Software Development
Software Development
Author
Discussion

2gb

Original Poster:

10 posts

238 months

Thursday 25th May 2006
quotequote all
Hi,

I have been involved in internet marketing for years and have had a few products developed offshore with verry few problems.

I am now wanting to develop a large scale application and am looking for someone who can do technical layouts and manage the offshore development team as I want no problems with development time or budgets.

Would my best idea be to hire somboday or do you know if UNIs etc offer a service like this?



Thanks in advance

Jared_m

252 posts

245 months

Friday 26th May 2006
quotequote all
Most of the larger offshore development teams will have their own project managers and architects but if you need face-to-face engagement then look for someone nearby.

You can try www.freelancers.net to find the people you need and there are a couple of other sites like it.

iaint

10,040 posts

261 months

Friday 26th May 2006
quotequote all
From limited but painful experience of off-shore development you should not rely on their project management or have any expectations of accurate communication.

While technical skills are high the off-shore outfits often have little understanding (or interest) in providing a maintainable solution and the initial few deliveries of product from them rarely interpret the specification/requirement in the way you intended.

While they will provide some Project Management it regularly turns out to be more like team leadership than PM. You should seek to provide some form of oversight that acts as a PM for you the client and spends a reasonable amount of time out there. This individual must understand both PM, technical issues and (critically) your business and the product as well as is possible.

Quality of work is rarely an issue if you ensure it's managed according to your expectations. You also want someone on hand there to report back accurately as there is a tendency for progress reports to be fine until a crunch date where it turns out everything's now 2 months behind schedule!

Pick somewhere nice for the offshoring and then you'll find it easier to find a good person to manage the work for you

s a m

509 posts

260 months

Friday 26th May 2006
quotequote all
I tried off-shore developers a couple of times with some small projects, things that we didn’t really do in-house at the time, testing the water as it were. Unfortunately it was a bit of a waste of time. They don’t really grasp how we do things in respect to processes, quality, maintainability and reliability. You have to test and validate every requirement independently; don’t rely on them saying it works or has been implemented in “the right” way. Check every little detail. Needless to say I had to learn how to do it, and build it my self. Never again

I recently got asked by a guy who had spent a few grand on a medial related site; it had been developed off-shore to his spec. He needed someone to clean it up, fix the bugs and improve some of the implementation. He said that the off-shore guys were cheap but got it all done way too quickly, then weren’t really interested in sorting out the problems. For the record, what was produced would have been a nightmare to maintain. I’m not sure what happened with that one, it went cold after a while so he may have dropped it.

Also a few of my contacts have tried it and not got very far with it. One got a quote from us to customise our order management product, but he decided to have one made from scratch by a company in India... As of two weeks ago, he is now using our product, with the customisations and seams happy as pie.

So personally I would avoid it, I know there are good people over there who can do better than the UK, but they won’t be 50p an hour and I expect they are harder to find.

May I suggest starting with a smaller project, don’t tell them there is a larger one, see what the quality is like then if acceptable go for the big one. It could save a lot of problems down the line.

(There are obviously crap UK companies too, so its not exactly UK=best every time!)

Added: The other background I have on this one is from the hundreds of spam emails I get from Indian development firms offering out-sourcing “relationships”.

I have looked at some of their portfolios and examples… not good.


Best of luck with the project though, if you find a good team you’re laughing.


>> Edited by s a m on Friday 26th May 11:08

2gb

Original Poster:

10 posts

238 months

Friday 26th May 2006
quotequote all
Thanks for all the replies

MY main issue is no matter where I hire the development team I have no idea of development time of products. I am clued up on web design but when it comes to applications I could be told anything and as I have no idea if its true or not.

I think what I might need is an indavidual with experience who can design the system layout and watch over the team. Do you think If I contact a local University they may be able to advice of a suitable candidate as Glasgow seems to be a little limited of developers.

I take note of your comments about offshore teams and have had problems in the past of the developers reselling all the code to anyone with $1000 and a domain name but I have found some of the romanian teams reliable with good communication

Thanks
Lee

J_S_G

6,177 posts

273 months

Saturday 27th May 2006
quotequote all
OK, it's a Friday night and I'm onto the second bottle of fizz of the evening, so apologies for the effect that's likely to have on what I'm about to write, but here goes...

Offshoring in general
I rarely see offshoring as the "cheap" option. I.e. I don't pick it as a cost saving exercise. I generally only advise software to be offshored where you either can't get the development resource locally, or you have nowhere to put them even if you could find them. This is based on projects where the software's "high value" - i.e. it's one of the core assets being produced, not just a churned out system that does little other than sprawl. Standard development costs all-in have tended to work out a good 30% cheaper when offshored, to be fair... But by the time you've had any deliverables reworked to be what you actually wanted, but didn't document 100% correctly (given the communication issues - language barrier, time gap, etc), and added on local management costs (more on this below), it's pretty much invariably been 15% more expensive at a minimum. And far worse at times. The main issue has always been "you get exactly what you ask for" - no questions about clearly incorrect requirements, etc.

Managing offshoring
The only management structure that I've found that works for offshoring is man-marking. I.e. for every remote manager, you have a local manager ensuring they're on top of things. It's this overhead that gives you visibility you're getting what you're want (i.e. project de-risking). It's also this that bumps the cost back up beyond local dev. costs (i.e. payment for that "insurance"

Skillsets, etc
Rather than just leaving the entire technical solution to some remote company, I'd be more inclined to split the delivery from the analysis and design. The "higher" technical skills - architecture, for instance - I'd recruit locally; get the relevant people close to the client (yourself/whoever) so they can really understand the problem, short-/long-term requirements, approach needed, etc. Documenting all that is the cheapest looking option, but something will be missed - some context will be lost. Sending these people out to work with the development team(s) can be a better option, along with a good project manager (possibly not a team lead, as the cultural differences could cause inefficiences in the team there).


If all that sounds a bit complicated and over-blown, it regularly is. I see it a bit like the shipping/importing... It's one of those models that can be quite successful when scaled to a decent level, but has high entry costs - if you've got the resources on your doorstep, there's less to go wrong. It's definitely not worthwhile on small projects.

With a large project, I'd take into consideration whether it's "large", or "large and complex" (very different things). Then I'd consider the requirements - whether they're generic (standard form-filling) or based on specialist knowledge (i.e. some intelligence has to be captured). Then there's the delivery approach - whether a straight out v1.0 bullet-proof release is needed, or a more iterative approach to give a quicker initial turnaround and ease cash-flow.

Just drop me a line if there's anything I can do to help - I've managed the development of some "not small" technical projects.

Again, apologies for 1/2 price fizz at Sainsburys taking its toll tonight...