New ways of cost benchmarking

Overview

I have been thinking about good ways to do cost benchmarking and that is an integral part of my role in the cost intelligence realm. I noticed throughout the years that cost managers are benchmarking projects using unit rate.

So, how can we use the currently avaialble technology to change the paradigm in this matter?

A more advanced analysis version would be creating a box and whiskers plot for a range of unit rates of all projects lying within the same category.

I think this vision is quite limited and does not show the bigger picture. Cost managers tend to focus on the final account unit rate at completion and use it face value.

The new approach

My journey in data analytics taught me the importance of creating a storyline to understand the wider scale of things.I questioned myself, every cost submission throughout the lifecycle of the project is true for its scope and time parameters.

This means that the different submissions could provide a wide range of number of true unit rates which are true taking into account their scope and time parameters. That made me think, why do not we add 2 columns within every cost submission to understand the timing and the overall asset/building quantity that led to deriving the total cost? You would have repeated quantities with different unit rates. These can be dealt with by applying inflation to the unit rates. Once you have understood the range of unit rates available, you could use the latest unit rate for every change in quantity. Accordingly, you could have a bigger picture for the unit rate corresponding to a certain project.

Direct application

A good example for what I explained above is viaducts, you could have a starting unit rate of £4,000/m2, mid point construction unit rate of £4,500/m2 and a final account unit rate of £5,000/m2. At the start of the project, the agreed programme could have been 36 months with a viaduct area of 3,000 m2. The mid point construction unit rate could be based on a programme of 42 months with a viaduct area of 3,300 m2. The final account unit rate could be based on a programme of 48 months and an updated scope of 3,600m2. This means that for that specific viaduct you have 3 true unit rates for different circumstances which could be plotted using a box and whiskers chart showcasing the range of pricing. The next step that I could think of is deriving the yearly change in unit rate. So, in the above example it would be £1000 (25%) over 4 years or 6.25% yearly.

Having a wide range of projects, you could have a wide range of yearly change of the unit rate. This could enable creating another box and whiskers plot showcasing the range of the yearly change in unit rates. Another functionality could be added by classifying the data collected ( similar areas together (+/- 5%) )

Final Thoughts

The caveat to that approach is that it requires loads of cost data every month/quarter in loads of projects. The data should be classified with loads of parameters added accordingly (scope, base data, any construction related parameters...etc)

What do you think?

#Cost_benchmarking

#Analytics