It’s hard to say, but it should be a lot less than you do. In most organizations, “Tail Spend” is 20% to 40% of total spend that should be managed by Procurement, but isn’t for various reasons.
It includes a whack of spend that includes, but is not limited to:
- one-time buys for a new hire
- short term buys to replace supply lost to a supply chain disruption
- temporary services
- one-time buys for a trade-show or event
- one-time MRO buys for replacement parts
- “small” purchases for recurring orders under a certain volume or cost
But how much of there really deserves to be there? In theory, none of it, but in practice, some of it will always be, but it should be less and less as time goes on … and certainly a lot less than 30%.
Let’s start with the above:
- one-time buys for a new hire should be a standard kit, which shouldn’t change more than once a year, and since volume can be projected based on hiring patterns, any significant spend is good fodder for, and should be, either a sourcing event or a pre-negotiated catalog-based buy
- if proper sourcing was done for a critical part or item, then it should be easy to switch supply to the secondary supplier with only a minor disruption
- temporary services that recur should also be on a master contract that should be strategically sourced
- trade-shows and events costs tens of thousands these days; if you’re holding the event, you should have an RFI to select the most cost effective venue and most of the items you buy are in bulk and should be auctioned or 3-bids-and-a-buy sourced
- replacement parts could be put on a master contract when you buy the equipment that you know will need replacement parts; you can define a max price and have the option to buy or go to a third-party if a third-party alternative comes along in the future and have the spend at least partially managed
- just because volume is only a few thousand or spend is under 100K or 1M does not mean the category shouldn’t be sourced; if there’s 15% overspend that’s 15K that could be captured even in a simple event that might only cost 5K of resource time or 150K that could be captured only in 15K of resource time; and if 2/3 rds of that could be captured by automation (automated auctions, etc.), why not?
The reality is that you should not have very much tail spend.