Analysing Group Results

The only way for an association to know where the industry is headed is if the results are analysed professionally. While analysing results may seem to be simple enough, in reality, there is more than one way to do it depending on what industry the results have been sourced from and what the association intended to do.

A thorough analysis can help to explain the differences between each group. That applies to both individual performance benchmarks as well as broader or general studies. When it comes to measuring individual business performance, it is worth discovering which participant performed the best on the whole which would set a precedent for a best practice if required.

Compare Each Reference Group Against the Other

The first thing you’ll want to do is to set up each group for comparison. Depending on the goals of your association you can compare different reference groups against each other. What this will do is give you a breakdown of the differences between each reference group

Based on what results you get per group for each variable it is possible to compare the result for each group and even between groups. So, you know which are performing the best. It will also help you find out why one group is doing better than the other which is useful information especially when it comes to counselling individual member businesses.

Note: Comparing groups can also be done using a Benchmark Tool. However, it is always best to double check the results before final publishing.

Compare Reference Groups for Each Period

One of the things associations would want to do is to compare the results they get for each reference group for a specific period or more than two periods. You can do this in two ways the first being to compare the results for every period which can be independent of how many participants you have. However, this can only be done if you have enough participants for each period which will ensure a 90% to 95% reliability. So, the smaller your group is, the higher should be the sample.

The other way is to compare panels of participants who have taken part in the benchmark for every period. That way you have the same sample covering each period which makes tracking development easier.

List Best Practices

You will want to define the association’s best practices and then list why it is the best. One such best practice is to call up a member business to provide additional information regarding their performance. You can also offer them a way to grow the business further to meet their goals.

Tip: Always ask businesses for permission to include their information in your report. Doing so is a very important best practice which shouldn’t be overlooked.

Explain the Results

The goal of explaining the results is to define the variables between each period and reference groups. So, by accumulating the results from many different variables, it is possible for an association to draw a number of conclusions for things like performance.

It is important that you are careful when drawing conclusions for something like performance. This type of result isn’t an indicator of how a position is achieved and what a business is doing to maintain it. Though that’s where your analysis comes in which should be included in the finalised market report.

Reporting on Group Results

How can you determine if a particular group has performed well or not for a specified variable? To do this, you need to use a broader median value and a mean value. Most participants will not find the minimum and maximum performance stats to offer much value when comparing themselves to others because these are often subject to various circumstances which can’t be controlled. So, you wouldn’t want to include these in your analysis. Though they certainly make it to the mean which can add somewhat of a bias to the mean figure.

Generally, deciles, median and quartiles aren’t influenced by the max and min values. So, these values are more usable and stable offering a sound basis for an independent analysis.

Tip: The final report, i.e. the one you send over to member businesses should have your comments for each figure to explain what the figure means. It is always good to assume that those reading the results need assistance in understanding it.

Dealing with Mean and Median Values

One of the things you are bound to run into is the mean being higher than the median or vice versa. If you see that the mean is higher than the median then you have a biased mean with higher values in the upper 10% if not more. The mean can also be used for analysis, but outliers may influence it.

If it is the other way around, i.e. the median is higher than the mean, in this case, the mean is low value biased in the lower 10% or more for performance. The mean can still be used for analysis but will be influenced by outliers.

Conclusion

Analysing benchmarks is the most essential part of making sense of all the results. It is only once you have laid the foundation of thorough analysis that the results should be shared with members. Though a private or internal review should be published or individual members seeking help should be assisted to get the most out of all the figures.