about "long waigts", and in some of the bigger cities you might have to wait a few hours if you walk into emerg. without a critical condition. But I don't have to reach into my pocket to pay for a visit to a clinic or my physician. And when my mother had to spend a month in the hospital, we did not have to mortgage her house. And - best for last - if you come down with a serious illness up here, you won't ever get told "sorry, you you don't have insurance, go somewhere else to die." Last time i went to a hospital, I was treated within 2 hours - and it was a non-life threatening injury. seeing the guy wheeled in on a stretcher, covered in blood, it was no problem letting him go first (they call that "triage"). Get the facts.
Please don't dominate the rap, Jack, if you got nothin' new to say.
I really can't believe this story. I wrote a integrated Payroll/HR application in 2 years (plus received a US patent for part of it) myself for a major steel corporation in the early 1990's using Clipper! It did all timecard entry, had user-defined union rules, and tax rules. It did taxes for US & Canada plus 21 states. All user defined and maintainable. Printed laser MICR-checks, W2's, direct deposit, retirement and pension calcs, etc. etc. etc. It was used for 15 years until the company was bought out. How fraking stupid are these people? As a side note. You have NEVER felt pressure as when 2000+ United Steel Worker checks are wrong and you don't know why! (It was (L)user error)
And why do so many Americans come to Canada and lie to get basic medical treatment?
.00001% of the population.
Here's the deal:
Most Canadians would rather not get stuck in some American hospital bankrupting themselves for survival. So most of us don't go to the US for care.
The thing about Waiting Lists is that because we have regular contact with our doctors (free contact) we can identify issues early on. It's called Preventative Medicine, rather than addressing problems after they happen, we identify them before they start. It sucks to wait 9 months for a knee replacement, but that knee problem was spotted potentially months earlier for a Canadian as opposed to an American who may have waited until they were forced to go to an expensive doctors visit because their knee doesn't work.
However there are some treatments that are more developed in the US, some of the *rare* cancer treatments. That's because there are 300million Americans, allowing for more specialization in health problems that affect a very small
And here's the kicker: Our Socialized medicine will pay for Canadians to go to the US for these treatments when they aren't offered here.
Some Canadians aren't going to the US for rare treatments though. There are also some who feel that good enough Health Care isn't fast enough. They spend their own dough in the US, saving my health system money and generally showing their stupidity.
And then there are the Boobs. I'm hard pressed to find anything comparable to the Beverly Hills Plastic Surgeon in Canada. Of course Socialized medicine shouldn't be buying Boobs as a general rule of thumb.
-------
What it really comes down to is LIFE EXPECTANCY
You can shout to me until your blue in the face that the United States has a superior Health Care system than Canada. It's certainly more expensive per capita, but 2 years after you're dead I'll still be here laughing.
The rock, the vulture, and the chain