I've been in the room when a multi-plant business gets looked at hard: a new owner, a refinancing, a network review, a sponsor deciding which plants carry the next chapter. The first serious question is rarely about strategy. It is “show me how each plant runs.” In a lot of mid-market food and CPG networks, the honest answer is a scramble: an analyst rebuilding throughput from shift reports, a plant manager defending a yield number nobody else can reproduce, and three versions of service level depending on who pulled the report.
That is not a finance problem or a banker problem. It is an ops visibility problem, and it shows up the moment someone outside the building starts asking questions.
If you can't show how each plant runs from one version of truth, the people across the table will build their own version.
This is about operations, not deals
To be clear about what this piece is: we are not M&A advisors, and nothing here is advice on valuation or deal structure. At Three Chain Catalyst we work as operators, not observers, through People, Process, and Systems. We make your next decision obvious. This article is about the operating decision underneath any network or ownership conversation: does each plant have a locked baseline your own team trusts?
That question doesn't only matter when a sale is on the table. The same baseline is what you need to decide which plant gets the next line, which product family moves, whether a co-manufacturer makes sense, or where capital goes next year. Sale readiness is simply the moment the gap becomes visible to outsiders.
What a per-plant baseline actually covers
We keep it to four measures, per plant, built the same way at every plant:
- Throughput: what each line actually runs against what it was scheduled to run, by shift and by week. Not nameplate.
- Yield: what comes out against what went in, with the definition written down so every plant calculates it the same way.
- Changeover: how many, how long, and what drives them, because changeovers are where schedule, capacity, and labor collide.
- Service: whether the plant shipped what customers ordered, when they asked for it, measured from the customer's side.
Four measures sounds simple. The hard part is “the same way at every plant.” In most networks, each plant has its own reasonable definition, its own spreadsheet, and its own story. Rolled up, the numbers don't add. They argue.
Why baselines drift plant by plant
Nobody sets out to build five versions of yield. It happens because each plant solved its own problem. One plant manager built a tracker that works. Another inherited an ERP report nobody fully trusts. A third runs on a planner's workbook, the same pattern as the 50-tab planning file, the hero planner, and the private blend sheet behind lean-to-trim decisions. Each choice is defensible on its own. Together, they mean leadership is comparing plants on numbers that were never built to be compared.
Then a network question arrives (consolidate two plants, move a product family, prepare for an ownership conversation) and the early weeks go to agreeing on the numbers instead of deciding what to do with them.
A baseline isn't a report you build when someone asks. It's how the plant already runs, written down the same way everywhere.
People, Process, Systems, one module at a time
The fix is not a network-wide system replacement, and it is not a diligence binder that goes stale the day it's printed. Your systems already hold most of what is needed. The fix is one owned decision surface per plant, built the same way across the network.
That is what Module Management is for: connected modules for manufacturing configurations, each one turning a stuck decision into a visible number with an owner, a date, and a next action. For a per-plant baseline, the natural starting points are Production & Manufacturing (throughput, yield, line performance), Production Scheduling (changeovers and schedule attainment), and the Executive Dashboard that sits above them so leadership sees every plant on one page. We install one module at a time, typically in about 30 days per module, data dependent.
Around the tool, TCC runs a three-stage process led by seasoned former operating executives: triage and stop the bleed, make the structural fixes so the problem can't re-form, then sustain until it holds without us in the room. At every stage we look at the same three things:
- People: build the definitions with the plant managers, planners, and supervisors who already know where the numbers come from, so the baseline is theirs, not ours.
- Process: set the routine. Who reviews the baseline, how often, and who owns the exception when a plant moves off it. On a regular rhythm, not only when someone asks.
- Systems: one current source of truth fed from the systems you already run, not rebuilt by hand before every board meeting or site visit.
What “ready” looks like
We will not invent a benchmark for your network in this article. The standard we hold in the room is simpler:
- Every plant reports throughput, yield, changeover, and service the same way.
- Anyone on the leadership team can see how a number was built and where it came from.
- A plant manager can explain a variance from the baseline without a dig through old files.
- When someone outside the building asks how a plant runs, the answer already exists. It doesn't get built for the occasion.
- Network decisions start from agreed numbers, so the meeting is about what to do, not which version is right.
When that is true, an ownership or network conversation stops being a scramble. It becomes a review of how the business already runs.
How to tell if this is your network
You don't need outside help to name the symptoms. Ask a short set of questions at the next operations review:
- If someone asked tomorrow for each plant's throughput, yield, changeover, and service, how long would it take to answer, and would every plant measure them the same way?
- Where do those numbers live: in a shared system, or in plant-level workbooks one person maintains?
- When two plants show different yields on similar products, can you tell whether the difference is real or just definitional?
- How much of your last network or capital discussion went to agreeing on the numbers?
- If the person who builds the monthly plant rollup left, who could rebuild it?
If those answers are uncomfortable, you don't have a deal problem. You have a baseline problem, and it is solvable well before anyone outside the building asks.
Where to start
Start with the plant, or the measure, where the network argues most. Sometimes that's a yield definition that differs by site. Sometimes it's changeover data that lives on a whiteboard. Sometimes it's a service number that depends on who pulled the report. A working session is how we decide which module makes your next network decision obvious first.
We don't open with a pitch deck. Bring the plant rollup, or the reason nobody trusts it. Show us how each plant reports today and which decisions are waiting on those numbers. We'll tell you plainly whether there is a module worth starting with, and which one.
If someone asked you tomorrow to show how each of your plants really runs, how long would it take to answer?
Bring the plant rollup.
A working session, not a pitch. Bring the plant rollup (or the reason nobody trusts it), and we will compare notes on what a per-plant baseline would look like for your network.