← Back to Field Notes

April 4, 2025 · The Bloomfield Team

Stop Losing Money on Requotes

A precision machine shop in Michigan tracked every requote over a 12-month period. Out of 480 RFQs, 134 required at least one revision after the initial quote was submitted. That is 28%. Each requote consumed an average of 2.8 hours of estimator time, and the revised quotes carried margins 6 to 11% lower than the originals. The shop calculated the annual cost of requoting at $312,000 in lost estimator productivity and compressed margins.

Requotes are treated as a normal part of job shop life. They should be treated as a systemic failure that has a specific dollar cost and a fixable root cause.

Why Requotes Happen

We have reviewed quoting processes at dozens of manufacturing operations. The same five triggers appear in nearly every case.

Missing or outdated material costs. The estimator uses the last known price for 6061-T6 aluminum bar stock, which might be from a supplier quote dated three months ago. The customer's purchasing team knows the current market price. When the quote comes in high, they push back, and the estimator has to reprice with updated material costs. This single issue accounts for roughly 35% of all requotes in shops that do not have a current material pricing system.

Tolerance misreads on the first pass. The drawing calls for a 0.001" position tolerance on a bolt pattern. The estimator reads it as 0.005" because the GD&T callout is small, the PDF is low resolution, or they are moving quickly through a queue of eight other RFQs. The quote goes out based on the wrong tolerance, the customer catches it in review, and the entire pricing structure changes because the tighter tolerance requires a different machine and additional inspection.

Historical job data that cannot be found or trusted. The estimator knows the shop ran a similar part two years ago. Finding that job record takes 30 minutes of digging through the ERP, and the cost data in the record does not match what the estimator remembers. So they build the quote from scratch, miss a secondary operation that the original job included, and the customer flags the omission after comparing the new quote to their records from the previous order.

Customer scope changes after submission. The buyer comes back and asks for a different quantity, an additional finishing operation, or a faster lead time. This is sometimes unavoidable. But in many cases, the initial RFQ contained signals about potential scope changes that the estimator did not catch because there was no structured intake process to identify them.

Capacity misalignment. The quote promises a four-week lead time. Production reviews it and says six weeks given current load. The requote goes out with a longer lead time, and the customer is already comparing alternatives.

The Margin Compression Problem

Every requote compresses margin in two ways. The obvious way: the estimator adjusts the price downward to win the business after the first quote was rejected. The less obvious way: the estimator spends less time on the revision because the queue has grown while they were reworking old quotes, and the rushed revision misses cost elements that the careful first pass would have caught.

The Michigan shop found that requoted jobs carried an average margin of 22%, compared to 31% on jobs that were quoted correctly the first time. Nine points of margin, compounded across 134 jobs per year, represents a substantial amount of money walking out the door through a process failure that nobody measures.

Fixing the Root Causes

For a deeper look at how quoting connects to the broader operation, see our complete guide to AI-powered quoting.

Material pricing needs a system, not a habit. Current supplier quotes should be accessible to every estimator at the moment they open an RFQ. That means a centralized material cost database, updated whenever purchasing receives a new quote, and flagged when a price is more than 30 days old. Some shops designate one person in purchasing to update the material cost sheet every Monday. Others build a shared drive where supplier quotes are filed by material type and date. The format matters less than the discipline.

Drawing review needs a checklist. Tolerance misreads are concentration failures. They happen when an estimator is working on their sixth quote of the day and moving quickly through features. A printed or digital checklist that requires the estimator to record every GD&T callout before building the price adds 10 minutes to the initial quote and eliminates the 2.8-hour requote that follows a misread.

Historical job data needs to be searchable by geometry and feature, not by part number. Most ERP systems organize job records by customer or part number. That is useful for repeat orders. For new work that resembles past work, the estimator needs to search by material, tolerance range, size envelope, and operation sequence. Making ERP data truly useful for quoting is one of the highest-return investments a shop can make.

Intake process for RFQs should capture scope uncertainty upfront. When an RFQ arrives, the first step is a two-minute review that flags any indicators of potential scope change. Is the quantity marked as "estimated"? Is the drawing revision current? Are there notes suggesting the design is still evolving? If any of these flags appear, the estimator contacts the buyer before quoting to clarify, which prevents the requote cycle entirely.

What This Looks Like in Practice

A fabrication shop in Indiana implemented a structured RFQ intake process and a material cost database over 90 days. No new software. They used a shared spreadsheet for material costs and a one-page intake form that the estimating team fills out before building any quote. Six months later, their requote rate dropped from 31% to 14%. Estimator utilization improved because less time went to revisions. Average margin on quoted work increased by 4 points because the first quote was built on accurate information.

The changes were procedural. The tools were a spreadsheet and a form. The underlying principle was that requotes are a symptom of information gaps at the front end of the quoting process, and closing those gaps before the quote is built is faster and cheaper than rebuilding the quote after it fails.

For shops ready to take this further, custom AI tools can automate the material pricing updates, surface historical job matches, and flag drawing features that historically trigger requotes. The manual process builds the discipline. The AI accelerates the execution.