This blog post follows on from the last article, where we looked at the various cost positions that are usually incurred in manual invoice processing. We will now calculate these costs for the sample company Musterfirma AG.
Note: Like the company, the data is fictitious and therefore not to be interpreted as reference data.
Calculation of the costs of manual data entry for invoice extraction
The direct costs of an invoice depend primarily on the speed of typing. In other words, it makes sense to measure the average keystrokes. You can imagine that your employees will perform quite differently. As a consequence, it advisable to have an approximately representative group of employees for this measurement.
For our example, let’s assume that an average invoice with 30 fields is completely entered into the system in 112 seconds. The derivation therefore is as follows: Our average calculation with 30 fields and an average of 8 characters per field results in 240 required keystrokes. If we assume that the average keystroke of your employees is 200, we arrive at exactly these 112 seconds (important note: We assume that no corrections were made for these keystrokes). If an invoice is incorrect and has to be processed again for various reasons, we calculate an average processing time of 6 minutes instead of 112 seconds. With these figures, we can now work.
As mentioned in the previous article, the Musterfirma AG has an annual document volume of 240’000 documents. If we imply the estimate for invoices requiring revision, this means that the company must reckon with approximately 30’000 invoices x 6 minutes + 210’000 x 112 seconds = 572’000 minutes or more than 9’533 hours of effort per year for data capture alone.
As I also noted in the previous article, that the average salary of a manager in accounts payable in Switzerland is about CHF 115’000 and that of clerks is about CHF 70’000. On top of that, you should add at least another 30% for social benefits and other expenses on the part of the employer (attention: Here you should make sure that you do not include costs that you have foreseen in the indirect costs).
If we assume that only the clerks are involved in the data entry work, which is not always true, this means that the approximate direct costs amount to CHF 461’437.77 (70’000 x 1.3 = 91’000/1’880 hours * 9’544 hours), if the clerks of the Musterfirma AG work 40 hours per week and have 5 weeks vacation a year. And these are only the labor costs for data entry with an average of 200 error-free keystrokes per minute or 6 minutes of revision of incorrect entries according to the assumption for the distribution of error-free and incorrect invoices. Basically, there are additional things to consider such as average missed discounts, fees for late payment and transaction costs for supplier rotation due to poor communication. However, we leave these out generously for once.
For the indirect costs, also called overhead costs, we calculate with an additional 20%. These costs therefore amount to CHF 92’287.55.
One cost component that is often forgotten is opportunity costs. In our case, these try to quantify what the monetary loss is if instead realistic alternative activities were carried out, which generated more value than data entry work. In other words, to determine these costs, we must determine the financial value of the alternative activities.
Our assumption is that, on average, 25% more value could be created if the clerks could perform other activities instead of manual work. These activities could include action planning and optimization, coordination with carriers and customers, disposal, assistance with projects or other things.
This results in opportunity costs of CHF 115’359.44.
If we now add all three cost positions together, we receive CHF 669’084.76 per year for the processing of 240’000 invoices. Per invoice, this makes at least about CHF 2.80, taking into account the assumptions made regarding costs. But don’t forget, as mentioned several times, these calculations are based on various assumptions, which do not have to be representative for your company.
In a final article, we will look at the costs of automated document extraction and make a direct comparison with manual data entry.