Food Manufacturing KPIs on an Executive Resume: What the Numbers Actually Tell You

Suppose a candidate for a VP of Operations role tells you they lifted OEE from 62% to 81% across three sites. It is a good line. It is also, on its own, close to unreadable.

Operational metrics dominate every senior manufacturing resume I see. OEE, yield, giveaway, scrap, right first time, OTIF, changeover time, unplanned downtime, cost per unit, recordable incident rate. The numbers are usually accurate. What they rarely tell you, unaided, is how much of the movement belonged to the leader and how much belonged to the asset, the capital budget, or the market.

That distinction is the whole of the assessment.

Taking a neglected plant from 62 to 81 is a real achievement. So is holding a mature, well-invested site at 88 and finding another point and a half. They are not the same job, and they do not reliably predict each other.

The first is a turnaround: basic disciplines reinstated, maintenance backlog cleared, a demoralized team given a reason to care. The second is a marginal-gains job: sophisticated, patient, and mostly about systems that already work. I have seen strong turnaround operators struggle badly in mature networks, where their instinct to intervene reads as disruption. The reverse fails too, more slowly and more expensively.

So the useful question is never how far the number moved. It is what condition the operation was in when they arrived and whether that matches your condition.

What got worse?
Real operating change has a cost somewhere. Throughput gains often show up as higher changeover waste for a period, or as pressure on the maintenance budget, or as turnover in a layer that did not want to work the new way. A candidate who can tell you what deteriorated, by how much, and for how long was close enough to the work to know. A candidate for whom everything improved at once is usually describing something they were adjacent to.

Who owned the capital?
An OEE lift delivered alongside a line replacement is a project and procurement story. The same lift delivered with no capital at all is an operating story. Both are valuable. They are completely different hires, and they belong in different businesses. Ask what the capital envelope was, and ask what they would have done with none.

How long did it hold after they left?
This is the single most revealing question I ask, and it is not asked often enough. Improvement that decays within two quarters of a leader's departure was carried by that person's attention rather than by a system. That tells you something important: not that the candidate is weak, but that they may need to be the one in the room permanently for the result to persist. If you are hiring someone to build capability and then move up, that is a problem.

Who else would tell the same story?
Ask for the name of the plant manager, the maintenance lead, or the quality director who lived through it. Strong operators offer these immediately and are relaxed about you calling them. The willingness is itself the signal, well before anyone picks up the phone.

Recordable incident rates, complaint rates, and audit scores do not behave like throughput metrics. They move slowly, and they are sensitive to reporting culture in a way that OEE is not.

A dramatic improvement in recordable incidents over a short period deserves a careful, non-accusatory question about whether reporting practice changed at the same time. Sometimes the answer is that a genuine near-miss culture was built, which is excellent and hard. Sometimes the answer is that people stopped reporting. The candidate's comfort with that question tells you a great deal, and the ones who have actually done the work tend to raise the distinction themselves.

The same applies to audit outcomes. A clean SQF or BRC audit is table stakes at this level. What is worth probing is what the operation looked like in the weeks before it, and whether the site ran that way year-round or assembled itself for the auditor.

Almost every operational number is a joint product of the person and the plant. A few are not, and I weigh these heavily.

Turnover in the layer directly below them.
Not site-wide turnover, which is driven by local labor markets and wage rates that the leader does not set. The retention of their direct reports is a close read on whether people want to work for this person.

Internal promotions made.
How many of the people who reported to them went on to bigger roles, inside the business or outside it? Leaders who develop people leave a visible trail that is easy to verify.

Whether their successor came from inside.
If the business could replace them from its own bench, it would build something. If the business had to go outside, it might still have performed well, but it would not have left capability behind. For a family- or founder-owned operation that cannot afford to run an external search every time someone moves, this is often the most important line of the whole assessment.

I do not treat operational metrics as evidence until I know the starting condition, the capital, the cost elsewhere, and the durability. Until then, they are claims, usually honest ones, that have not yet been made comparable.

It also changes what I put in front of a client. Two candidates can present near-identical numbers and be suited to opposite briefs. Deciding which is which is the work, and it is the reason I would rather present four genuinely comparable people than ten resumes with impressive figures on them.

If you would like a 30-minute conversation about your leadership pipeline or a specific gap, use the button below.