← Back to Field Notes

Systems

March 7, 2026 · The Bloomfield Team

What Manufacturers Get Wrong About CRM

A $14 million job shop in Michigan bought Salesforce in 2023. $38,000 implementation. $24,000 annual license for eight users. The sales manager built custom fields, configured pipeline stages, migrated contact data. Six months later, two of eight licensed users logged in regularly. The other six had gone back to email, phone notes, and the spreadsheet on the sales manager's desktop.

This story repeats across manufacturing companies of every size. CRM adoption in manufacturing runs 25 to 40%. In SaaS companies, 70 to 85%. The gap exists because CRM software was designed for a type of selling that does not happen in manufacturing.

The Fundamental Mismatch

CRM software was built for companies selling the same product to many buyers through a repeatable process. A SaaS company sells the same subscription to every customer. Predictable stages: qualification, demo, proposal, negotiation, close. Relatively uniform deal size. Product does not change between customers.

Manufacturing sales work differently. A job shop sells a different product to every customer. Every RFQ is a custom project with unique material, tolerances, quantities, delivery requirements. The "sales process" is a quoting process depending heavily on technical knowledge: what the shop's machines can hold, how similar parts ran before, where the difficult operations are. Deal size varies from $800 to $800,000. The relationship between salesperson and buyer is often as technical as it is commercial.

For a deeper look at how these ideas connect across the shop floor, see our complete guide to AI in manufacturing.

When a manufacturing team uses a CRM designed for SaaS selling, friction starts immediately. Pipeline stages do not match the quoting workflow. The contact model assumes one buyer per deal when manufacturing involves a buyer, engineer, quality manager, and sometimes a program manager. Deal tracking orients around revenue forecasting when the team actually needs quoting performance data: win rate by customer, win rate by part type, average turnaround, reasons for lost bids.

Five Things Manufacturers Need That CRMs Do Not Provide

Quoting history tied to customer records. When a buyer calls about a new project, a manufacturing salesperson needs the complete history: what was ordered, quoted, won, lost. Specific parts, pricing, margins, delivery performance. A standard CRM shows deals with dollar amounts and dates. Manufacturing needs a technical record: part numbers, materials, tolerances, quantities, quality outcomes.

Technical capability matching. Evaluating an RFQ means knowing whether the shop can make the part. Can the mill hold the tolerance? Does the shop have the fixturing? Has a similar geometry been run? This lives in the ERP, on the floor, and in the estimator's experience. The CRM connects to none of it.

Quote-to-win analytics. The most valuable data: conversion rate from quote to PO broken down by customer, part type, material, complexity, turnaround time. This reveals where the shop wins and loses and drives intelligent decisions about which RFQs to pursue. CRMs track pipeline stages. They do not track quoting performance because they were not designed for businesses where every sale starts with a custom engineering exercise.

Job performance feedback. A salesperson who quotes four-week delivery needs to know whether the shop delivered in four weeks. A salesperson who won at 30% margin needs to know whether that held or eroded during production. This feedback loop between quoting, production, and results is critical. CRMs do not connect to production data. The loop does not exist.

Repeat order intelligence. Manufacturing customers reorder. 500 valve bodies last year likely means 500 this year. A system analyzing order history and flagging reorder windows would be enormously valuable. CRMs track opportunities. They do not analyze patterns or predict timing because that requires ERP data.

Why the Sales Team Stops Using It

CRM adoption fails for a practical reason: the system requires more effort to maintain than the value it provides. A salesperson spending 20 minutes daily logging activities in a CRM that does not help win more work, does not surface technical information, and does not connect to quoting or production data is spending 20 minutes on a system with zero ROI from their perspective.

The information they need lives in the ERP, estimator's files, quality system, production schedule. The CRM contains none of it. The salesperson maintains two systems: CRM for management reporting, their own notes for actual selling work. Eventually CRM maintenance drops because it provides no return on time invested.

Management concludes the sales team lacks discipline. The real conclusion: the CRM is not useful enough to justify the time.

What Manufacturing Sales Actually Needs

A system starting from data already in the operation: ERP, quoting system, quality records, customer communication history. Organized around the customer relationship and quoting workflow, not a generic pipeline.

When a salesperson opens a customer record, they should see: three-year revenue by part family, quote win rate, average margin on won jobs, delivery performance, quality issues, outstanding quotes, last five conversations including technical discussions, predicted reorder windows from historical patterns.

When an RFQ arrives: the last time similar parts were quoted with pricing and outcome. Comparable jobs from history with actual costs and margins. Current material pricing. Machine availability for the estimated production window. Quality notes and customer-specific requirements from previous orders.

None of this requires manual entry. It already exists in the systems the shop runs. What has been missing is the layer pulling it together in the context of the customer relationship and current selling opportunity.

Building the Right System

The path forward is a custom customer intelligence tool connecting to existing data sources and presenting a unified view of each customer relationship, including technical and production dimensions generic CRMs cannot model.

The sales team uses it because it makes their job easier. Management gets visibility because data generates automatically from production and quoting activity. Quote-to-win analytics become available for the first time. And the $24,000 per year going to a CRM nobody used goes to a system actually reflecting how manufacturing sales work.

The manufacturers that figure this out will sell more effectively with the same team. The ones buying generic CRMs and wondering why adoption stalls will keep losing to competitors who understand that manufacturing sales is a different business.