To: The board of Rezillos Engineering (Rezillos)
From: An Accountant
Date: December 20X8
Subject: Performance reporting and benchmarking at Rezillos
This report evaluates the current performance report used for the annual board review. Next, the customer rating survey method and reporting are examined. Finally, the method of benchmarking to be used for the divisions is evaluated and the divisional benchmarking exercise is completed and initial conclusions drawn.
Performance report for annual board review
The current report has a number of strengths and weaknesses. These will be discussed according to whether the report:
– measures performance towards the overall aims of the company; and
– is well-presented.
The current mission of the group can be broken down into a number of parts:
– The overall objective of the company is to deliver sustainable growth in value to the shareholders.
This is to be done by:
– working in partnership with customers;
– to deliver solutions which are
o     innovative; and
o     value-for-money;
– utilising the skills of the highly-trained workforce.
Beginning with the overall objective of Rezillos, the report does not directly measure shareholder value and so does not report its growth which are the primary goals of the organisation. This could be done directly by economic value added (EVA™) or total return to shareholders. Also, the report is wholly historical and contains no information about the future prospects necessary to judge whether the growth in shareholder value is sustainable. The current report uses period profits and return on capital employed as its main measures of performance. These can suffer from being short term unlike economic value added.
Overall, the report does not give information about the performance of the strategies which are in place to deliver the overall objective. These will require non-financial measures and so do not appear in the report which is dominated by standard accounting information.
There is no measure in the report of how the company works with customers. The company does collect data on interaction with customers at a divisional level although measuring ‘working in partnership’ is a vague term which should be more accurately defined in order to be measured, possibly by measuring the number of joint product development projects undertaken.
An indirect measure of the customer value-for-money is being offered through the gross profit which, when compared to the industry average, gives a partial measure of value but without data on the price/quality mix compared to competitors, it is difficult to be conclusive about this. Revenue growth is given but should be compared to industry growth in order to give an impression of the attractiveness of our offering to the customer.
No measure of innovation is offered in the report either in terms of the number of new products or the revenue or profit generated from them. This appears to be an important fault as an engineering firm such as Rezillos will only be viable in the long term if it continues to innovate.
There is no information on the skills or training done with Rezillos’ workforce and given the emphasis on innovation and the sector in which Rezillos operates, these appear to be important factors.
There are measures appropriate for these strategies available within the divisions (see the benchmarking exercise later). These could provide easy solutions to many of these gaps.
The report treats the divisions in the same way as the group, using profit and comparison to industry average margins and budgets as the main assessment tools. This may not be helpful as the divisions are in different markets and so may not be easily comparable. This problem could be solved by providing the targets set for each of the divisions for the major financial indicators.
It is unusual, given the preponderance of financial data in the report, that there is very little about the assets and liabilities of the company or its liquidity. This may be acceptable if the efficiency of capital use and the danger of insolvency are negligible but this appears unlikely as shareholder value is a key measure and the ROCE is narrow.
In terms of presentation, the data are clear and in a form which would be easily recognisable to those used to reading accounts. However, it is common to provide a narrative commentary with such a report in order to highlight the key features in the report such as major deviations from target or performance well outside industry norms.

Customer survey ratings
In a recent analysts’ meeting, Rezillos has been questioned about its disclosure that it has an average 7·0 customer rating. The average customer rating is correctly calculated as 7·0 from the sample data. However, there are a number of questions which could be raised over the method of calculation and sampling.
The first issue which might be raised is what does ‘average’ mean in this context. The method used for the calculation is the arithmetic mean but average can also be considered to be the mode (most common rating) or the median (the middle value of the sample, which here would be the 5th value). Both of these alternative methods of calculation would give the lower value of 6·0 for the rating.
The calculation method does not take into consideration the size of the account. The rating of the large accounts might be considered more important and so given some extra weighting in the calculation.
A larger area of concern would surround the sample selected for the calculation. There are a number of questions to be raised about this:
1 A confidence interval and level should be supplied for the rating. If the sample is a small one or the confidence level required is high, then the confidence interval might be shown to be unacceptably large.
2 The method of sampling is not disclosed. Ideally, it should be a random sample.
3 There are various ways in which the sample might be biased:
(a) Only customers who are expected to give a favourable response may have been selected.
(b) It could be that only the customers who chose to respond are included in which case the sample will often be populated with those at the extremes of opinion on Rezillos’ service.
(c) The majority of customers in the sample are from Beeland (six of nine) and only one from Teeland. This will not allow the rating to identify divisional performance, which, given each division has responsibility for customer support, may create a bias. In order to do this, random samples from each division should be taken separately.
(d) There are only three customers sampled with an above average account size ($20·5m). It is common in such surveys to ensure that all of the major accounts are sampled.
Overall, the criticism appears justified and in future, Rezillos should consider disclosing more detail of the method of sampling and the sample size in order to build trust with the investing community.

(i) Methods of benchmarking
There are broadly three methods of benchmarking relevant here. The proposed benchmarking exercise is an internal one comparing divisions within the same organisation.
The other methods are external (or competitor) benchmarking where comparison is drawn with competitors. This is valuable in identifying areas where the other companies demonstrate competitive advantage and also areas for improvement with a similar business. However, although this method can suggest areas where Rezillos can catch up with its major competitors, it will not identify how to gain advantage over these rivals
At a practical level, the difficulty with this method is obtaining the information and even if a competitor can be persuaded to share information, it will often only give strategic improvements, not operational ones, as such detailed information is unlikely to be in the public domain.
A third method of benchmarking is functional benchmarking with a world-class company from another business sector. Rezillos could share detailed operational data without the worry of loss of confidential information directly to a competitor. The difficulty lies in translating lessons learned from one industry to another, so it is often done for generic activities such as logistics. Functional benchmarking against a company from outside Rezillos’ industry sector could be challenging as it will require the use of information from another company which will likely use different systems to collect data.
The major advantage of internal benchmarking is the ability to obtain detailed operational information and so to share best practice amongst the divisions. This will show the different divisions the advantage of being part of a larger company and assist in integrating them.
This method of benchmarking suffers the drawback that it will often involve non-financial data whose production is often less robust than the financial systems involving subjective judgements. Also, it will not necessarily identify world-beating performance. Its internal focus may lead the company to ignore competitor performance. However, as a one-off exercise to harmonise and improve the divisions’ performance, it seems to be suitable for Rezillos now.
(ii) Divisional benchmarking exercise
The benchmarking has been completed as follows: