Production Tracking: 10 Metrics Every Factory Owner Should Watch

Factory owners often receive plenty of updates but still lack a reliable picture of production. A supervisor reports that work is running, a spreadsheet shows a quantity from yesterday, and the dispatch team is waiting for a final date. For an Indian small manufacturer coordinating suppliers, job workers, shop-floor teams, and customer commitments, good production tracking replaces these separate opinions with a small set of current, consistent numbers.

What good production tracking should achieve

Production tracking metrics for small manufacturers should help people act, not simply create more reports. Each metric needs a clear definition, one source of data, an owner, and a decision connected to it. A number that nobody trusts or responds to adds administrative work without improving output.

Begin at the level that matches your operation: pieces per shift for components, kilograms per day for process manufacturing, lots per week for batch work, or job cards per month for project-based production. Compare similar products, machines, and shifts rather than combining unlike work into a misleading average.

1. Planned versus actual output

This is the clearest daily measure of whether production is meeting the plan. Record the planned quantity and good quantity completed for each shift, machine, line, or job. Calculate plan attainment as actual good output divided by planned output, multiplied by 100.

A low result is only the starting point. Capture one reason such as material unavailable, machine breakdown, operator shortage, quality hold, tool setting, or plan change. Over several weeks, the reason pattern is more valuable than the percentage because it shows where management attention is needed.

2. Schedule adherence

Output can look healthy while the wrong jobs are being produced. Schedule adherence measures whether the work planned for a period was completed in that period. Track completed planned jobs or operations divided by total planned jobs or operations.

This metric exposes frequent priority changes, unrealistic plans, missing material, and work carried forward from one day to the next. For mixed production, measure it by job or operation rather than by total pieces. Completing a large easy order should not hide several overdue customer jobs.

3. Work in progress and WIP age

Work in progress (WIP) is material that has entered production but is not yet a finished item. Measure both its quantity or value and the number of days each job has remained open. An ageing view, such as 0-2 days, 3-5 days, 6-10 days, and over 10 days, makes blocked jobs visible.

Rising WIP usually means work is being released faster than the next operation can absorb it. It occupies space, ties up cash, increases handling, and makes shortages harder to see. Review the oldest jobs first and record whether they are waiting for a machine, material, inspection, decision, or outside process.

4. Cycle time and manufacturing lead time

Cycle time measures how long an operation takes from start to completion. Manufacturing lead time measures the full journey from job release to finished production, including waiting between operations. Tracking both prevents an important mistake: improving machine speed while jobs continue to spend days waiting in queues.

Use the median or a normal range for repeat items because one unusual job can distort the average. Compare actual time with the routing or standard time, then investigate repeated differences. Long setup, small batch interruptions, tool search, inspection queues, or unplanned rework may matter more than the machine's running speed.

5. Machine downtime

Record minutes of planned production time lost because a machine could not run, along with a short reason code. Downtime percentage is downtime minutes divided by planned production minutes, multiplied by 100. Keep planned maintenance separate from breakdowns so teams do not avoid necessary maintenance to improve the number.

For a small factory, accurate reason codes are more useful than an advanced calculation based on estimates. Start with breakdown, setup or changeover, no material, no operator, power or utility issue, tool problem, and quality hold. Review the largest recurring loss on critical machines instead of treating every machine equally.

6. First-pass yield

First-pass yield (FPY) is the percentage of units that complete an operation correctly the first time, without rework, repair, or repeat inspection. Calculate it as good units passed first time divided by total units processed, multiplied by 100.

FPY reveals hidden effort that final output alone misses. A batch may eventually meet the order quantity but consume extra labour and machine time along the way. Track FPY at the operation where the defect occurs, with the drawing or specification revision, so corrective action addresses the actual process.

7. Rejection and rework rate

Rejection rate measures unusable output; rework rate measures output that requires additional work before acceptance. Record both by quantity and, where practical, by cost. One rejected high-value assembly matters more financially than several inexpensive pieces.

Use a short, controlled list of defect reasons and add free text only when necessary. Review trends by product, operation, machine, supplier batch, and shift. The purpose is to find repeated causes, not to blame an operator. A stable reporting culture gives management better data than a target that encourages teams to hide rework.

8. Material consumption variance

Material variance compares the standard or BOM quantity for actual output with the quantity really issued or consumed. It captures excess scrap, wrong issues, unrecorded returns, process loss, inaccurate BOMs, and measurement differences.

For example, if a job should use 500 kg for its completed quantity but records 535 kg, the 35 kg difference needs a reason. Do not assume every variance is waste: it may reveal that the standard is outdated or usable material was returned but not entered. Consistent material issue and return records are essential before this metric can be trusted.

9. Labour productivity

Measure good output per direct labour hour for comparable work. Where products vary significantly, use standard hours earned divided by actual direct hours instead of raw pieces. This avoids comparing a complex assembly with a simple component.

Labour productivity is affected by much more than individual effort. Material readiness, machine condition, instructions, batch size, layout, inspection response, and planning all influence the result. Use the metric to improve the system around employees. Analyse overtime separately, because a higher daily output achieved through excessive overtime may increase cost and fatigue.

10. On-time, in-full delivery

On-time, in-full (OTIF) delivery measures orders delivered by the confirmed date and in the complete promised quantity. An order counts as successful only when both conditions are met. Measure successful orders divided by total orders due during the period.

This is where production performance becomes customer experience. Link each missed delivery to its main cause: late material, planning change, capacity shortage, quality problem, outside job work, documentation, transport, or customer change. A production team can complete work on time while dispatch delays the order, so ownership must cross departmental boundaries.

How to build a production dashboard people will use

Do not launch all ten measures at every workstation on the first day. Start with planned versus actual output, schedule adherence, active delays, and quality loss for one line or product family. Use the existing job card process as the source instead of asking supervisors to maintain a separate report.

1. Define each metric. Write down the formula, unit, update frequency, source, and responsible person.
2. Record data close to the event. Operators or supervisors should update production, downtime, and rejection during the shift, not reconstruct them at month-end.
3. Set a baseline. Observe normal performance for several weeks before choosing targets.
4. Review at the right rhythm. Use output, delays, WIP, and downtime daily; review quality, material, productivity, and delivery trends weekly or monthly.
5. Connect every exception to action. Assign an owner and expected resolution date to important delays and repeat losses.
6. Improve data quality visibly. Correct missing and inconsistent entries without allowing reporting work to become more important than production itself.

Start with decisions, not software screens

Choose three decisions the factory owner or production head must make every day. Then identify the minimum facts required to make them confidently. This creates a focused first dashboard and shows where a simple digital production tracking system can remove manual consolidation.

Once the data is reliable, add deeper measures and automatic alerts. The best system is not the one with the most charts. It is the one that helps the team notice a risk early, understand its cause, and act before the delivery date is lost.