CRM for Metalworking: The Complete Guide (2026)
Metalworking pre-sales is heavier than most industries, and opportunities slip away while you are busy putting out fires. I wrote this guide to fix that: the CRM systems I recommend by shop size, what AI genuinely changes in pre-sales, and the pipeline structure I set up at metal companies.
Have you tried to grow your metalworking business by attracting new customers? If so, you know how hard it is. First contacts and follow-ups drown in the chaos of daily email and urgent projects, and promising opportunities quietly disappear.
I wrote about this before:
I wrote about this before:
So, what do you do? Many firms fix this with a Customer Relationship Management (CRM) system. This guide gives you my recommendations by shop size, what AI genuinely changes in pre-sales, the pipeline structure I set up at metal companies, and how I would start this week.
What is CRM, really?
CRM stands for Customer Relationship Management, and it is a function, not a box you buy. It can live on a whiteboard, on cards on a notice board, or in software. The moment you can no longer track your opportunities in your head, the function is asking for a system, and the right question is never "which product". It is: what problem am I trying to solve?
The goal is plain: revenue growth. A good CRM gives every opportunity a place, a status, and a next action, so following up stops depending on memory. For a busy owner, that is the difference between structured new business and hoping you remember the trade-show conversation.
Why a metal shop needs one
Metalworking pre-sales is heavier than most industries: feasibility checks, drawings, certifications, sometimes a site visit, all before a quote can even go out. That work crosses departments, and without a central system, details get lost exactly where the money is.
One thing before the list. When shops compare systems, the conversation drifts to money. Wrong lens. It is about ownership and effort: how much work is it to have this system, to expand it, to integrate it, to maintain it? Or as I ask it in projects: do you need a pickup truck, a shipping truck, a jumbo jet, or a space shuttle? Every tier below is a vehicle that is right at its own size.
My favourites for 2026, by shop size
Small shops (the pickup truck): start with Odoo, Attio, or Twenty. Midsize with marketing in the mix (the shipping truck): add HubSpot. Larger firms (jumbo jet and up): the Microsoft route, or Salesforce.
| System | Best for | Free tier | Open source | AI in 2026 |
|---|---|---|---|---|
| Odoo | Small shops, growth path into ERP | Yes | Yes (Community) | Lead scoring, agents, transcription (Enterprise) |
| Attio | AI-native pipeline, lean teams | Yes | No | AI fields, research agents, call intelligence |
| Twenty | Open source, own your CRM | Self-host | Yes (AGPL) | Agents, chat, MCP support |
| HubSpot | Marketing + sales suite | Yes | No | Breeze assistant and agents |
| Dynamics 365 Sales | Microsoft-ecosystem firms | No | No | Copilot, sales qualification agent |
1. Odoo
Open source Small shops Grows into ERP
Odoo gives you a full CRM in the open: the Community edition is open source and self-hostable, and the hosted One App Free plan runs CRM with unlimited users. The bigger story is the path: the same database grows into quoting, inventory, and ERP when you are ready.
Know that the new AI features live in the Enterprise edition. The Community CRM is still a complete pipeline tool, and for a small shop that wants one system to grow with, this is the strongest starting point on the list.
2. Attio
AI-native Lean teams What I run myself
Attio is what I run myself today. It is the cleanest picture of where CRM is going: AI built into the data itself. Fields that fill themselves, research agents that complete company profiles, call recording that writes qualification notes straight onto the deal.
It is built with tech companies in mind, and a lean metal shop with an office team that lives in the browser will feel right at home.
🔗 attio.com · 🎥 Product demo
3. Twenty
Open source Self-host Build on it
Twenty is my open-source favourite. Free forever when you self-host, a modern interface, and since version 2.0 its own AI agents plus an app platform to model your own objects and workflows. Hundreds of contributors ship weekly releases.
It sits in my open-source tools list for a reason: this is the CRM you own outright, and the one you can extend into exactly your process.
4. HubSpot
Full suite Midsize + marketing Free on-ramp
HubSpot is the suite play: marketing, sales, and service on one CRM. The free tier is a genuine on-ramp, the Breeze assistant is included across plans, and the heavier Breeze agents (prospecting, research, service) come with the paid hubs.
If you want your website, campaigns, and pipeline in one system with one login, evaluate this one.
🔗 hubspot.com · 🎥 In-depth demo (video)
5. Dynamics 365 Sales
Microsoft ecosystem Larger firms Copilot inside
Dynamics 365 Sales fits organizations already deep in Microsoft 365 and Business Central. Copilot is built in, and the Sales Qualification Agent researches and scores inbound leads on its own. Plan for partner support: this is an ecosystem decision, and it pairs naturally with the Business Central story.
Salesforce remains the enterprise standard with the largest app ecosystem, and for most shops under a few hundred people it is a space shuttle for a pickup-truck job.
A few more names that come up in every conversation:
Pipedrive: the classic visual pipeline, rebuilding itself around agents. I ran my own pipeline on it for years.
Zoho CRM: a broad suite where you bring your own AI model.
Salesflare: a small Belgian CRM that fills itself from your inbox and calendar. I used it for years as well, and it taught me what a small office really uses.
Database builders (Airtable, NocoDB, even a disciplined spreadsheet): they can carry a simple pipeline, with the usual trade-off of total flexibility and no guardrails.
What AI actually changes in pre-sales
Three things are real and shipped in 2026:
- Research and qualification agents. An inbound lead arrives, the agent researches the company, scores it against your criteria, and drafts the first reply. Microsoft and HubSpot both ship this today.
- Fields that fill themselves. Attio's AI attributes add a column the AI completes per record: industry, size, fit. Enrichment without a data-entry afternoon.
- Meetings that write themselves down. Call recording that produces structured qualification notes on the deal record, in Attio, Odoo Enterprise, and HubSpot's paid tiers.
CRM and ERP: together or separate?
Your ERP can hold opportunities too, and for larger companies that can be the right call. I often advise keeping pre-sales separate, for one big reason: most opportunities are not customers yet. Mixing maybe-deals into your ERP pollutes the customer database, and ERP screens are built for orders, not for the fast visual triage a pipeline needs.
And when that moment comes, the receiving system deserves the same scrutiny. My ERP comparison covers that side:
The pipeline that works in metal
Standard CRM templates assume software-style sales. Metal pre-sales has longer cycles, feasibility work, and technical review. This is the stage structure I set up, in whichever tool you choose:
- Lead capture: every potential project lands here. Trade-show chat? Lead in. Website enquiry? Lead in.
- Opportunity assessment: does it fit your company, your capabilities, your strategy? Is there enough potential to spend hours on?
- Technical review: engineers check feasibility, materials, certifications, special requirements. Sometimes that means requesting drawings or visiting the site first.
- Quoting: the detailed quote, built on the review. Speed matters here.
- Negotiation: terms, timelines, budget, until there is a basis to work on.
- Closed won or lost: record the outcome and the reason. The lost reasons teach you the most.
Keep it flexible: some deals skip stages, some sit in technical review for a month. And measure expected revenue per stage, conservatively, so the pipeline tells you something true about the coming quarter. For the quoting stage itself, the right software is its own decision:
Or vibe-code your own
Building your own CRM has become a realistic weekend project, and plenty of shops now do exactly that. Tools like Lovable and Bolt generate a working app on a Supabase backend from plain-language prompts: your pipeline stages, your fields, your dashboards, with login and database included. Want a head start on an open core instead? Self-host Twenty and extend it: its app platform exists exactly for that.
The ownership logic applies here too: build where it gives you an edge, use proven tools for the rest. A custom CRM makes sense when your pre-sales process IS your edge. The open stack that carries this pattern:
How I would start, this week
- Pick the vehicle for your size from the favourites above. When in doubt, start with the free tier of Odoo or Attio: you can be live today.
- Set up the six pipeline stages from this guide. Nothing more.
- Import your open quotes and the leads currently living in your inbox.
- Run it for a month with weekly ten-minute reviews. Then decide what you actually need next.
Questions I get about CRM in metalworking
Do I need a CRM if my ERP already has one?
What should a small shop start with?
Can AI really do my follow-ups?
Is an open-source CRM safe for a business?
Can agents run my CRM for me?
The point is
A CRM will not win deals for you. It makes sure no deal is lost to forgetfulness, and in metal pre-sales that alone justifies the effort. Pick by size, not by feature list. Install the pipeline. Keep pre-sales out of your ERP. Let the AI handle the paperwork while your people handle the judgment.
And about my own work: I do not do CRM-selection engagements. But in every factory project, the CRM ends up integrated into an agentic workflow with the rest of the shop, and that integration is exactly what my services cover.
Which stage leaks the most deals in your shop today: capture, follow-up, or quoting speed? Tell me, I read every reply.
