The Numbers Management Should See Every Day

In many manufacturing operations, management sees the numbers — in the weekly recap or monthly report. The numbers are often accurate. But there's one fundamental problem: by the time those numbers are seen, the moment to act has already passed.

A recap is a mirror showing what already happened. Management needs a window showing what's happening now — and better still, signals showing what will happen if nothing changes.

A concrete example: two scenarios, two outcomes

Take a real scenario from a warehouse & delivery operation at an automotive-components manufacturer. Missing stock rises from <1% to 4.5% over two weeks.

Without a system: the missing stock increase is detected at the monthly recap. Goods losses have already been occurring for 3–4 weeks. Customer claims have already come in. Nothing can be fixed retroactively — the loss is documented, not prevented.

With a system monitoring stock accuracy per location daily and comparing it to set standards, the deviation is detected within 2–3 days. The warehouse team has time to investigate — misplacement, input errors, or shrinkage — and intervene before the first customer claim arrives.

The financial difference is significant. In the W&DCS we built, stock accuracy went from >5% missing to <0.5% — and customer claims fell from 10+ per month to zero. Not because the team became more diligent, but because problems became visible while they could still be solved.

What a real-time dashboard actually looks like

The operational metrics hierarchy for manufacturing

The difference between leading and lagging indicators

This distinction determines whether management can act preventively or only reactively.

Monthly recaps consist almost entirely of lagging indicators. A good operational system exposes leading indicators in real time — so problems can be identified before they become recorded losses.

What numbers should be on a management dashboard?

For manufacturing warehouse and delivery operations, at minimum five numbers should be visible every day:

• Stock accuracy per location — not a total aggregate, but per zone/rack so anomalies are visible
• Today's delivery status vs schedule — how many on-time, how many delayed, how many unconfirmed
• Customer claims received today — not a monthly compilation; each claim is a signal
• Aging / dead stock — stock not moving beyond a threshold must be visible, not just in audits
• Admin throughput vs target — if the admin process slows, deliveries are affected too

A common problem: lots of data, few decisions

When a new system is implemented, the result is often a dashboard full of numbers. But not all numbers are decisions. And too many numbers disconnected from context create noise, not clarity.

A good system doesn't display all available data. A good system displays numbers that, when you see them deviate from target, you immediately know what action needs to be taken — and by whom.

The goal of a system isn't to collect data. The goal is to make the right action the easiest choice to take.

Principles that apply to all operations

• Every process that generates a loss when it deviates should be monitored in real time, not weekly
• Metrics seen by a supervisor should differ from those seen by a director — different granularity, different purpose
• Automatic alerts are better than scheduled reports for anomalies requiring a fast response
• Comparison against targets, not just absolute numbers — numbers without context are meaningless
• Trends matter more than snapshots — one bad day might be noise; a 5-day trend is a signal