← Back to Field Notes

January 21, 2025 · The Bloomfield Team

Building a Quoting Workflow That Scales

A $6 million job shop quotes 30 RFQs per month. One estimator handles all of them. The process works because the estimator has fifteen years of institutional knowledge, a personal spreadsheet with pricing history, and a relationship with every repeat customer. When the shop grows to $10 million and the RFQ volume increases to 55 per month, the process breaks. The estimator is overloaded. Quotes take longer. Win rates drop. The owner considers hiring a second estimator at $95,000 per year.

The hiring solves the capacity problem temporarily. It creates a new one: inconsistency. Two estimators using different reference data and different pricing logic produce different quotes for the same work. The shop now has two quoting processes running in parallel. Neither one is documented well enough for a third estimator to replicate.

A quoting workflow that scales solves both problems simultaneously. It separates the institutional knowledge from the person who carries it, makes that knowledge available to anyone who quotes, and maintains consistency as the team grows.

The Four Components of a Scalable Quoting Workflow

Centralized job history. Every quote the shop has ever built and every job the shop has ever completed should be searchable by geometry type, material, tolerance band, customer, and operations. When a new RFQ arrives, the system surfaces the most relevant comparables automatically. The estimator does not search their memory or their personal files. The system delivers the context. This is the foundation that the AI-powered quoting approach is built on.

Standardized pricing logic. Shop rates, material markup rules, outside processing handling, and margin targets should be codified in the system, not in individual estimators' heads. This does not mean every quote is priced identically. It means every quote starts from the same cost foundation. The estimator applies judgment on top of that foundation to account for customer relationship, capacity constraints, and competitive pressure. The foundation stays consistent.

Structured customer intelligence. Each customer's order history, typical volumes, negotiation patterns, quality requirements, and preferred lead times should be accessible at the moment of quoting. When Estimator B quotes a job for a customer that Estimator A knows well, Estimator B should have the same context. That context includes which prices the customer has accepted in the past, which they have pushed back on, and what their typical order-to-quote ratio looks like.

Feedback loops that close. When a quoted job completes, the actual cost data should flow back into the quoting system. Quoted cycle time versus actual cycle time. Quoted material cost versus actual material cost. Quoted setup time versus actual setup time. These feedback loops are what make the quoting system smarter over time. Without them, the system is a database. With them, the system learns from every job the shop completes.

What This Looks Like in Practice

The RFQ arrives. The system identifies the customer and loads their profile: 47 orders in the last three years, 92% repeat rate on reorders, typical negotiation of 5 to 8% off initial quote, prefers 3-week lead times, AS9100 quality requirements. The system surfaces four comparable past jobs from the shop's history, showing quoted price, actual cost, and resulting margin for each.

The estimator reviews the drawing, confirms the routing, and builds the quote using current material pricing that the system has already pulled from recent purchase orders. Setup time and cycle time estimates come from actual floor data on comparable parts. The estimator adds their judgment on pricing strategy based on current capacity and the customer's profile, and the quote goes out within four hours of the RFQ arriving.

When the second estimator receives a different RFQ from the same customer the following week, they see the same customer profile, the same historical data, and the same pricing foundation. The quotes are consistent. The process is repeatable. The shop can add a third estimator without any loss of accuracy or consistency.

The Scaling Math

A shop quoting 55 RFQs per month with one estimator averaging three days per quote is constantly behind. The queue never clears. Win rates suffer because quotes arrive late. The estimator is working overtime to manage the volume.

The same shop with a structured quoting workflow that compresses the research phase from two days to two hours can handle 55 RFQs with one estimator working normal hours. The $95,000 second estimator hire becomes unnecessary. The existing estimator quotes faster with better information. Win rates improve because quotes arrive sooner. Margins improve because historical data informs every price.

If the shop does eventually hire a second estimator to handle continued growth, that person becomes productive in weeks rather than months because the system carries the institutional knowledge that would otherwise take years to acquire. The knowledge management infrastructure that supports the quoting workflow also supports onboarding, training, and succession planning.

Where to Begin

Start by exporting your complete job history from the ERP: job number, customer, part description, quoted price, actual cost, and key operations. Structure that data so it can be searched by the attributes that matter for quoting comparisons. Connect it to your quoting workflow so the estimator sees relevant history before they start building each new quote.

That single step, making historical job data searchable and accessible during quoting, delivers more value than any other investment in the quoting process. Everything else builds on top of it.