Is it about trust?
Is the issue only trustworthiness? I wouldn't be surprised to find out that companies ignore advice after being told what the cost of doing it properly would be.
Large companies across the UK increasingly turn to independent consultants when they want advice on the design and specification of a data centre. Almost all of them then ignore some or all of that advice, according to research released today. One hundred interviews were conducted among senior IT professionals at UK …
It's always possible that the problem isn't with the consultants themselves, but with the expectations generated from the sales process of getting them in to a client in the first place. If there is too much hype and too many promises made by the consultancies sales people then it's going to be impossible to live up to the commitments made by them, no matter how good you are as a consultant - or how quickly you work.
One problem may be that some consultants tend to be biased towards specific technologies because of their background and sometimes because they have an interest in promoting the products of a specific manufacturers e.g. other parts of their organisation engage in hardware sales. In short, some are not as independent as one would like.
Of course, another common reason for a particular choice maybe that a relative of one of the directors is in a sales role at the chosen supplier or that the IT manager has been offered a nice job at the chosen supplier.
I reckon that, lacking specific data center knowledge, companies are engaging consultancies and then using their proposal as a framework from which they can skew it to their own preferences - rather than just guessing.
Otherwise if they're just not taking advice they have paid for, why would they pay for the advice in the first place? They must be getting *some* value from it somewhere.
Mind you, they could just be skimping on cooling and ramping up the number of racks.
I disagree with the fact that consultants are the issue or are focused on a technology. I've done a DC build and the primary issue was getting budget for what the client considered not to be a 'critical' area such as power monitoring. I personally am not tied to any vendor and would always see what best market practice is. In one particular build I commissioned a company to build custom power distribution units so that all power could be controlled and monitored.
i know if I was doing a major overhaul of my data center i would have consultants in. however, i wouldn't have them in to tell me exactly what to do. it seems a responsible and learned admin would have a picture of what he intends to do for the upgrade or data-center build. then you get a consultant or 2 in and get some opinions. i, personally, wouldn't even detail my plan to the consultants. i'd let them (and yes i'd probably get 2 in there) completely design as they wish, then i'd take my plans as well as both of theirs and develop a solid system. the in-house sys-admin should usually have a better feel for the inner workings and details that may make what the consultants advise not work practically for the situation. if the in-house admins are just throwing out the ideas based on money, then that's irresponsible; but to take a consultants advise in entirety without any questioning or personal research is equally irresponsible. this article seems to make a few assumptions that aren't fully justified in what is written.
If you get supposedly knowledgeable people in then you shouldn't ignore their opinions. If you and they differ then either they're wrong (and you need to find out why) or you're wrong (ditto) or you aren't giving them enough info to be right or wrong. Or there's just a difference of views on priorities.
In any case you should - indeed must - drill down to the crux of this divergence; a 'responsible and learned admin' should be able to state exactly why they're rejecting advice.
1) accountant determines that they aren't going to pay for competent I.T. staff, thus requiring the use of external consultants. But that's OK, because the consultants don't come out of his(*) budget, and he got promoted when they were able to cut the obvious I.T. budget by 90% because they hired a bunch of chimps on work-release from the local zoo.
So they get consultants to make recommendations and provide detailed instructions on how to implement those recommendations.
2) accountant for the consultants looks at the recommendations by the techs, and adds an arbitrary 50% overhead for un-needed services. Because of course if they get the contract he'll get promoted, and if they don't get the contract it's not him that will be blamed.
3) accountant for client looks at consultant recommendations and chops much more off than his colleague added, thus getting promoted because he's "saved the company money". Of course, when the system fails and the data centre catches fire the money to fix it doesn't come out of his budget, so he's fine.
4) techs get the blame because everything is cocked up. Attempts to show that the accountants were actually responsible are shuffled aside, since said accountant is probably giving blowjobs to the CEO - who came from either sales, or sales+accounting, and is thus unwilling to allow accountants to get the blame for any corporate failure.
(*) all accountants I've met who do this kind of shit were male. Not saying the women accountants don't do this, just saying the ones I know and want to personally throttle with my gloved hands and a strand of concertina wire were all male.
company: "We want 5 9's reliability. Spec it." (The part not stated is "without spending any money".)
consultant: "That'll be be bignum dollars." (provide specs to turn out a kick-ass data center).
company: "Ummmm... yeah. Well thanks for the advice."
Part of this consulting should be to figure out what requires "5 9's" and what can get away with "4 9s" or so (99.999% reliability amounts to 5 minutes a year, 4 is still only 52 minutes a year.) But if someone just flat out says they wants highest reliability on everything, that's what'd be specced out, and it'll cost a lot.
Advice that doesn't sit well with the company doesn't get into the plan.
Advice that does, does.
The plan goes to steering committee where people who think they understand technology try and improve it. Then finance committee which thinks its sole purpose is to knock zeros off the bottom line ("if we're not filling these racks immediately can just cable the ones we need and do the rest later?").
Once the danger of a little knowledge and the danger of no knowledge have been negotiated you end up with what you knew you were going to get before spending 000's on consultants but somehow those in control now think it's acceptable.
Consultants aren't there to give advice that will be taken seriously. They are there as a big baby-blue safety blanket for the management who don't understand what they're managing and don't trust those who do.
What a way to run a business.
The post above, "shoot the accountants", explains much of the workings. Consultants can be blamed (despite their recommendations not being followed) as non-employees of the company when management decisions led to something gone very wrong. From US Space shuttles and California gas pipelines blowing up to massive oil well spills, engineers warned management, and management failed to act properly.
Biting the hand that feeds IT © 1998–2019