PDA

View Full Version : Outsourcing



SCOUT
3/9/2006, 11:08 PM
I know outsourcing technical jobs is a really hot topic these days, but I would like to hear some opinions on a real life example. I would like to preface this with the fact that my company has almost no interest in outsourcing our developers or anyone else for that matter.

I work in HR and I receive solicitations from companies all the time to outsource various roles. The most recent one is for a firm that is based in India. They provide candidates that we can interview over the phone prior to working for us. They would work US business hours and have a us contact number. They provide free internet connectivity and remote monitoring of their work. They also handle all administrative functions for their developer pool including training, retention, performance management, payroll etc.

Here is the kicker... Their Java Developer rate is $9 per hour :eek:

We currently have Java contractors making in the $70 per hour range.

I won't bore you with the math but if a company has even just 6-10 developers, that is a pretty sizable savings.

Thoughts?

GottaHavePride
3/9/2006, 11:18 PM
Thoughts?

I need to learn me some Java.

jkm, the stolen pifwafwi
3/9/2006, 11:35 PM
i can never remember the name of that company, but they will NOT be beat on price. i've seen them go as low as $4 an hour.

i could go on and on about outsourcing. the key item is that as a company you have to evaluate your current relative strengths, especially in the areas of PM, requirements gathering, dev, test, and support. if the first 2 are weak, there is no benefit to outsourcing, as a matter of a fact you'll lose money as they will delay and change order you to death. if the first 2 and the last one are strong, you can reap a ton of rewards because these areas don't change as rapidly with technology as the other two sectors.

we have a dev center in china and those cats are a lot better at development than most of the indian centers because they are able to a) systematically problem solve and b) recognize holes in requirements and ask about them up front.

personally, i automatically recommend outsourcing production deployment tasks. it is by far your riskiest item in IT as the business requires you to do it at the witching hour. bastards. :mad:

jrsooner
3/9/2006, 11:41 PM
Here is the kicker... Their Java Developer rate is $9 per hour :eek:
We currently have Java contractors making in the $70 per hour range.
My daddy had a quote... "You get what you pay for".

I've been with companies that have done both. I've been on the SQA side of software development, and I've noticed that most of the time you don't get "quality" from the outsourced groups. Usually your deadlines go unmet with all the revisions and the meetings to finally get the thing right. They usually develop exactly what you say, then if you got something wrong, they'll bill you for the additional work. Where as if it's in house your dev team will probably speak up about "blatant" errors. The key to a good quality team is catching the errors at the start of a project not at the end. The sooner you recognize that problem, you'll spend less dollars fixing the gap where if you catch it at the end, you are in it full $$$+. That's on the dev side. If you outsource your process and controls group, you'll get a ton of "canned" processes that don't really fit your company, and you'll spend a ton of cash trying to get it to fit into what they say you have to do based on their can scripts and checklists.

BajaOklahoma
3/10/2006, 12:09 AM
If I am having a computer issue, my stress level goes up amazingly when I can't understand the person on the other end of the phone. I don't care how nice the person is, if I can't communicate effectively, I get madder.

OTOH, I like the guy who comes to the house to work on my computer. He is going to be on my Christmas list next year.

crawfish
3/10/2006, 08:43 AM
Just see if it's less expensive when a bunch of us developers form a posse and burn down your office building. :mad:

;)

IB4OU2
3/10/2006, 08:57 AM
I had one of the "outsourced" persons call me yesterday trying to sell me some type of technical publication. I couldn't understanding anything she was saying and after several attempts of asking her to repeat what she was selling I finally said I wasn't interested and terminated the conversation. I don't know what company she was representing but I don't think she's making them much money.........:rolleyes:

Cam
3/10/2006, 07:09 PM
i could go on and on about outsourcing. the key item is that as a company you have to evaluate your current relative strengths, especially in the areas of PM, requirements gathering, dev, test, and support. if the first 2 are weak, there is no benefit to outsourcing, as a matter of a fact you'll lose money as they will delay and change order you to death. if the first 2 and the last one are strong, you can reap a ton of rewards because these areas don't change as rapidly with technology as the other two sectors.
Spot freaking on.

BeetDigger
3/10/2006, 07:12 PM
I've been experimenting with outsourcing. For the past three weeks a guy over in India has been posting for me. Now, I get work done but miss out on none of the nonsense. Woot.

Okla-homey
3/10/2006, 07:14 PM
All I'd add is don't feel bad if that's where the numbers lead you after serious consideration. Free markets baby! Just think what kind of crapola Detroit would still be cranking out if the Japanese hadn't hit the domestic US scene in the 70's.

yermom
3/10/2006, 07:17 PM
All I'd add is don't feel bad if that's where the numbers lead you after serious consideration. Free markets baby! Just think what kind of crapola Detroit would still be cranking out if the Japanese hadn't hit the domestic US scene in the 70's.

yeah because the cars in the 60's were awful ;)

jkm, the stolen pifwafwi
3/10/2006, 08:12 PM
All I'd add is don't feel bad if that's where the numbers lead you after serious consideration. Free markets baby! Just think what kind of crapola Detroit would still be cranking out if the Japanese hadn't hit the domestic US scene in the 70's.

the quality of outsourced code is beyond suckitudinal

Okla-homey
3/10/2006, 10:58 PM
the quality of outsourced code is beyond suckitudinal

So, that said, what's the appeal? Free patches for life or something?

RedstickSooner
3/10/2006, 11:09 PM
So, that said, what's the appeal? Free patches for life or something?

Same as any industry with such a disparity in hourly rates. You hope that the cheaper work, while shoddy, will be good enough to just scrape by.

Same reason you go to a doctor who'll charge you five bucks a visit. Yeah, he probably sucks, but surely if you're persistent, he'll improve your health *some*, right?

Same reason you go to the two dollar hooker (which is how you end up needing the five dollar doc) -- wrap that soldier in two biohazard suits, and surely you won't catch anything, yeah? I mean, c'mon -- it's two bucks! How can you go wrong?

Savings like that are hard to pass up. And you also sorta figure that even if they code badly, after they've worked for you long enough, surely they'll have had enough practice coding that they'll improve, right?

Just think how much it'll suck for American programmers in another decade or so, when the Chinese and Indians are actually *good* at it.

My wife's in the business. Thank the gods she's no programmer, though. Instead, she's one of those folks who can sell an umbrella to a fish -- and sales ain't never gonna get outsourced overseas. Not in my lifetime, anyway.

jkm, the stolen pifwafwi
3/10/2006, 11:48 PM
the appeal is in the bottom line. that alone shouldn't be the determining factor, however it is because of the "corporate ladder" dilemna.

moving straight up the ladder in a single department is an arduous task. the ones that succeed either waste 1/2 of their lives or are lucky enough to be a golden boy. executive level folks are enamored with the grass on the other side - they'd rather take a chance on their appraisal of a candidate after an hour interview then promote the guy where they know every strength and weakness. its ignorant, but its the way it is.

this leaves you with 2 options -> abandon the firm or jump to another ladder within the company. because of a fear of change, the second becomes a better scenario than the first (even though the first will get you a heck of a lot more $$$). so people jump from the business side to IT and then back to the business. that means that you end up with a lot of business folks in upper management in IT. since coding is totally foreign to them, they look at java and c# as if it were excel or outlook. this line of thinking means that if java is on your resume and you can answer this stupid list of questions i found on the internet, then you must be a great coder. the problem is that you can't see the difference between a great coder and a horrible coder until the first cut of code. by then, its just cheaper to stick with it.

for the record, this also applies to hiring consultants as managers. they have a chip on their shoulder about consultants and want to beat down their rates. they expect to get a $150/hour consultant for $60/hour rates, just because they know how much the consultant makes.

Cam
3/11/2006, 05:48 AM
As stated above, the appeal is the immediate bottom line. Unfortunately, what's not taken into account is the massive amounts of $$$ that shoddy coding costs a company when a system or application goes down, repeatedly. Not only do you have the cost of change tickets, but you also have the cost of downtime by 100's, if not 1,000's of employees that's just not taken into consideration. They're different line items, so it's somebody else's P&L that you're jacking with so why should you care?

It's been my experience that you see this a lot in companies that let IT run the business instead of the business running IT. What I mean by that is the business brings to IT a list of their top 10 items they need to achieve their goals. IT comes back with "We'll do 3 of them, #'s 4, 7, & 8". They don't/won't see that the business would've gotten more bang for their buck by getting #'s 1 & 2 completed. They're only looking at their costs.

The companies were I've seen this happen the most have zero business experience in their IT department. They don't understand the side of the end user of their apps because they've never been the end user. IMO, you need a mix of business and IT experience in the IT side to be successful. Not 50/50, but maybe 75/25 IT/business. Depending on the individuals, that ratio would be sufficient to get the goodness of IT and business knowledge that would allow enough dialogue to do the right thing for the health of the over all bottom line.

JKM is 100% on about hiring coders. You never really know what you've got until their first project is underway. Most of the time, it's nothing more than blindly throwing darts at a dart board and hoping you hit the bullseye.

OUinFLA
3/11/2006, 07:30 AM
OTOH, I like the guy who comes to the house to work on my computer. He is going to be on my Christmas list next year.

Isnt new technology amazing.
Down here for years, most women have had the same thoughts about their "pool boy". Now, it's the "computer tekie".

12
3/11/2006, 07:49 AM
Here, it's the "Vet Assistant."

sooner n houston
3/11/2006, 08:35 AM
I've been "outsourced" so many times now I've lost count.
Funny thing is there is now a swing back to doing it "in house" and to everyone's suprise, there aren't enough coders to handle the demand so wages are on the uptick again. I would bet 50-75% of the people I've known in IT have moved to the functional side of the house in the past 5 years and aren't about to move back into IT.
One good thing about WalMart, they have 3,500 full time people in their IT dept. with an additional 1,500 of us "contractor puke" types as well. They don't seem at all intersted in outsourcing!

ChickSoonerFan
3/11/2006, 10:57 AM
Isnt new technology amazing.
Down here for years, most women have had the same thoughts about their "pool boy". Now, it's the "computer tekie".

uuummmm...no.

There is quite a difference in the "pool boy" and the "computer tekie"

just sayin

;)

jkm, the stolen pifwafwi
3/11/2006, 12:09 PM
It's been my experience that you see this a lot in companies that let IT run the business instead of the business running IT. What I mean by that is the business brings to IT a list of their top 10 items they need to achieve their goals. IT comes back with "We'll do 3 of them, #'s 4, 7, & 8". They don't/won't see that the business would've gotten more bang for their buck by getting #'s 1 & 2 completed. They're only looking at their costs.

honestly, i think this situation is a corporate culture thing. yes, at microsoft, IT determines what the business wants (normally based on what coolest to work on). but at cingular/(at&t wireless at the time) the business told you what and when they wanted it and you had to deliver (most of the time with no requirements).

the problems that i've seen with the business ranking projects is that they are normally ranked by political weight, not by impact. the higher the exec, the more weight on his dumb pet project. its one of the reasons why placement of your PMO office in the org structure is so hard. people are not going to **** off the wrong guy.



The companies were I've seen this happen the most have zero business experience in their IT department. They don't understand the side of the end user of their apps because they've never been the end user. IMO, you need a mix of business and IT experience in the IT side to be successful. Not 50/50, but maybe 75/25 IT/business. Depending on the individuals, that ratio would be sufficient to get the goodness of IT and business knowledge that would allow enough dialogue to do the right thing for the health of the over all bottom line.

i don't necessarily agree. i think the issue is that architects tend to be technology geeks instead of process geeks. they push down technology decisions, not process improvements. they knock all sorts of things out of the system because of "performance degredation". i've lost track of how many times i've said "it isn't incremental performance gains, its overall user cycle time". hardware is cheap, the key item is to minimize your error rate and maximize your throughput. its one of the reasons people pay my rates, because i design systems like i am a super user not someone who has to be on the cutting edge of technology ;)

sanantoniosooner
3/11/2006, 12:11 PM
I'm going to hire a guy from India to root for the basketball team for me.

Czar Soonerov
3/11/2006, 12:14 PM
I outsource all my farks. It's great, they can do a fark for a fraction of what it would cost me to do it.

Cam
3/11/2006, 07:25 PM
honestly, i think this situation is a corporate culture thing. yes, at microsoft, IT determines what the business wants (normally based on what coolest to work on). but at cingular/(at&t wireless at the time) the business told you what and when they wanted it and you had to deliver (most of the time with no requirements).
It is 100% a culture thing.


the problems that i've seen with the business ranking projects is that they are normally ranked by political weight, not by impact. the higher the exec, the more weight on his dumb pet project. its one of the reasons why placement of your PMO office in the org structure is so hard. people are not going to **** off the wrong guy.
At my company, it's not political, it's ALL about the P&L, nothing else. Our former CIO made a mandate that IT Opex would be no more than 1% revenue. Once his PM's decided to stick to that mandate, down time increased for every in-house supported app. Since he left, the amount of charters has increased 300% and we're actually able to get headcount to do the work that we need done. Hell, we even offered to pay for multiple projects and were told no because it would raise the IT spend. At that point, it became political because the CIO wanted to save face.


i don't necessarily agree. i think the issue is that architects tend to be technology geeks instead of process geeks. they push down technology decisions, not process improvements. they knock all sorts of things out of the system because of "performance degredation". i've lost track of how many times i've said "it isn't incremental performance gains, its overall user cycle time". hardware is cheap, the key item is to minimize your error rate and maximize your throughput. its one of the reasons people pay my rates, because i design systems like i am a super user not someone who has to be on the cutting edge of technology ;)
Designing an app based upon the users needs. That's a pretty cool concept. ;)

I've sat in quite a few meetings where developers flat out said no, you don't need that functionality to do your job. I've offered to set up shadow time for them, but sadly none have taken me up on the offer.

There's no reason why process improvement and technology advancement can't walk hand in hand.

It's probably a mixture of both of our scenarios. I can't necessarily disagree with your points, but I'm not in the IT department meetings, so I'm not privy to exact reasons for a yes or no.