← Back to Field Notes

February 10, 2025 · The Bloomfield Team

Why Your Delivery Dates Are Wrong Before the Job Starts

A delivery date gets committed during the quoting process or at order entry. The estimator checks the schedule, makes a judgment call on material lead time and shop load, and types a date into the system. Two weeks later, the job is late. The shop floor takes the blame. The actual failure happened at a desk, before anyone touched a machine.

Roughly 68% of late deliveries in job shops originate from front-office decisions. Material ordered late, capacity overcommitted, lead times underestimated, or customer requirements misread at intake. By the time the job reaches the floor, the delivery date was already unrealistic. The shop is executing on bad assumptions.

The Quoting Handoff

The estimator wins the job with a three-week delivery commitment. The PO arrives. Someone in order entry loads the job into the ERP with a ship date of three weeks from PO date. The estimator originally quoted three weeks from material receipt, which happens five days after order. That five-day gap propagates through the entire schedule and nobody catches it until the job is due.

This handoff failure between quoting and order entry is one of the most common sources of late delivery. The information in the estimator's head about material availability, outside processing lead times, and shop load assumptions rarely makes it into the system in a structured way. The person entering the order works from the PO and the quote document, which typically lists a delivery window without the underlying assumptions.

Material Lead Time Assumptions

Estimators carry mental models of material lead times. 4130 steel round bar, two weeks. 6061 aluminum plate, one week. 17-4 PH stainless, four weeks. These mental models were accurate once and drift over time. Market conditions shift. Supplier inventories fluctuate. The price of nickel moves and suddenly Inconel lead times stretch from six weeks to twelve.

Shops that quote delivery dates based on assumed material availability without confirming stock with their supplier are betting the job on outdated information. The fix takes five minutes: confirm availability and lead time with the supplier before committing a delivery date. Every time. On every job where material is not already in stock.

Capacity Loaded Beyond Reality

The most common scheduling mistake is loading the shop to 100% of theoretical capacity. A five-axis mill that runs one shift has 2,080 available hours per year. Nobody gets 2,080 productive hours. Planned maintenance takes 5 to 8%. Setup and changeover takes 10 to 20%. Unplanned downtime takes another 5 to 10%. Quality holds and rework consume time. Programming and prove-out for new jobs consume time.

Realistic capacity for a one-shift job shop machine is 55 to 70% of theoretical hours. Shops that schedule against 100% of theoretical capacity will miss delivery dates consistently because the math never worked. For a closer look at how scheduling errors cascade, see the scheduling mistakes that kill on-time delivery.

Outside Processing Blind Spots

Heat treatment. Plating. Anodizing. Grinding. NDT. Every outside process adds lead time that the shop does not control. Standard practice is to estimate outside processing at one to two weeks. Reality varies. A plating house running at capacity might take three weeks. A heat treat facility might have a two-day turnaround on Tuesdays and Thursdays and a ten-day turnaround the rest of the week.

The shops that hit delivery dates consistently have standing relationships with their outside processors and know the actual turnaround time for the specific processes they use most. They call before quoting. They confirm before releasing. They do not guess.

The Accumulation Problem

Each of these errors adds a small amount of risk. Material lead time off by three days. Shop load underestimated by one day. Outside processing estimate off by two days. Individually, each gap seems manageable. Stacked together, they create a job that was six days late from the moment the delivery date was entered into the system.

The shop floor cannot recover six days. Overtime might recover one or two. Expediting material or outside processing might recover another day or two. The remaining gap becomes a late shipment, an apology call to the customer, and a note in the OTD report that blames production.

What Accurate Delivery Dates Require

The fix is a system, not a hero. Five inputs need to be confirmed before a delivery date gets committed to a customer.

1. Material availability confirmed with the supplier, with actual lead time documented
2. Shop load checked against realistic capacity, accounting for jobs already scheduled on the required machines
3. Outside processing lead times confirmed with the actual vendor who will do the work
4. Engineering review completed to identify any special tooling, fixtures, or programming that adds front-end time
5. Buffer built into the schedule for variability, typically 10 to 15% of total estimated lead time

Shops that implement this five-point check before committing delivery dates typically see OTD improve by 15 to 25 points within two quarters. The improvement comes from making fewer promises that cannot be kept. The data that feeds these five checks already exists in most operations across ERP records, supplier correspondence, and scheduling systems. The visibility to pull it together at the moment of commitment is what makes the difference.

Late delivery starts with bad data at the front end. Fix the inputs and the outputs follow.