Cutting costs generally means cutting corners.
Cutting corners generally means getting stuff wrong.
Outsourcers are to blame for causing major disruptions at one in three major UK companies over the last three years, according to research by consultancy firm Deloitte. The global survey, which included the responses of 107 UK companies across different sectors, also found that 80 per cent of UK companies are dependent on …
"Clearly you need to a team of expensive consultants to write a report to confirm this and proffer suitable amelioration strategies."
Actually, you do. Because, IME, management will always rather listen to what an external cunsultant says than to their own expert staff. Even if it's exactly the same.
Actually, you do. Because, IME, management will always rather listen to what an external cunsultant says than to their own expert staff. Even if it's exactly the same.
Funny, but as one of those highly paid experts, I was brought in to a client to help advise them on their architecture. While at this client, they asked me a question about finding anything on the 'net that addresses their problem.
I told them that I didn't think it existed, however I could prepare a write up if they wanted...
They turned me down.
I don't think that they realized that I was one of those people that do the write ups that other search for on the 'net. Or did the work that got written up because as a consultant I was under an NDA.
The point is that even customers don't listen to their highly paid experts.
Posted anon for the obvious reasons.
"management will always rather listen to what an external cunsultant says than to their own expert staff. Even if it's exactly the same."
That's because it costs them more. Price = value. The consultant can ask the expert and then pass the answer on along with a big bill. It's called adding value.
But I'd also add that cutting corners and costs means not actually having the quality people to actually fully scope and understand any new or incumbent solution in the first place.
Wipro and HPE are the worst I have come across for a lack of joined up total thinking when it comes to the project management, design and delivery of complex e2e enterprise solutions.
They do all like their little independent silos.
Because how many of those outsourcers staff actually work for them?
Not damm many.
The actual domain experts will probably be in sub-contractors, or sub-sub-contractors.
And under strict instructions not to let on that in fact your ERP/CRM/Mainframe/Payment system is in fact not being installed by Atoss/Crapita/SCS but Joe Blogss Contracting (who actually know WTF they are doing).
Well maybe, having worked on a few contracts, the things that go wrong are usually down to the relationships, and it's rarely, wholly the fault of the outsourcer.
1. The relationship between users and IT changes, it becomes all about moving money, before outsourcing the money isn't visibly real before, after it is, suddenly as a user you have to think about costs you didn't before.
2. Purchasing Departments are naturally adversarial and are incentivised to save money by arguing bills, thus making change processes more complex.
3. Outsourcers front load investment, and then try and reduce that exposure after the fact, often promising bid teams investment that then fails to materialise leaving the delivery team exposed and disincentive.
4. You never get the best quality from the lowest bid.
5. Outsourcing contracts are hugely complex, after all, a minor government department may have 200 odd systems.
6. Client rarely know everything they have in their estate, and then try to make the outsourcer take the liability for stuff they didn't know about.
7. Why would anyone think putting another 2 layers of management between the user and the systems be efficient and cost effective.
8. Contracts take so long to negotiate that they are out of date before they start.
and so on. Having said that outsourcers don't help themselves.
Another key thing is that if the outsourcing business is of any appreciable scale, the chances are that the outsource vendor cannot do the work any cheaper. Whilst typically the vendor has better access to cheap offshore labour, productivity is usually lower in those markets due to things like unsocial hours working, high staff turnover, focus on low salaries even in the low cost location, limited language skills, and reluctance to train staff who will shortly move on. After all that, offshore may still have a lower cycle cost in direct labour terms, but the real killer is that the vendor has to charge higher margins than most client organisations would need to "pay themselves" to cover capital employed, has a balance sheet dripping with goodwill that has to be paid off by the clients, they incur significant setup costs, and not always cheap bid costs and ongoing account management by onshore staff. In aggregate, somebody else doing your processes will cost more than you doing them yourself.
As the vendor needs to be cheaper per transaction when the new arrangement launches, they are then incurring a net loss, and they have to backload those losses and the additional costs mentioned above, and the way they do that is through variations, changes, and non-standard order costs. That is the vendor's model in ITO and BPO. They know that cost-reflective pricing up front would result in zero sales, so every deal is a loss leader, with the certain knowledge that every organisation will find needs for significant change over a five to seven time period, and that's when they start to fill their boots. Also, because businesses have very short term organisational memory, within three years everybody will have forgotten the costs of in-house delivery, the standards of service of in-house, and the business case for the outsource will have been shredded or lost. I speak from experience in a European organisation in exactly this position, being reemed out by a large US based IT outsourcer.
4. You never get the best quality from the lowest bid. (And the others..!)
old woman Otto is a bear -- you are accused of heresy on three counts -- heresy by thought, heresy by word, heresy by deed, and heresy by action -- *four* counts. Do you confess?
I once heard tell of a well - known broadcaster (NNNPD) that outsourced the maintenance of the broadcast equipment at one of its locations. Not long after, just as the local news was starting, something in the programme chain failed and the studio 'phoned the maintenance men who had, of course, been TUPEd. "We''ll deal with it first thing in the morning". "But you used to provide support within minutes" "Yes but the contract states that we now have a 12 hour response time". (Or something similar)
> Outsourcers blamed for cocking up programmes at one in three big firms
Which could be (mis?)interpreted as meaning that the other two in three cock-ups were due to the firms' own in-house staff, and therefore be an argument for outsourcing, since apparently outsourcers may be only half as incompetent as in-house staff.
I'm not saying it is so, just that the language is imprecise enough to allow this possible interpretation.
If a client is outsourcing something, then it frequently means that they want a particular solution, but don't have deep expertise in that solution.
Which means their understanding of the solution they've chosen is limited.
Which means that if something goes wrong, they are limited in evaluating HOW it went wrong. The unscrupulous will just blame the contractor for doing what he was told to do.
After all, the usual story a contractor will tell you is making a support call, looking at the client's equipment, and saying to himself, "Holy mother of fuck, what have these people been up to?"
If you understand the small print.
I used to work at a national telecomms company which had a major drive to reduce overheads by outsourcing everything it could.
First was the DBA support, which was handled by TUPE'ing staff over to a major international IT company. After which, while we were dealing with /the exact same people/, things took a noticable turn for the worst, as they rigidly stuck to the letter of the contract.
Database performance problems? We can kill the queries, but we can't comment on what's causing the performance issues. Not in our job spec, guv.
Roll a patch out of hours? Not in our job spec, guv.
And so on. Some of this was driven by people being jobsworths, or by people upset about the forced transfer. But most of it was due to the simple fact that anything not in the job spec could be charged for, and their new employer wanted to maximise revenue.
At the same time, increasing amounts of system support was being handed over to outsourced ownership - generally offshore suppliers with one or two heads onshore. In the short term, support quality absolutely nosedived, not least because the supplier was again looking for ways to maximise revenue.
We need to reduce the number of processes? Well, you're using a spreadsheet to track X: let's ban use of that spreadsheet and call it a process so it counts towards our SLA.
In the end, people spent more time arguing about the letter and spirit of the contract than anything else.
I'd like to say that quality improved as the outsourcers gained experience, but truth be told, that's where the other issue with outsourcing comes in: the people doing the work have no reason to be "loyal" to your company, and can often get a better paid job once they've built up a bit of experience.
So you get into a cycle of training people up, building relationships and then having to start again from scratch.
And let's not get started on the joys of relying on outsourced teams for software development...
Or maybe we're all just talk and no action.
I once had this brilliant business idea; a product I could prototype myself in the metaphorical garage, with what seemed to me like a pretty good marketing, sorry, "go-to-market" story about what it was and why you'd want to pay lots of money for one. I spent six months or so of spare time working on this idea and getting quite over-enthusiastic. Anyway, during this period I heard a round table discussion of some sort ("Bottom Line" on BBC R4, perhaps?) that included a guy from Google who made what seemed to me an astute observation:
Any fool can have an idea. It's all about the excution.
Needless to say I realised I was in the "any fool" category, rather than the "future Dyson, Curry or Zuckerberg" category.
Maybe I'm not the only one here that could say that...
"Well, there's the source of your problem: believing that a contract has spirit as well as letters."
A contract may not have spirit, but the people implementing it do.
I've seen it both ways - the DBAs may have worked to rule, but I've worked with contractors and outsourcers who have been willing to go beyond the letter of the contract, whether that's working a bit over or helping with something they're not meant to.
And conversely, when I've been in an outsourced position, I've been willing to do the same. Sometimes with management approval, sometimes not.
There's arguments for and against this of course - not least the fact that it can leave you exposed from a legal perspective - but it's been my experience that working like that generally makes things run smoother and better!
"A contract may not have spirit, but the people implementing it do.... I've worked with contractors and outsourcers who have been willing to go beyond the letter of the contract"
True. We used to have a saying in the PCG forums that the contract is only for when things go wrong. If it's your own business there's a lot of value in having a happy client. It brings renewals or new clients because of your reputation; most of my contracting career I had pimp-free contracts for this very reason. (The sad thing, of course, is that the IR took advantage of the spirit of such arrangements to characterise contracts as IR35-caught.)
Offshoring and all that crap.
The beancounters see a profit on the bottom line and they are happy.
God help the poor sods who have to use the crap that the offshoring company produces or rather does not produce.
This report needs to be copied and the CEO's of all FTSE 500 companies (and beyond) made to read it TWICE.
Then the Beancounters need to read it three times.
Perhaps then one or two disasters might not happen,
I am probably dreaming or it the three pints of ESB that have been drunk tonight.
Back in the day this was only possible (in the UK) in large nationalized industries that had developed 1000 page rule books of what to do in every given situation.
But now it's all on a contract the first question is "What does the contract say?"
I once saw a large (900mm dia ?) aircon duct installed immediately in front of a switchboard's doors. Don't remember if it was a weekend job or a case of no 'local' staff visiting the job. Contractors organised by external consultants, perhaps they thought it was not being used anymore, well, it was !
Trouble is, consultants have no skin in the game, if something goes badly wrong, its another contract for them to fix it. And the guy who made the mistake, and hopefully learnt something, is elsewhere.
Look here's what happens.
Say its IBM. They write up an outsourcing deal that converts the company's tech to IBM employees.
Over the first phase, IBM slowly converts the employees to 'lower cost resources' and RIF the former employee. Due to the cost reduction, IBM makes a profit. As initial contract ends and its time for renegotiation, customer sees this and negotiates a lower price. IBM further reduces quality of staff and brings in less experienced and cheaper resources. Again next cycle same thing. And if they can't get cheaper resources, they reduce the number of resources. Overworked under educated staff tend to make mistakes and write crap code because they don't know what they are doing.
And you can bet IBM's lawyers wrote in enough legalese that it ends up being the client's fault and not IBMs.
Alternatives to this is the company going with IBM , then bringing in TATA or someone who claims that they could do the same but for less money. Bean Counters don't get it. They are like home cooks who can boil water thinking that they could work as a sous chef in a 3 star or better restaurant.
gee that sounds familiar
but with the following ending.
The company now has an accurate inventory of its systems, and at the end of the contract they decide to split the contract into 10 smaller ones going to different outsourcers. The local IBM entity that did all the work is not even allowed to submit a bid - but the equivalent IBM entity from India is. 2 subcontinental entities win two of the contracts, Accenture another and the rest go IBM's Indian entity.
The IBM India entity does not have manpower so subcontracts the local entity, who are shedding (expensive) TUPE'd staff who understand the business backwards forwards and sideways and replacing with (cheap) graduates with no understanding of the problem domain..
is that companies don't often realise how much "added value" they get from their in-house staff, and it only clicks once an external company either doesn't do it, or charges them for it.
It's not universal, but in general, in-house IT staff deal with an application as if it's theirs, and treat it as such ( although if they are just twiddling about with config of a 3rd party app it's not as clear cut as in-house written stuff, but that's an argument for another day) If it goes wrong, they know they'll have to deal with it, and so take a bit more care. if it's not what the business need, they'll question things, as it'll save them rework in the future.
From experience, the outsourcing companies treat it as someone else's, as it is. If it meets the requirements it's fine, and it's in their interest for the requirements to be incorrect, as that means added change requests... It doesn't need any added value that might make it slicker to use, leaner, or easier to maintain in future. If it throws out the odd error to the users, that's fine as long as it's in SLA etc. Its a bonus if it is to requirements but has loads of usage issues, as the client then has to pay them more to change it so it actually does what they wanted initially.
"It's not universal, but in general, in-house IT staff deal with an application as if it's theirs, and treat it as such ... If it goes wrong, they know they'll have to deal with it, and so take a bit more care. if it's not what the business need, they'll question things, as it'll save them rework in the future."
More than that. The in-house staff know that what the system does is what ultimately earns the money that pays them. They have skin in the game that outsourced staff half the world away don't.
When I was a contractor at Lloyds I watched in disbelief at the process that unfolded during one case of outsourcing.
Some sharp suited, quick talking American salesmen turned up and essentially conned senior executives into buying a product that was not fit for purpose and incredibly difficult to maintain (frankly it never worked properly in the first place). The executives where on the whole, IT and security illiterate so it didn't take much effort to pull the wool over their eyes though there were suspicions that some of them were in cahoots with the company concerned. They dumped their own technical people in favour of the outsourcers.
The product was poorly documented, the contract was written with deniability clauses built in and the requirements were poorly framed.
Initial promises of cost cutting have now turned into a damage limitation exercise as the full cost of ownership becomes apparent. The outsourcers have sold their product and the support agreement to an another company who now hold the bank to ransom for every small change required.
Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha! Ha!
Serve 'em right. I have no sympathy with big companies who outsource IT. IT is a CORE FUNCTION.
Biting the hand that feeds IT © 1998–2019