Cost and Benefit Analysis
for implementing an ERP system
As soon as the top team are confident that they have a reasonable understanding of the new planning process in an overall sense, it is time to carry out a cost and benefits analysis. It should be stressed that this is an analysis not a justification because we will not proceed if the analysis is unfavourable. It should also be made clear that we will re-visit the analysis after the project is complete to ensure we have cashed all the benefits. The benefits stage of the analysis looks at three main areas of potential benefit, increasing sales, reducing costs and reducing inventory.
One of the main focuses of a Business Excellence project is customer service. We find that our "make to stock" Business Excellence companies achieve high levels of stock availability, typically over 99%. Any make for stock company in a competitive situation should be able to estimate the possible extra sales they could capture if they always had their products on the shelf. For other types of business, assemble to order, make to order or engineer to order, the benefits from a Business Excellence project typically come in the form of reduced lead time, a 50% reduction in 12 months is common, and on-time delivery close to 100%. The potential for increasing sales if lead time could be reduced by half and on time delivery improved should be estimated. It would not be realistic to believe that all the potential increase in sales could be realised, so the potential benefit from increased sales should be calculated as the gross margin on a proportion of the potential increase. For the example shown in Figure 1 we have taken half of the potential benefit. Some companies will get increased sales from both the off the shelf and lead time improvements but for most it will be an either/or situation.
The most basic improvement that Business Excellence companies achieve is a stable, managed master schedule out to the cumulative lead time. When a manufacturing department has stable, reliable schedules so that supervision can spend their time managing and improving their processes instead of chasing material and shuffling priorities, there will be a saving in both direct and indirect costs. Idle time or waiting time, if recorded, is typically only the tip of the iceberg. An estimate of the potential direct saving alone should be made and again we will use just 50% added into the project benefits for our example.
Direct material is frequently a large proportion of the cost of sales. There are many potential savings in this area that arise from improvements in master scheduling. If you give your vendors stable schedules they can achieve the same manufacturing cost savings as you. The greater visibility will enable your suppliers to invest in better equipment and to take on value engineering work with you. Single reliable sources of material, rather than multiple but unchecked sources, opens up the possibility of point of use delivery which saves inventory and administration time. We find that purchasing department can typically achieve a 20% reduction in material costs with as little as 3 months forward visibility of requirements. We will again be as realistic as possible and use only half of this estimate of potential savings on purchased material for our example.
Inventory is the last main category of quantifiable savings. Merely replacing a seat of the pants planning system with a faster, computerised system will often result in inventory rising, as the temptation is to react to the messages to purchase material but not always get around to actioning the other messages. The ignored messages will include re-scheduling out works orders and therefore, purchases. If there are overdue items on the schedules, all the other material which depends on the overdue items will be scheduled. Eliminating all overdue items on the schedules alone will significantly reduce raw material and work in progress. When a managed master production schedule is driven by a sales and operations plan, agreed by the senior team, finished unit inventory will be much better controlled so that slow moving stock will be gradually eliminated. The illustration below shows and example.
Click the image on the right to download a .csv spreadsheet which you can import into any spreadsheet program and then insert your own figures.
With inventory there are two types of savings. The cost of holding inventory, including the cost of the storage space and its overheads, inventory management etc., is an annual cost saving. In addition there is a one time capital saving which will offset the one time capital cost of the project. It is reasonable to take half the potential annual saving in inventory holding cost. For the capital saving, a quarter of the capital is a more realistic figure as there may be equipment purchases necessary to realise some of the capital savings described in the chapter on continuous improvement. The benefits analysis table show some typical figures for a company with a £40m p.a. turnover.
Unfortunately there is a cost associated with this benefit so it is now necessary to estimate the cost of the project. First there will be hardware and software costs. At this stage the choice of system may not have been made. It would do no harm to put in a fairly high estimate obtained by talking to any of the software companies. There will be a one time capital cost of the software which will vary enormously depending on the complexity of the software and size of the company, typically from about £7,000 to as much as £40,000 (US$10,000 to US$60,000) per seat / user licence depending on the software and the database licence fee.
Whilst every effort should be made to resist the temptation to customise the package, there may have to be some changes which should be allowed for and some packages may be cheaper for the software but much more expensive to configure. There will also be an ongoing maintenance cost to install the upgrades along with further customising (typically 15% to 20% per annum of the software cost). The programming cost will depend on the size of the company but there could be two man years in the customising and one person on a continuous basis after that. On going
Many companies do not go any further than these costs to arrive at the cost of the project. It would, indeed, be possible to change the current planning system for the new planning system without any additional costs. Unfortunately, such implementations fail to achieve any of the benefits itemised above. All of the benefits come from the second half of the investment.
You have to allow for costs associated with getting the bill of material and inventory record accuracy above the 98% that is vital to get meaningful data from the system. For inventory records, the cost of cycle counting will be on-going but bills of material will generally remain accurate, once the process for maintaining the bills of material has been established, without any significant on-going cost. The same applies to routings.
An allowance should be made for external education and training in the initial phase. As well as the initial cost of education and training, there will be an on-going cost of training new recruits and for adjusting the process as the business changes. An allowance of 25% per annum of the start up training and education cost is a good estimate.
Many companies start out with the intention of not having a full time project leader let alone a full time project team. At this stage it is not necessary to decide whether you will have these full time people but the justification should include them. As a rule of thumb, a project team of three people is necessary for companies with up to 300 employees. One more project team member is needed for every 200 employees. Deciding who the team will be can be delayed until later.
Finally there should be a "miscellaneous" allowance of about 10% of the cost so far. This is one cost that is almost always overspent!
The last stage of the analysis is to bring together the costs of implementing and compare it with the net annual benefits. The pay back period can then be calculated and compared with the company norm. It is very unusual if the pay back is not one of the best the company has seen for a while.
The final figure to calculate is to divide the net annual benefit by 12. This gives the cost of one month delay. This is a powerful motivator to get on with the project, get the best people possible onto the task forces and avoid delays. The cost of delay of £98,500 per month in this example puts the costs into perspective. The cost of the full time project team, for instance, becomes less of a decision. A full time project team will considerably speed up the project invariably and pay for their cost many times over. The use of outside consulting help is a significant and cost effective way to speed up the project.
One of the most common causes of delay is the decision on which software package to use. The fact is that success or lack of it has almost nothing to do with software. There are some basic functionality requirements that are described in the Software Selection Guide most of the better known packages have this functionality now, after that it is down to support and training.
Phil Robinson www.bpic.co.uk