Best method in conducting expansive inventory sampling 2535
Seeker last edited by
Currently, we own approximately 4000 tangible fixed assets (computer hardware systems) at over 400 sites across the U.S. This year we have decided to do a sample count.
I am wondering what the best approach is to get this done in terms of both efficiency and effectiveness. We are considering a geographical cluster approach, but I’m wondering what I should consider in terms sample design, size, selection. I have never had to pull together a sample before and I want to ensure that I have a solid approach that provides a comfortable degree of confidence for the company. But, I don’t even know where to start.
Can someone please provide some advice and direction to me?
ilyq8429 last edited by
First, you need to consider the NBV of each item, I think 60~80% coverage may give you high level comfort. That mean high-value asset must be included in the sample sheet. Then to calculate the quantity.%0AIf I were you, i will conduct a full scope count.%0ASecond, sample site need include all the significant site(whose tangible asset is more that 5% of the total asset,just for example), this is more like control test.
milan last edited by
Hi,%0A Physical Asset Verification %0ASince you have IT equipment in many locations (400) throughout the US, it might not be practical to verify the equipment existence by physical verification.%0ADirect observation is a strong form of evidence, but verifying assets by physical inspection is also costly, time-consuming, and not fully efficient (travel time, distance involved), etc.%0AYou might consider calling the location where the physical asset is located and asking someone from management to send you images of the IT equipment assets on hand. Specifically, you can ask for an image of the serial number affixed on the asset tag that is found on each piece of equipment.%0AMost people own digital cameras and/or have a camera with one built-in so the level of assurance achieved is probably more than acceptable and should give you a good idea if the equipment is still on hand and in use.%0AThe traditional method is to verify the existence of a physical asset by personal observation. However, with technology advancements, I would argue that you can more reliability and increase your scope more efficiently by using alternative approaches and making use of current technologies.%0A Sampling %0AAudit sampling can be a useful procedure to determine the accuracy of asset balances such as equipment or fixed assets like computer hardware.%0AYou can determine the number of assets to include in your sample selection by using a statistical sampling table. %0AA good sampling write-up may be found at:%0Awww dot nao.org.uk/publications/Samplingguide.pdf This resource provides sampling guidelines in plain English and includes tools to help you to determine your sample size based on the confidence level and other criteria.%0AA quick read of the document suggests that you might be well-served to use cluster sampling to select your sample. Regardless of the approach that you choose, I suggest that you make use of the sampling rules to avoid unnecessary auditing. You can easily achieve 80% confidence level with reasonable precision by examining only a small sample of the population.%0AThe document also includes useful guidance so that you can use the built-in data analysis functionality in Excel for sample selection.%0AGood luck and please ask more questions if you need as this forum has no shortage of competent and quite savvy audit professionals.%0AMilan
harrywaldron last edited by
In addition to the excellent suggestions above, some of these links below might help.
It’s important to collect information accurately and to design a standard capture format so that you merge items from each location easily. Sampling and verification of results are also beneficial to ensure you are receiving accurate information from each location. Finally, from a design standpoint, use the ‘keep it simple’ approach with a visual example, as it’s easy to misinterpret instructions. Good luck
http-and-#58;//www.google.com/search?hl=en-and-q=inventory sampling techniques
kymike last edited by
I’ll defer to the IT guys to respond in detail to your questions.
It seems to me that if you are inventorying computer systems in use, you could devise a way to do so electronically. Don’t they all log in to your network regularly? If so, could the log-in be considered evidence of existence? Software exists to inventory applications installed on all computers located on a network, it should also be able to tell you which computers log in regularly and which do not. I would focus my physical testing on those that do not regularly log in.
harrywaldron last edited by
^ Excellent idea kymike … Yes, network topology mapping software would help facilitate this process well where the parent company has access.
One issue I see is that if some of these units are being actively leased out to customers – the parent company will not have any access if the customer has this locked down so that it can’t be seen from the outside (maybe a very brief survey - checklist type form could be used).
In my earlier response, I also thought more along the lines of computer leasing firms where there might be parts stored in 400 ‘warehouse’ locations.