A discounting disaster averted at the expense of one's own employment A tale of discounts and process improvement via the magic of Excel, Access and a fair bit of electronic duct tape we imagine. Welcome to Who, Me? "James" is the Regomized reader of record today, and continues the theme of running the risk of doing a job just that little bit too well with an ancedote from the end of the last …
I once worked as a consulting electrical engineer on a large UK airport upgrade many moons ago, one of the problems we had was the timescales. The client wanted to open early and we just couldn't get the fire alarm system in and tested quick enough.
After speaking to the electricians about the kit we were using I realised we could switch to a more expensive per unit alarm which was far faster to fit. I ran it past the architects and they agreed - it would result in an overall reduction in cost too (as time is money with contractors).
Job completed on time.
I was suspended by my firm for a month and given a first and final warning as our contract paid a percentage of the overall final cost and I'd just accidentally saved the client a fortune.
I left 3 months later to take up a job in IT. Clearly I learned nothing!
Yep, cost-plus accounting has a lot to answer for.
I used to work writing user documentation for a software company. But eventually they realised that if the customer looked up the answer to a question in my docs, the company got nothing from that; whereas if they raised a support query, that was billable hours. The job became a lot less satisfying after that.
To be fair I should have known how the job was priced as I was involved in the bid. However even if I'd realised at the time it wouldn't have changed my mind over what to do.
The client may have likely been a one-off customer but the architect who was getting it in the neck from that client was someone we had a solid long standing relationship with and so by helping them we more or less guaranteed ourselves more work in the future. Sure we may lose out a little (still profit) on this contract but profit is profit.
I think the problem was the partners had plans for the slightly fatter cheques they were waiting on, those new company cars would need to wait a few months.
But eventually they realised that if the customer looked up the answer to a question in my docs, the company got nothing from that; whereas if they raised a support query, that was billable hours.
One of the conclusions which Eric S. Raymond completed missed in The Cathedral and the Bazaar was that for closed-source software companies, support is a cost while for open-source companies it is the main or only revenue stream. As a result, the former have a financial incentive to make their software easy to use while the latter have a financial incentive for bugs and other awkwardness.
The company I worked at was "open source", at first, in the sense that anyone with the right software and skills could view the code anytime they liked. But, as my then-boss said, the database had almost 1000 tables and more than 3 million lines of code, poorly (if at all) commented and inconsistently designed, and if anyone seriously has time to wade through all that, good luck to them.
His successor as CEO was less liberal in outlook - and a marketer rather than a coder - and one of the first changes he made was to encrypt all the code. It was shortly after that I left.
Some FOSS organisations do monetise support. But that was only one half of the OP. The other was support being a cost for closed source. It's not if they monetise support. That's why as a system manager I had support contracts with HP & Informix and why, when I was freelance at some of my clients had support contracts with their vendors.
That's why cost+markup agreements are not common. The most common cost+ agreement is cost+fixed payment and usually that fixed payment is small compared to the cost. Of course, I've been in lots of cost+markup projects and they always go overtime and overbudget. It's just not fixable to keep a contractor under a milestone and cost level when they just make money without any problem.
Not providing documentation doesn't just push customers towards buying support packages, it can also be used to sell training. A certain vendor of middleware has a terrible Enterprise Service Bus offering that a client insisted we integrate with. We tried to understand how it worked, but it seemed to be deliberately unintuitive and the training courses were incredibly expensive. We eventually gave up and had to walk away from a multi year, multi million pound contract (there was a lot of internal politics at the client that helped scupper things as well).
Many years sinceupon, I worked for a purveyor of industrial control systems. We'd won a couple of nice big jobbies which everyone, including the client, acknowledged would have significant scope for changes as the projects unfolded. Change wouldn't be in the architecture, just mainly the size. To give the client's beancounters a dense of prediction, the engineering content was agreed as a fixed % of hardware cost. IIRC 50%.
Simple enough and the project continued with both parties entirely happy.
Then the customer realised there was a shedload of installation hardware from all manner of sources still to ordered. As we had been helpful, seemed to have a good buying support and the fact that the customer was in a bind for handling that part himself, we were asked if we'd be up for a further contract to do his bulk materials purchasing - cable, trays, all manner of bits and pieces.
We were quite happy to do that and agreed. Naturally there had to be a contract and one was duly prepared, which more or less mirrored the existing contracts.
Including how variations were handled.
The moment anything got added to the bought out hardware contract - and there was quite a bit of that was the project evolved - there was an automatic 50% added to cover "engineering".
We were delighted.
Dunno about the client.
Anon cos although it was donkeys years back, someonemight recognise it.
It's a well-known fact (though conveniently ignored when it's not your own money you're spending) that contracts are often (usually) won on the bottom line: many contractors price their tender on the basis of the minimum that meets the bid spec, even if that struggles to meet the cost and irrespective of whether it is actually feasible, and makes profit on the inevitable variations.
Before retiring, I could be asked to take part in bid reviews on major (multi million $) contracts and, despite advising that the work wouldn't meet the actual client requirement, was told to keep quiet and just check that the bid ticked the right boxes on paper. I soon learned that was work to keep clear of as I could later be called back to help clear up the mess. Well paid but not worth the hassle. I wish that had been taught at university as I would have been able to spend more time with my family as the kids grew up. We often learn too late that there's a lot more to life that earning more than we actually need.
> Buy the job then, so long as the contract was set up properly, screw then on variations. Then screw them again on the support contract. Then screw them yet again on any upgrades.
My grandfather made and sold airplanes. He knew how to set a number for costs based mainly on weight. But then he sold it under his cost. Because the revenue for repair parts would typically be twice the original cost to make the airplane.
Rule number one of operating system design. Don't fuck up existing applications.
Maaaany years ago, I recall a colleague having a "discussion" with a potential supplier over prices. My colleague designed consoles which back in the day would be row upon row of analogue gauges - you know, where a little mechanical needle moves around over a scale. He'd gone to a supplier of a particular gauge he was interested in using, but couldn't get them to quote "a" price.
See, in our industry, we expect a price for a part. But in the supplier's primary industry (aviation), a part to be built into a new 'plane was considerably cheaper than the same part being fitted as a replacement. Hence, they refused to quote without knowing what the part was going to be used for.
I have to say I'm forever grateful to one university professor I had, who had actually moved from industry into academia. He drilled into us that you never put your name to something that you were not willing to back up 100%. A very important lesson as in our industry (aerospace) you can go to prison if your mistake leads directly to people dying.
I've had managers try to demand I just sign something before. It's usually ended with me informing them of its so important then it requires the signature and if it fails they can be the one go to prison. Amazing how the risk of personal consequences can suddenly change people's minds about cutting corners...
I have to say I'm forever grateful to one university professor I had, who had actually moved from industry into academia. He drilled into us that you never put your name to something that you were not willing to back up 100%. A very important lesson as in our industry (aerospace) you can go to prison if your mistake leads directly to people dying.
One of the very first lectures in my engineering degree included "Physicists discuss their mistakes in follow-up papers. Engineers explain theirs to coroners."
Quote
"Amazing how the risk of personal consequences can suddenly change people's minds about cutting corners.."
Yeah I've had experience of that in the aerospace stuff I've done and the company I currently attend does.
Called over to inspect a pump bearing part, nope its wrong.. as is the 600 they made yesterday, ... mangler comes over with the usual stuff "gotta deliver them tommorrow", said I'm not signing them off as they be reject parts, mangler tries to force my hand, and told him he'll have to sign the parts off so when they come back its his fault, and if they make it onto an aircraft and the crash is traced back to faulty pump parts, its his name on the paperwork......
Amazing how his tune changed to 'can we fix/remake them?'
Anon to protect the guilty ...
At a previous employer we used water for process cooling - so had large quantities of water at temperatures where legionella bacteria can thrive. A similar time ago, a local entertainment venue had a public health issue due to inadequate management of water in the air conditioning, a couple of members of the public died from legionnaire's disease, and a council minion was thrown under the bus to cover for higher ups' refusal to pay for what was needed to avoid the problem.
A colleague told me that each quarter when he went to the factory manager to have the order for water testing/dosing signed, he'd face the usual "do we really need it ?" type questions. Just uttering the name of said entertainment value would immediately get the order signed.
This, of course, is not just IT contracts. Lowest bid school meals contracts seem to end up with the poor kids getting crap food, as the catering companies find ways to meet the obligations without actually giving the kids a decent meal (" nutritional standards" are not the same as "edible meals"). Educational support service lowest bid contracts end up replacing skilled and expert teachers with barely qualified pairs of hands that have had a 20 minute training course. No beancounter is going to retain these staff if they can get around it, or replace them with someone equivalent if they move on/retire. Cleaning contracts end up with the school looking filthy because a significant part of the work was implicit stuff that the cleaners used to just get on and do, but isn't specified in the letter of the contract. e.g. when Hoovering they'd get rid of the dust underneath the bookshelves, just past the edge the of the ((contracted) carpets. And so on. And of course the estimate for the time it takes to do various jobs is always 10% less ( and cheaper) than it it really is.
This, of course, is not just IT contracts. Lowest bid school meals contracts seem to end up with the poor kids getting crap food, as the catering companies find ways to meet the obligations without actually giving the kids a decent meal (" nutritional standards" are not the same as "edible meals").
Smol's Law (named after a colleague of mine): "A professional is someone who knows exactly how bad a job they can get away with. An amateur does the best job they possibly can for the love of it."
To be fair, this is partly the fault of poorly funding the school systems. A relative of mine used to be an architect who did (among other things) school buildings. One story was a school that told him exactly what they wanted and how much it would cost - which was about 25% below the going rate. He rearranged everything to be the cheapest possible, but simply couldn't get their "required" list at the specified price. It finally came down to him unrolling the plans in front of the board, handing them a red marker, and saying "Pick a building to cross out."
Yeah, seen this in many industries. Original contract price will barely cover the cost of the contract, the VOs (Variation Orders) while the contract is in progress is when the profit is made.
Which always makes me wonder, why did the orignal design / architect / plan not take into consideration all these points, which could have saved the client a huge amount of money.
No proper planning / end user feedback / some other cock up? No feedback loop to learn from for the next job being planned for?
Which always makes me wonder, why did the orignal design / architect / plan not take into consideration all these points, which could have saved the client a huge amount of money.
Simple actually. It's not about saving the client money but extracting as much profit as possible from them in any way they can. I learned to stay away from suppliers who feel they need to pack their bottom line with over-priced add-ons.
Because the end client would never accept the real cost. So everyone down the line lies to get the contract.
And while it's smart to try and save money by everyone from the end client to the sub-sub-sub-contractor, there is a relevant old saying: "It's good to save money in business, but you can also save yourself right out business."
"Which always makes me wonder, why did the orignal design / architect / plan not take into consideration all these points"
It's easy to put this down to incompetence or unwillingness. However I think the deeper reason is in the nature of such contracts. In order to specify fully you'd need to complete a good proportion of the design which by itself is part of the tender.
It's just almost impossible to specify precisely what you need in a reasonable sized document that takes a proportionate time to prepare.
In educational contracts, like the ones I listed and others, and I know from these pages and chatting with people in other industries it's the same elsewhere, contracts are written and agreed by people who a) think that they know what the job/project/whatever entails, from a skeletal description and b) are convinced that it can't be as complex and/or costly as they've been told. At least part of this, I'm convinced, is that they are overpaid and underworked, so assume everyone else is, too.
How else do you think do people like Crapita, Sungard, IBM, Deloitte, E&Y, Accenture etc all make their money on their management consulting gigs? It's T&M, contract variations, add-ons, etc. This is how the Passport Office thing overran, this is how Child Support Allowance overran. Management consultants know this gig perfectly well, civil servants with nil commercial experience do not.
And you have to bear in mind that the Civil Service (at least, the bits I'm familiar with) are not renowned for the good pay, working conditions, and unfailing support from politicians looking for someone to scapegoat for their own policy failures. So, especially in certain areas such as IT systems, the people who do understand stuff tend to move on and up - or over to the other side who are paying much much more.
Hence, if you look behind many failed gov. IT projects you tend to find inexperienced CSs who were doing the best they could but were simply outflanked by the "professional liars" (a.k.a. sales and contract people) from the usual suspects; or you find that role was taken by consultants for whom their own future prospects are more important than the success or failure of the project.
The problem is recognised in some quarters, but when you have politicians setting pay for political reasons (e.g. Rishi's pay freeze for you lot while we pocket a nice rise) then the departments concerned can't actually fix it because they don't have the ability to do so. This year, after a decade of "restraint" (a.k.a. below inflation rises), a year of "pay pause (a.k.a. no rise at all), this year departments are capped at a maximum which is well below inflation again.
In my own area, I'm involved in a flagship project, working on safety critical systems, and most sections are under-staffed. Meanwhile, some are leaving us, and going over to work for the contractors for much more money. I'd like to say that I have all the skills and knowledge and experience I should have to be doing the technical assessments I am, but that would be ... slightly optimistic.
We used to make a product in two versions, 'conventional' and (let's say) 'advanced'. Projects were usually specified in such a way that we could submit two bids; one for exactly as described and a second for an alternative way of completing the same overall task. The consultants would assess the offers and put forward recommendations to the end-user.
Our 'advanced' offer was slightly cheaper in terms of installed equipment but generated massive savings for the end-user as the building structure was much simplified..... but that was in another contractor's scope of supply.
You can guess that the 'advanced' equipment was not ordered as both the main contractor and the consultant were paid on the basis of the overall project cost.
Working for a large consultancy company, we had a contract with a large telecoms company to move kit and apps from one DC to another. We seemed to be losing money, so I had a look at the contract. They had a convoluted algorithm which defined the type of servers/apps to be moved and the costs associated - which was produced by a third party and obviously no-one had actually analysed it.
So, I got my handy Exel spreadsheet (no MS Access though) and set up the definitions and then applied it to the servers we had moved or were going to move. It turns out that the telecoms company had applied it wrong and it saved us about £1m - which put us nicely back in profit.
Did I get any thanks or rewards? What do you think? Of course not.
Alan
All the code we write, the services we maintain, the automation we create, and the efficiencies we make - it all goes to make more money for the higher ups, and IT people continue to be seen as a cost they can't wait to reduce and outsource to India.
As someone going through this pain right now, a beer.
Division in other side of planet:
(HQ) Mgt: "You have two developers. Cost saving. Sack one"
Local IT manager: "But these guys already can't keep up. We need more resources"
Mgt: "Have x0,000 Dddelloittteee Dollars for support. Sack one" (fictional company, name made up)
... Meanwhile back at the ranch ...
LIM: "Hmmm, Dddelloittteee no longer respond, it's only been a month"
Mgt: "Oh, you used all your Dddelloittteee Dollars"
Last Local Developer: "I will need 80% pay rise due to increased stress AND time in lieu"
Mgt and LIM: "Faaaark"
Excel was said to reveal that the actual savings to the business was as measured in local currency was -100%. (so, 100% higher costs) Productivity reduced to 25% compared to 1 year ago. Customer satisfaction reduced to "statistically insignificant".
Yep. You automate a load of reporting reducing required headcount by 3 and not even a thank you. This is what I don't get with companies. I did that for the accounts function however I could have done it for billing and other departments. Departments at multinationals that don't want to reduce their own budgets unless forced and good old office politics.
The Access DB was. However once you RTFM and discovered you could connect to 'real' databases Access became a really useful tool. Until the other tools in the dept got the yips at anything Access. I modified about 50 or so Access apps around the company once to use the data from MSQL 4.2 via stored procedures so I could modify the DB structure without having to re-write all the apps. After I moved on someone panicked and got rid of the apps and never managed to replace them before the company did something drastic though I know not what as is been totally subsumed by what was a lot smaller company when I was there.
I agree. Access COULD be a useful tool so long as you understood and accepted certain things.
I used it once to create a simple app that would read a file, generate a few reports, then attach said reports to some emails. Compared to doing it all manually, copying and pasting subsets of data from the file by hand, it worked great. Another time I created a front-end to a Sharepoint site. Considering the network connection at the site kept going down -- at least 1-2X/day on average, lasting anywhere from a few minutes to a few hours -- it was great. I could keep right on recording data, Access would store it locally, and then I could push it all in a batch to the Sharepoint server when the network came back up.
The trick is in knowing when something has graduated from Access and needs to be done as a proper app. A test all too many people fail.
An acquaintance of mine built an entire automated futures trading platform on Access. By the time he left the company, with a fat wad of cash which has kept him in idleness and Porsches ever since, his platform was managing well over £1bn.
I conclude that those who disparage Access are those who don't know how to use it. Dunning-Kruger strikes again.
"his platform (sic) was managing well over £1bn. (p/) I conclude that those who disparage Access are those who don't know how to use it."
And you can use a ride-on lawnmower to do your grocery shopping. If someone says you shouldn't do that, it doesn't necessarily mean they don't know how to use the lawnmower. I use MS Access every day, but I'm pretty careful not to let the management know everything it can do, because then they _would_ do those things.
I'm not sure things graduate from the access front end capabilities. I implemented pivot tables and that seemed to be beyond the limit of most managers understanding. I spent some time isolating the access front end from the DB - as I said all date was read or input via an API of stored procedures so users couldn't randomly lose data in tables of their own creation or delete others. I've long said laziness it the mother of invention Access was the lazy, but fucking useful, way of giving users controlled access to their data - even if the permissions were controlled at the MSql server end. Interestingly no-one complained about not being able to access other peoples data through Access but when I tried to implement a system agreed by management where data (including documents) was restricted to 'owners' and permitted readers the excuses used to access other peoples data when they were on holiday/sick were interesting to say the least even thought there was a clear and easy to implement policy for just these events.
We had someone (or someones) pointing Access to Sybase backend and it caused no end of havoc as it locked whole tables. Not sure if that was fault of Access or ODBC drivers, but the end result wasn't pretty. it is possible that it does something stupid (while thinking it would improve things0 causing lock escalation from page level locks to table locks.
Cluebat was swiftly administered for the guilty for doing that without even asking if that was allowed/good idea.
I worked for a major UK TV company.
Pointed out massive potential liability because they were using several thousand copies of Office software (guess which one) unlicensed. Was ignored. Raised again.
Was told my job was at risk and to not mention it again.
two weeks after I left they got "fined" £15 million from MS.
Karma is sometimes a bitch.
So, I did about a week-long stint at a job for a new startup company. You talk about your amateur hour. I didn't know this until after I started, but they were growing and had leased out a second office in the building. They couldn't move in right away because the landlord was splitting a much larger office in two by throwing up a wall and the contractor hired to do that wasn't done yet. So, the "executives" at the startup just sat around doing nothing despite the move-in date getting ever closer. They didn't bother ordering desks for the new employees they were hiring, they hadn't established Internet service for this new suite, never mind order the networking equipment or even have someone come in and start wiring things up. I also found out they were running some expensive Cisco networking equipment, which they had just plugged into the wall outlet. No UPS or any sort of planning for if the power were to go out.
I took a look at the new suite and asked them where they were thinking about putting the network switch. They pointed to a closet that kind of gives new meaning to the tern closet. One rather thin person would have a hard time fitting in there and closing the door. There was no ventilation unless you opened the door, and you can just image a Cisco managed switch costing a couple grand being stuck in there, overheating, melting, and best case scenario it just fries itself. I decided that I had plenty of other things to concern myself with, so I could fight that particular battle at another time.
After about a week I had finally started to get some things somewhat organized and moving. I had a switch ordered, tracked down the specific wifi AP they insisted on because some person or other in the management decided they liked it, gotten a contractor to come out and start wiring the place, Internet service established... then they gave me the boot.
As an added bonus, while dealing with a technician AT&T sent out to set up service in the new suite -- technically the second tech, since AT&T splits everything into business units and the first tech both wasn't allowed to do certain things on top of being so new he wouldn't have known how even if he was -- we discovered that multiple people had broken into the utility room using a flathead screwdriver to just pry the door lock into the open position since apparently no one knew how to get the key. Then, once in the utility room, saw a brilliant bit of jerry-rigged electrical work where someone wrapped a bare copper wire around the metal spine casing of the fuse box to serve as a ground. Someone even wrote on the wall next to it "Nice ground". All in all, looking back, they probably did me a favor by firing me.
Congratulations for doing some "doing" stuff successfully. We're now going to punish you by removing you from "doing" and making you do management.
I had a friend where this happened to him. He so successful as a programmer, his innate and enjoyed talent, they banned him from programming and forced him to be a manager, something he has no aptitude for or enjoyment of. Why do so many companies do this? Strip talented people away from using their talent and force them to do something they have no wish, aptitude, ability or satisfaction for?
For many companies, the only reward they know is a promotion to management.
Those companies dont usually have a parallel technical career ladder, where the person gets promoted to a more senior development/technical role with increase in pay and benefits.
And eventually Peter principle will strike since that person will be promoted to a job for which neither the skills nor aptitude is present.
This seems to be in every industry. Private or public.*
It has, I think, a number of roots, which include that the senior managers seem to believe;
1) If you're good at doing something you should be (good at) making sure other people do it well too.
2)That management is a higher calling to which you should be adopted
3)That the job you are doing is all the same for everyone no matter how good you are at it so there can't be too many grades within it
4)Your job costs enough as it is, without paying a horny handed oik like you even more money
5) If you re to be paid more for your contribution it must be because you are doing something new, different and superior.
*In schools you get whisked out of the classroom- the job you trained in and love doing, to sit in an office looking at or writing business strategy etc reports. In the NHS you get whisked out of doing clinical work, ditto. It is crazy. You spend j3 or 4-years training as some kind of therapeutic worker. You spend just four or five years doing what you chose to train for- and maybe dreamt of doing since childhood. Then they stick you in an office well away from any patients.
This post has been deleted by its author
This reminds of a colleague, now sadly deceased, who did electronic design for BT, and previous to that, the Post Office. There was a piece of test kit used to check the proper working of telephone lines, and report faults before there were any customer complaints. This was purely an internal development. The "brain" of this kit was based on a Z80 CPU, plus various cards for RAM, I/O, etc. My enterprising pal squeezed the whole lot onto a single card, potentially saving millions when the kit was built and deployed all round the country. The customer (another BT department) phoned my pal one day, to see how work was going. He said "I have redesigned the computer control, and just based on parts cost, it will save you millions".
This caused major ructions, and my pal was called before higher management, to explain his reckless actions. The trouble is of course, that no one could really complain about saving the overall company money. I guess what was really at stake was corporate nest building and budget slurping, and you would have to be some kind of spin doctor genius to show that in a good light.
the trick is to implement the speed up but not to tell anybody. the resulting extra time is then your own. simple and effective solutions implemented by peasants only ever expose the inadequacies of the corporate aristocracy, promptly resulting in the demise of the peasant. the art of office self-defence.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
