How the Service – Profit Chain works
At the heart of Profit Chain management is ‘managing by fact’, which becomes more feasible if you know how much your profit would increase as a result of specific improvements in employee commitment or customer loyalty. Using research to collect the right employee and customer metrics makes it possible to build powerful Profit Chain models which identify how much of the company’s profit comes from various employee and customer behaviours and how much from other variables. Appropriate research will also identify the key drivers of those behaviours and highlight the issues that should be addressed for optimum improvement in company performance.
However, building Profit Chain models is not simple. If it were, everybody would have one! There are two critical success factors in building a working Profit Chain model. The first is collecting reliable measures of the relevant employee and customer behaviours, and that has been the most important factor in our development of this proposal. The second is using the appropriate statistical techniques and interpreting them correctly when developing the model, but since this is academic for most companies, because they haven’t collected the right measures in the first place, this article will concentrate on the first critical factor. There are three criteria that will determine whether the right measures are being collected:
- Asking the right questions
- Using the best data collection methods
- Comparability between employee and customer measures.
Each of these will now be considered in turn.
1. Asking the right questions
There is always pressure to include too many questions, especially on employee survey questionnaires. Everybody has something they would like to ask. From a research perspective it is therefore good practice to judge the proposed questions on the basis of what they will tell us about the employees / customers themselves, especially about their propensity to engage in behaviours that will be beneficial to the company. In this respect the critical factor is how the questions were originally devised.To provide a reliable measure of employee or customer satisfaction, and therefore a valid lead indicator of their future behaviour, the survey process must mimic the way that people make their satisfaction judgement. It is now widely recognised that people make satisfaction judgements by considering whether their requirements (as employees, customers or other stakeholders) have been met. An accurate measure of satisfaction must therefore identify the extent to which the company meets the requirements of its employees and customers. To produce this measure the questions asked must cover the most important requirements that employees / customers have. The only way to achieve this is to allow the employees / customers themselves to determine the questions asked. This is typically done through exploratory research e.g. focus groups, or depth interviews for customers in B2B markets.
2. Using the best data collection methods
Although poor wording can impact on data reliability, by far the main issue is the choice of rating scale. Some questionnaires use a 5 point Likert scale for collecting scores. Verbal scales, where each point on the scale is given a verbal description (e.g. ‘strongly agree’, ‘agree’ or ‘very satisfied’, ‘satisfied’ etc) are ordinal in function. They give an order from good to bad or satisfied to dissatisfied without quantifying it. In other words, we know that ‘strongly agree’ is better than ‘agree’ but we don’t know how much better. Nor do we know if the distance between ‘strongly agree’ and ‘agree’ is the same as the distance between ‘agree’ and ‘neither agree nor disagree’. Therefore, verbal scales have to be analysed using a frequency distribution, which simply involves counting how many respondents ticked each box. It is not statistically acceptable to use means and standard deviations or to apply most multivariate statistical techniques to establish the relationships between variables in the data set – a massive problem for using the data subsequently to develop a Service – Profit Chain model.Interval scales use numbers to distinguish the points on the scale. They are suitable for most statistical analysis techniques such as means, standard deviations and correlations because they do permit valid inferences concerning the distance between the scale points. For example, we know that the distance between points 1 and 2 is the same as that between points 3 and 4, 4 and 5 etc, so it is reasonable to conclude that a score of 4 represents twice the magnitude of a score of 2. For a scale to have interval properties it is important that only the end points are labelled; the labels (e.g. Completely satisfied…….Completely dissatisfied) simply serving as anchors to denote which end is good / bad, agree / disagree etc.
To summarise, interval scales permit far more statistical analysis than verbal scales and are essential for the multivariate statistical techniques used to build causal models such as Service – Profit Chain relationships.
A second issue on rating scales concerns the number of points on the scale. More points yield greater variability, which is better for analytical purposes for two main reasons. First, scales with more points discriminate better between top and poor performers so tend to have greater utility for management decision making and tracking. Second, it is easier to establish ‘covariance’ between two variables with greater dispersion (i.e. variance around their means). Covariance is critical to the development of robust multivariate dependence models such as identifying the drivers of employee commitment or customer loyalty, or establishing the relationship between employee satisfaction and customer satisfaction. 10-point scales are therefore much better than 5-point scales and are not feasible for verbal scales where each point on the scale requires a verbal description.
Comparability with customer measures
To establish reliable links between employee and customer measures, the data must be comparable. This will be determined by the scale used on the questionnaire and the timing of the surveys. Obviously, both surveys should use a 10-point numerical scale. This is easy to implement, even for companies that currently use a verbal scale. Timing, however, can be more difficult.For example, a large retailer collects customer data on a continuous basis with a rolling programme of store surveys, but the employee survey follows the traditional approach of a one-off annual survey. Consequently there is a risk that comparing customer and employee measures could be invalidated if the time difference and seasonality between the two surveys is too great for individual stores. We therefore recommend one month each year when the major customer and employee surveys take place. Companies wanting to track customer satisfaction on a monthly or quarterly basis can usually do so quite adequately with smaller scale tracking surveys in the intervening months.
Summary
To summarise the points made in this article:- The ability to develop workable Profit Chain models is one of the main new sources of competitive advantage to emerge in recent times, but one which has been sadly under-utilised by UK companies.
- To accurately quantify the relationships between employee satisfaction, customer satisfaction and profit, employee and customer surveys must be totally compatible.
- To ensure that the questions asked provide valid measures of employee and customer satisfaction, exploratory research should be conducted with subsequent survey questions based on the factors that are most important to employees / customers.
- Due to the demanding statistical requirements of Profit Chain modelling, the required rating scale is the 10 point numerical scale.
- Compatibility with customer data will also be important in other respects, especially timing. The customer and employee measures that are being compared must be collected at the same point in time.