← Back to Field Notes

August 20, 2025 · The Bloomfield Team

8 Red Flags in Your Quoting Process

Most manufacturers know their win rate. Fewer know why it is what it is. The quoting process sits between the RFQ and the purchase order, and everything that goes wrong in that gap shows up later as lost bids, thin margins, or jobs that cost more than the quote predicted. Here are eight signs that your quoting process has problems worth fixing, ranked by the financial impact we see most often.

Red Flag	Typical Annual Impact	Difficulty to Fix
Average turnaround over 3 days	$500K - $2M in lost bids	Medium
No win/loss tracking	Unknown (that is the problem)	Low
Quoted vs. actual cost variance over 10%	$150K - $400K in margin erosion	Medium
One person holds the quoting knowledge	$200K+ risk per departure	High
Material costs from memory	$80K - $200K in pricing errors	Low
No historical job reference	$100K - $300K in missed efficiency	Medium
Requotes exceed 15% of volume	$60K - $150K in wasted labor	Medium
No bid/no-bid criteria	$40K - $100K in misallocated effort	Low

1. Your Average Turnaround Is Over Three Days

Response time is the single largest predictor of win rate in job shop quoting. Manufacturers that respond in two days or less win 35% of bids. At five days, that drops to 12%. Every day over two costs you bids you were qualified to win. The math on this is so consistent across industries that it should be the first number any shop owner looks at when evaluating their quoting process.

2. You Do Not Track Win/Loss by Customer, Part Type, or Estimator

A shop quoting 40 RFQs per month without win/loss data is operating without a feedback loop. Which customers convert at the highest rate? Which part types carry the best margins when won? Which estimator's quotes convert most consistently? Without this data, pricing strategy is based on intuition. With it, you can allocate quoting effort toward the highest-probability, highest-value opportunities and adjust pricing on segments where you consistently lose or consistently leave money on the table.

3. Your Quoted-to-Actual Cost Variance Exceeds 10%

When the difference between what you quoted and what the job actually cost exceeds 10% on a regular basis, the quoting process is either missing cost components or using outdated assumptions. The gap between quoted and actual costs typically comes from three sources: underestimated setup time, outdated material pricing, and missing secondary operations. Each of these is a data problem, and each is fixable when the right historical information reaches the estimator during the quoting process.

4. One Person Holds All the Quoting Knowledge

If your quoting process slows down or stops when a specific person is on vacation, sick, or unavailable, you have a single point of failure in your revenue pipeline. This is one of the highest-risk conditions in any manufacturing operation because quoting cannot wait. RFQs that sit unanswered for a week because the lead estimator is out do not wait for their return. They go to the competitor who responded on Tuesday.

This is also a knowledge loss risk. When that person eventually leaves, the pricing relationships, customer histories, and job-specific knowledge they carry leave with them.

5. Material Costs Come From Memory or Old Spreadsheets

Material prices in 2025 move faster than most shops update their cost tables. An estimator working from a spreadsheet last updated in Q1 could be 15% off on 6061 aluminum, 8% off on 4140 steel, and 20% off on any copper alloy. A single mispriced material on a $30,000 job can erase the entire margin. The fix is connecting the quoting process to current supplier pricing data, either through direct integration with supplier portals or through a regularly updated material cost database that the estimator accesses during the quote.

6. Estimators Cannot Quickly Reference Similar Past Jobs

An estimator looking at a new RFQ should be able to find three to five comparable past jobs within two minutes. If the search takes 20 minutes, or if they rely on memory to recall similar work, the quoting process is running without its most valuable input. Historical job data, including actual cycle times, setup times, material usage, and quality outcomes, is the foundation of accurate quoting. Making that data accessible at the point of quoting is the highest-leverage improvement most shops can make.

7. Requotes Exceed 15% of Your Total Volume

A requote means the original quote had a problem: a missing operation, a wrong material, a changed specification, a pricing error. Each requote consumes estimator time that could have gone to a new RFQ. If more than 15% of your quotes require revision, the root cause is almost always incomplete information during the initial quoting process. The estimator did not have the customer's full spec, the current material price, or the historical context to get it right the first time.

8. You Have No Formal Bid/No-Bid Criteria

Every RFQ that your estimator works on has a cost: typically 1.5 to 3 hours of skilled labor. If your shop quotes everything that arrives without evaluating fit, you are spending estimator time on jobs you are unlikely to win or that do not match your capabilities and margins. A simple bid/no-bid framework based on customer history, part type fit, required turnaround, and target margin saves your estimating team 15% to 25% of their time, which they can redirect to the quotes most likely to convert.

For a deeper look at how to restructure the quoting workflow, see our complete guide to AI-powered quoting.

The Common Thread

Every one of these red flags traces back to the same root: the estimator needed information that was unavailable, outdated, or took too long to find. The RFQ, the customer history, the job records, the material costs, the quality outcomes, the setup data. All of it exists somewhere in the operation. The question is whether it reaches the person building the quote at the time they need it, organized around the specific job in front of them.

Shops that solve this problem see win rates climb, margins improve, and turnaround times compress. All from the data they already had.