If you’re missing data, either because data on your current supply chain is incomplete, or because you’re recommending new lanes, locations or markets, there are many ways you can approach this problem.
𝗘𝘅𝘁𝗿𝗮𝗽𝗼𝗹𝗮𝘁𝗲 𝗳𝗿𝗼𝗺 𝗲𝘅𝗶𝘀𝘁𝗶𝗻𝗴 𝗱𝗮𝘁𝗮
You can take averages of costs at similar warehouses or on similar transport lanes. You can use regression analysis to find cost drivers. This works well if you're growing in a way that's similar to what you've done before.
- Simple extrapolation
Take an overall (or regional) average of what you’ve done before.- E.g. Average cost per warehouse for a region
- E.g. Average cost per mile for a transport lane (see this article for how-to tips)
- This is easy to do, and often good enough for high-level strategic modelling
- Regression Analysis
For more nuanced insights, apply regression analysis to identify cost drivers. This method reveals patterns in your data and can incorporate more variables.- E.g. If you have both very short and very long distance shipments, regression analysis can identify a distance independent cost for the short shipments
- E.g. Regression analysis lets you explore the impact of shipment size as well as distance
Learn more about regression analysis here and here
𝗕𝗲𝗻𝗰𝗵𝗺𝗮𝗿𝗸𝘀
Extrapolation is best used when you are looking at activities similar to the cost data you already have. If you don't have any data, or you're doing something a bit new, industry benchmark costs can be more useful. There are many independent benchmark providers for all types of logistics costs. They can give you an idea of typical costs in the market (usually for a fee).
- Using benchmark data for new markets or activities gives you a good initial idea of the costs you will face for those activities.
- Using benchmark data for your existing operations can identify areas where you may be paying too much right now and may be able to save money
Your company may already have a contract one or more data providers. If not, high-level benchmark data is available embedded in SC Navigator, and more detailed data is available on request.
𝗚𝗼 𝗼𝘂𝘁 𝗳𝗼𝗿 𝘁𝗲𝗻𝗱𝗲𝗿
Both the previous methods give reasonable high-level approximations, suitable for exploring a large range of possible future supply chain configurations. This makes them very good for identifying potential improvements in your supply chain.
You will need real quotes for actual prices before you commit to implementing those improvements. But going out to tender is a big (and slow) activity for your procurement team. You need to narrow it down another way first!
Using extrapolation or benchmarks to find your preferred scenario(s) first means you will get to initial results far faster, and request far less when you do eventually go out for tender.