Building a Production Tracker You Can Trust

Customer service opens the tracker and sees “assembly.” The supervisor knows that only part of the job reached assembly; the remaining pieces are waiting for rework. Neither record shows that inspection still needs to release the finished quantity.

Both people have some of the truth. Neither can safely promise the whole order from the headline status.

A production tracker records the progress of jobs and operations, including quantities, locations, blockers and relevant completion dates. A useful shared view helps the team answer where work stands without reconstructing the story through several calls.

The difficult part is deciding what each status means and capturing the events that keep it credible. A digital screen can be just as misleading as an old whiteboard if partial work, pauses and quality holds are hidden behind one label.

This guide explains what to track first, how to handle split progress and exceptions, and how to pilot a tracker the floor can update and management can interpret.

A trustworthy production view explains quantities and blockers as well as the headline job stage.

Key Takeaways

• Production tracking records what happened. Scheduling sets the intended sequence and timing. Connect the two without treating them as the same function.
• Define job, operation and quantity states before designing the dashboard.
• Partial completion, rework, quality holds and waiting time need explicit handling so a job does not appear further advanced than it is.
• Capture events near the work, retain their history and show when information is stale or a connection has failed.
• Labour and material records can support cost comparisons, but they need agreed capture and costing rules.
• Assess existing tools before building, then pilot one value stream with representative exceptions and measurable acceptance checks.

In this guide:

• Define what the tracker must answer
• Make statuses and quantities meaningful
• Catch missing work and emerging risk
• A worked split-job example
• Capture labour and materials consistently
• Compare tracking tools
• Make updates practical
• Pilot and connect the tracker
• Frequently asked questions
• How OpsMavix can help

1. Define What the Production Tracker Must Answer

Start with the questions that consume time or cause poor decisions. “We want one screen” is a layout request; it does not establish what the records need to explain.

Useful questions include:

• Which accepted jobs have entered production?
• What quantity has reached each operation?
• What is ready for the next step?
• What is blocked, and who is resolving it?
• How much finished output is released for dispatch?
• Which dates need review after the latest progress update?

Distinguish tracking from scheduling

The tracker records starts, stops, quantities and outcomes. The scheduler uses that progress alongside resource availability and remaining work to decide what should happen next.

An operator’s completion event can update the tracker without automatically creating a feasible revised schedule. Likewise, moving a card into tomorrow’s queue does not establish that production started today.

For the planning side, see manufacturing scheduling software for small shops.

Decide the required level of detail

A simple product line may need one job record with a small set of stages. A high-mix job with parallel operations may need operation-level quantities and dependencies.

Choose the detail that supports decisions. Recording every minor activity can burden operators, while one status for a complex job can conceal the quantities that are still waiting.

A tracker can sit within a broader manufacturing execution system. IBM’s MES overview describes a wider scope including production execution, work orders and quality information. A focused tracker need not implement that entire scope on day one.

2. Make Statuses and Quantities Meaningful

A label such as “complete” needs an agreed meaning. Does it mean the operator finished processing, inspection passed or the finished quantity reached stores?

If different teams interpret it differently, the shared view can create false confidence.

Define a useful state model

Not every shop needs all these states. Keep the distinctions that prevent a real misunderstanding.

Retain more than the latest label

Store the relevant event time, operation, quantity and actor or source. Corrections should retain a reason and traceable history.

A status changed at 16:00 might describe an event that occurred at 14:00. Distinguish event time from entry time where that difference matters.

Keep quantity measures distinct

Ordered quantity, started quantity, good output, rejected quantity and released output are different measures. Avoid treating them as interchangeable completion percentages.

Also distinguish waiting work from active work. A job can physically occupy a cell while waiting for tooling, inspection or a material decision.

Preserve specification context

Make the approved drawing or revision accessible against the job. Define what happens if that specification changes after work starts.

A tracker should show the agreed hold or review, rather than letting an obsolete operation remain apparently ready.

3. Catch Missing Work and Emerging Delivery Risk

A tracker cannot show a job it never received. The accepted-order-to-production handoff needs a control separate from progress updates.

Reconcile demand with jobs

Review accepted demand that has no linked job, unless a documented fulfilment route means production is not required.

Also show released jobs without a scheduled operation and jobs waiting on missing information. An unscheduled record should remain visible rather than disappearing because it is absent from a work-centre queue.

Separate lateness from risk

An overdue flag identifies a deadline already passed. An at-risk flag requires an assessment of what remains and whether it can finish in time.

Inputs may include:

• Remaining operations and expected durations.
• Material and specification readiness.
• Resource calendars and committed capacity.
• Waiting and subcontract lead times.
• Inspection, packing and dispatch cutoffs.

If the tracker lacks that information, show the unresolved blocker and need for review. Do not present a confident completion prediction based solely on the current stage.

Make alerts actionable

A useful exception includes the job, reason, owner and next action. “Job red” is less useful than “inspection release pending; quality lead to review before today’s packing cutoff.”

Review stale records too. A job apparently on time but not updated for two shifts may need verification before a customer promise is made.

A production exception should identify what is waiting, who owns the response and which commitment may be affected.

4. A Worked Example: One Job at Several Stages

Consider a hypothetical order for 100 fabricated units following cutting, assembly and inspection. Assume units move individually through the stages and the snapshot describes mutually exclusive positions.

Cutting produced 100 units. Of these, 80 passed its checks and 20 require rework. Assembly has completed 50 of the 80 passed units. Inspection has released 40 of those 50.

What the current snapshot should show

The headline “assembly” would hide both the rework and the release position. “50 complete” would also be ambiguous: 50 have finished assembly, but only 40 are released finished output.

Avoid adding overlapping totals

The cumulative measures are:

• 100 units processed through cutting initially.
• 50 units completed through assembly.
• 40 units released after inspection.

Adding them gives 190, but there are not 190 physical units. They are overlapping measures of progress through different operations.

Use cumulative operation output to analyse progress, and distinct current-position quantities to explain where the units are now. Label both clearly.

Show the next decision

If partial dispatch is allowed, 40 units may be ready for that process. If the customer requires a complete 100-unit delivery, the order remains unfulfilled.

The supervisor still needs a rework plan, assembly capacity and inspection timing for the remaining quantity. A completion percentage alone does not answer whether the promised shipment can leave.

If the shop uses indivisible batches or shared subassemblies, its quantity model may differ. Test the tracker against the actual flow rather than forcing this illustrative model onto it.

5. Capture Labour and Materials Consistently

Progress records can support estimate-versus-actual analysis when time and material movements are linked to the correct work.

That is an additional capability, not an automatic consequence of displaying job status.

Define what a timer measures

Elapsed operation time, operator labour and machine runtime are different measures. A timer left running during a pause may overstate active labour. Two operators working together can create more labour hours than elapsed hours.

Agree how setup, breaks, waiting, multitasking and shared labour are handled. Keep corrections visible.

Odoo’s Shop Floor time-tracking documentation provides one example of operators recording time against work orders. Assess the capture rules and user workflow in the selected tool.

Link material events without duplication

Record issues, consumption, returns and scrap under their agreed meaning. Issuing material to a job does not always mean it was fully consumed.

Where inventory and production are separate applications, decide which owns each movement. One physical issue should not reduce stock twice because both systems independently post it.

The related inventory controls are discussed in why stock never matches the system.

Hypothetical direct-cost comparison

Suppose a job estimate includes eight labour hours at £25 per hour and £300 of material cost.

Estimated direct cost = 8 × £25 + £300 = £500.

Actual recorded labour is ten hours and recorded material cost is £340:

Recorded direct cost = 10 × £25 + £340 = £590.

The difference is £90, or 18% above the estimate. This simplified example excludes machine charges, overheads, subcontracting and other costs.

Investigate whether the variance came from rework, a scope change or an unrealistic estimate. It does not establish full profitability without the remaining relevant costs and revenue.

6. Compare Whiteboards, Spreadsheets and Production Software

A well-managed board can be sufficient for a simple flow. The choice becomes more important when several people need consistent status, quantity and history information.

Check existing products first

Current manufacturing software may already offer suitable work-order tracking. Configuration, better references or simpler screens can sometimes solve the gap.

Compare that option with a dedicated tool and a custom layer using the same jobs and exceptions. A standard product is not necessarily rigid, and a custom system is not automatically easy to maintain.

Keep ownership practical

If building custom, define source access, documentation, account control, data exports and support responsibilities. Hosting and third-party connections remain ongoing dependencies.

A shared application reduces reliance on informal memory only when the team understands its rules and another competent person can maintain it.

7. Make Updates Practical and Data Quality Visible

The people doing the work should help design the update path. That improves the chance that events are captured correctly and promptly.

Test the actual station workflow

At the intended cell, check whether an operator can:

• Identify the correct job and operation.
• View the relevant instruction or revision.
• Start or pause work with an appropriate reason.
• Record good output and rejects without ambiguity.
• Confirm the event saved successfully.
• Correct an error through the authorised process.

Use the actual terminal, tablet or scanner. Accessibility, gloves, shared devices and poor connectivity can change what is practical.

Do not preserve every existing habit

Fit the system to the real manufacturing process while improving weak controls. A habit of reporting completion verbally at shift end may need changing if timely quantity records are essential.

Explain why each required field exists. Remove fields that nobody uses to make a decision.

Handle outages and corrections

Define what happens when the system or connection is unavailable. A temporary record needs later reconciliation so the same event is not entered twice.

Retain correction history rather than forcing users to hide mistakes through another inaccurate entry. Apply permissions appropriate to starts, quantity changes, release and cancellation.

Review reliability routinely

Check a sample of records against the floor during the pilot and periodically afterwards. Track missing updates, stale jobs, contradictory quantities and unresolved exceptions.

The goal is dependable information with visible limits. Routine verification remains useful even when the team trusts the tracker.

Order, production, inventory and reporting connections need clear ownership so one event produces the intended records once.

8. Pilot One Value Stream and Connect It Carefully

Start with a bounded product line or value stream where unclear progress causes repeated work or delayed decisions. Preserve the upstream and downstream references needed to understand its jobs.

Establish the baseline

Measure status-chasing time, missing jobs, stale updates, blocked-work age and completion against clearly defined dates.

Separate causes: material shortages, capacity overload and unrecorded work should not be combined into one generic “late” category.

Define record ownership

The order system may own customer demand, inventory may own physical movements and the tracker may own execution events.

Document how changed quantities, cancellations, completed output and failures cross those boundaries. Connections to inventory and wholesale orders should preserve stable references and handle retries safely.

Agree acceptance checks

Test a normal job, partial completion, parallel work, rework, quality hold, revised specification and cancellation.

Also verify that:

• Accepted demand without a job remains visible.
• Paused work is distinguishable from active work.
• Released quantity differs correctly from processed quantity.
• A stale record is flagged for review.
• A repeated event does not duplicate output or material consumption.
• An authorised correction retains its history.

Expand after the pilot is dependable

Review the result with operators and supervisors. Confirm support, daily exception ownership and training before adding more areas.

A focused pilot may become useful in weeks, but data preparation and integration complexity affect timing. Agree milestones around demonstrated outcomes rather than assuming every factory follows the same rollout schedule.

Frequently Asked Questions

What is a production tracker?

It records the progress and outcomes of production jobs and operations. A useful view shows quantities, stages, blockers, locations and relevant dates, with enough history to explain the current position.

Is it the same as a scheduler?

No. Tracking records actual progress; scheduling determines intended timing and resource use. Connecting them helps planners revise work using current information.

Can one job be at several stages at once?

Yes, where quantities or operations split. Represent the relevant positions rather than forcing the whole job into one misleading status. Distinguish cumulative operation output from current physical quantities.

Will the tracker automatically warn us before a job is late?

Only if it has appropriate risk rules and inputs such as remaining work, readiness and capacity. A due-date field alone can identify overdue work but does not establish a reliable completion forecast.

Can we keep using a spreadsheet?

Possibly, if the flow and controls remain manageable. Review update delays, conflicting records, missing history and the effort needed to track partial work. Those symptoms help establish whether a different tool is justified.

Can it show actual job costs?

A connected implementation can support costing using recorded labour, material and other agreed inputs. Define the capture and rate rules, and distinguish a direct-cost comparison from full job profitability.

How do we encourage operators to use it?

Test the update path at the actual station, provide training and use the records consistently in daily planning. Keep the actions useful and straightforward while retaining the controls needed for reliable data.

Do we need an ERP replacement?

Not necessarily. Existing software may support the required tracking, or a focused tool can connect to it. Assess the wider requirements before deciding whether a broader suite is appropriate.

Where should we start?

Choose a recent job whose progress was unclear. Map its events, quantities and blockers, then pilot the required controls on a bounded value stream with realistic exceptions.

How OpsMavix Can Help

OpsMavix helps manufacturers connect production events to a shared view the team can understand. We start with how jobs and quantities move, then identify which records and controls are missing.

We can help you:

• Map job stages, operation dependencies and quantity states.
• Build practical production tracking workflows.
• Show blocked, stale and missing work with clear ownership.
• Connect labour and material records to the relevant job.
• Link inventory and order information without duplicate posting.
• Create management reporting from the agreed records.
• Pilot one value stream with training and acceptance checks.
• Provide a documented handover and defined support arrangements.

Bring a recent job that looked complete on the board but was still waiting on the floor. We will trace the status, quantities and handoffs to identify what the tracker needs to show and how the team can keep it current.

Book an Operations Leak Audit to map where production visibility breaks down and what to improve first.