[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
On 30/06/2021 09:30, Giles Coochey wrote: > > On 30/06/2021 09:09, Michael Everitt wrote: >> On 30/06/2021 08:27, Giles Coochey wrote: >>> >>> >>> On 30/06/2021 08:10, Michael Everitt wrote: >>>> On 29/06/2021 22:45, Simon Waters wrote: >>>>> On Tuesday, 29 June 2021 15:41:52 BST Exwick, Trevor wrote: >>>>>> https://hpe.wd5.myworkdayjobs.com/Jobsathpe/job/Exeter-United-Kingdom/HPC-Ap >>>>>> >>>>>> plication-Analyst_1085763-2 >>>>> The job posting is a little weird. I assume it is supporting the >>>>> Met Office Cray >>>>> and then they mention SAP.... >>>>> >>>>> >>>> Last I heard (from an inside source) that the Met Office's new >>>> supercomputing >>>> provider was .. wait for it ... Microsoft.... >>>> >>>> "good value for money" was the argument I heard they got the >>>> contract ... > On the "good value for money" comment, most decision makers are utterly > useless at calculating the TCO between a subscription based model and a > in-house outright purchase model. They factor in overestimated savings > savings in HR resource, but it means that they lose general in-house > expertise that is often used in multiple places. They factor in End of > Service dates for hardware refreshes, when often hardware is re-used > elsewhere in non-critical environments, or hardware vendors give > considerable discounts in return for older kit and so on. Many tech > suppliers are now moving to Subscription models, which also have high > levels of vendor lock-in, and ensure they make it just about expensive > enough to revert to in-house again to prevent the customer getting out. > And once a customer is sufficiently invested into the lock-in, prices > begin to hike. > Subscription costing. An aside. Loss of in house ability is a critical failure. I often dabbled in 'subscription methods of selling'. Very few companies and most certainly not most HMG establishments can calculate 'real cost or benefit of subscription service or benefit'. Insurance companies can, and they are major selling agent when life risk or major (think burst undersea pipeline) equipment failure could occur. Interesting point. The "Yorkshire Vet" was first to use it in veterinary practice but he lost a lot of money on it as only one farmer took it up, as the sales agreement 'we answer when you call ' for unlimited visits'; caused a lot of call outs to that farmer. So he abandoned subscription selling. I used it to service petroleum and chemical installations. However it was based on one annual visit to check, inspect and recommend, then a 30% discount to parts purchased. We priced it at 'cost plus 10%' for visit (a big loss in money per ONE annual visit; but a much greater return in escaping legal liabilities & court cases , inquiries etc. if an incident occurred). If recommended spares , and refitting was not taken,, then the user not us carried a very big penalty in legal costs and liabilities for any incident. We also 'rented equipment; and serviced and operated it'. Equipment, software, instruments, inspection equipment and 'gold masters', technicians [security cleared] etc. as a service. This was priced at very much higher rates, (gross profit estimate about 160%) with replacement and upgrades built in on a 10 to 15 year life cycle, with *some parts* being replaced at 3 month intervals. This 'consumable spares problem' was what we tried to solve by the 'service offering' as escaping 'very toxic "substances" ' were the biggest user problem with HMG procurement not seeing fit to replace when we could 'just detect wear' which was our criteria for replacement, due to nature of escaping substances. H2S is a very very much less deadly substance than the toxins that establishment handled. The establishment was in the news recently due to an incident in Salisbury. Pricing for service. A) Include the very big insurance premium, a burst undersea gas pipeline is a big expense (we dried undersea pipelines, lots of computer modelling involved, as you are 'guessing' a circumstance perhaps 40 miles away from the instrument point). B) Ensure you have adequate stock of long time to manufacture spares (include carrying cost of money at 20% per annum). C) Train and train and train your operatives (I assume a code check is equivalent in software) Assume a 3 month wage bill per person per year overhead cost. D) Amortize goods supplied over one third of expected service life. E) Test goods extensively. Sometimes a 'trial unit' was tested to destruction. Roughly: 'Service' supply was an annual cost of about 80% of a 'new supply'. However customer's insurance bill could drop by 'hundreds of thousands of USD per year'. Two government sites out of many bought service, most did not. One in three chemical companies bought service. Strangely, the service customers never had a 'critical failure' with loss of life. Others did. -- regards Eion MacDonald -- The Mailing List for the Devon & Cornwall LUG https://mailman.dcglug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/listfaq