Addressable Spend Is Not Available Spend

The money was there and it was not available

A team sizing an opportunity assessment found remaining spend in a category it had already worked once. Roughly half the cost in that category had been taken out about a year earlier. The remaining spend was substantial, and a large block of it sat with a single supplier.

Then came the detail that decided the question. Multi year agreements had been sent to that supplier's customer, and nobody on the call could confirm which ones had been signed and returned. The transition from the prior award was also still settling, with material being imported from overseas and supply still stabilizing months after the decision.

The team took the category out of scope for the moment, concentrated on the one where the client itself believed opportunity existed, and routed a separate request, help finding better specifications across all packaging, into its own engineering work rather than folding it into the sourcing wave.

That reads like ordinary scope discipline. The reasoning underneath it is the part worth keeping, because most teams never get to it and commit to the number instead.

Three gates a spend file cannot show you

A spend file is a record of purchases that already happened. It is a good record. It is also silent on every fact that determines whether a dollar in it can move this year.

The first gate is term. End date, auto renewal, notice window. A category under agreement for two more years is not an opportunity, it is a calendar entry.

The second gate is execution, and it is the one that surprises people. A negotiated term you cannot locate is not a constraint you can plan around in either direction. You cannot rely on it and you cannot rule it out. In this case memoranda carrying two and three year terms had gone out and the signed copies could not be accounted for, which means the category's availability was genuinely unknown rather than merely unfavorable. Unknown is the worse state, because a team will treat unknown as open when a savings number is being built.

The third gate is transition state. A category that changed suppliers recently is still carrying unfinished work: qualification, specification confirmation, routing, and the ordinary variability of a new supply chain settling in. Reopening it early trades a known gain for an unknown one and spends supplier credibility you will want later. A prior win that is still transitioning has not finished becoming a result, and that alone is a reason to leave it closed.

There is a quieter fourth gate that caused real confusion on this one. The flexible material had been split into three or four categories whose labels did not clearly distinguish finished bags from roll stock from lidding film. Nobody could say with confidence which bucket held what. A category boundary that does not match how the material is specified, quoted, and bought cannot be sourced as a single thing, and an opportunity estimate built on it is an estimate of nothing in particular. The dollars are real. The category is an artifact of how the data was grouped.

The failure mode follows from arithmetic. Category size is the one number that is always available, so it becomes the number that sets the commitment. The three facts that gate availability each require someone to go find something, and they never arrive before the kickoff deck does.

Put the gates before the kickoff

Require the three facts in writing before a category enters a wave. This is not a judgment call, it is a short list of tasks: locate the agreement, read the term, date the transition. When one of them cannot be answered, that is the finding, and it is worth more than a sizing estimate because it is actionable immediately.

Put recently awarded categories on a hold list with a review date rather than dropping them. A category under contract is not dead, it is scheduled, and the review date is what keeps it from being rediscovered as fresh opportunity in two years by someone who does not remember the first round.

Route adjacent demand to the capability that fits it. A client asking for help finding alternative specifications is describing engineering work with its own deliverable and its own verification. Running it inside a sourcing wave hides both, because the wave reports savings and the specification work produces a changed drawing. They are different products and they should be sold, staffed, and measured separately.

Then track one ratio across waves: targeted spend at kickoff divided into spend that reached a signed award. When that ratio sags, the instinct is to look at negotiation. Look at availability first. Availability failures are diagnosable before kickoff, which makes them the cheapest category of failure a sourcing function has.

Keep the three states distinct while you report. Negotiated savings, implemented change, and verified result are not the same, and a category that produced the first two is still capable of giving back the third if it is reopened before it stabilizes.

What a well run category pipeline reads like

Every category entering a wave carries four stated facts: agreement end date, location of the executed copy, transition complete date, and the buying unit as the supplier quotes it. Categories under contract sit on a dated hold list that someone reviews on a cadence. The ratio of targeted spend to awarded spend is known for the last several waves and the variance is explained rather than averaged. Nobody is estimating savings on a category whose contract status is an open question, and when that status is unknown, closing the question is itself a tracked task with an owner.

Spend an afternoon in the contract file before the next wave

Pull the executed agreements for your largest categories and confirm the term and the signature. Teams that do this usually find two things in the same hour: a category they assumed was locked is actually open, and one they had already put in the plan is not available at all. Both findings change the wave. Neither one requires a single supplier conversation.