How to Manage a Small Manufacturing Operation Without ERP

Walk into any 30-person job shop or contract manufacturer and ask what software they use. The answer is usually some combination of QuickBooks, spreadsheets, and a whiteboard. Maybe a $40/month inventory tool. The owner has been told for years that what they really need is ERP. NetSuite. SAP Business One. Microsoft Dynamics 365. Sage X3. Each one comes with a $150,000 to $400,000 implementation, a six-month deployment, a full-time admin to maintain it, and per-user pricing that turns punitive past 25 seats.

For a $2M revenue manufacturer with 12 employees, that math does not work. So the owner keeps running on spreadsheets, and the spreadsheets keep failing in slow, expensive ways — inventory off by 30%, work orders that get lost, purchase orders that miss lead times, jobs that ship two weeks late.

Here is the truth nobody in the ERP sales cycle wants to say: small manufacturers do not need ERP. They need ERP-equivalent capabilities, assembled differently. This post is the operator's blueprint for that. The six capabilities every small manufacturer needs, what each one actually requires at SMB scale, how to assemble them without a six-figure ERP project, and the honest line where you do still need a real ERP.

"Small manufacturer" in this post means: 1 to 50 employees, $500,000 to $20 million in annual revenue, discrete manufacturing (not process industry like chemicals or pharma), and not in a heavily regulated traceability regime (aerospace AS9100, medical device FDA 21 CFR Part 820 with full lot-genealogy, defense ITAR). That covers a huge swath of real-world manufacturing — job shops, contract manufacturers, food and beverage at SMB scale, furniture, custom fabrication, electronics assembly, packaging, and dozens more.

If you are bigger than that or regulated, skip to the last section. The blueprint here is for the businesses that ERP vendors do not really want as customers but tell yes to anyway because the deal commission is still good.

Strip an ERP down to what an SMB manufacturer actually uses every day, and you get six functional areas. Inventory management. Bill of materials. Production work orders. Purchase orders. Sales orders. Quality and basic traceability. Each one needs specific things at SMB scale. None of them require a $300,000 deployment.

Three pools, all tracked separately. Raw materials sitting in receiving and the warehouse. Work in progress on the floor — partially built sub-assemblies, kits staged for production. Finished goods ready to ship.

At SMB scale, what you actually need: SKU-level tracking with quantity on hand, location (which bin, which shelf, which line), unit cost, and reorder point. Lot tracking for raw materials that have expiration or version sensitivity. Automatic decrement when production consumes inventory — without this, your inventory drifts 30% within a quarter.

What you do not need at SMB scale: real-time RFID tracking, AI-driven demand forecasting at the SKU-day level, bin-by-bin physical inventory automation with conveyor integration. Those are real things ERPs sell. They are not your bottleneck. Your bottleneck is whether anyone knows how many of part-4421 are on the floor right now.

A small manufacturer with proper inventory tracking — even just a clean digital list with locations and reorder points — outperforms a competitor with NetSuite who never adopted it past basic finance.

A BOM is the recipe for each SKU. "To make one Widget Model A, you need 4 of part-1, 2 of part-2, 1 of part-3, and 15 minutes of assembly labor on Line 2."

What SMB manufacturers actually need: BOMs at the SKU level with quantities, including labor allocations and machine time if relevant. Version control — when you change a BOM (substitute supplier, swap a component), the change is dated and traceable, and existing in-flight work orders use the BOM version that was active when they started. Cost roll-up — the BOM should automatically calculate the standard cost of the finished SKU based on component costs.

What you do not need: 12-level deep BOM hierarchies with sub-sub-sub-assemblies modeled separately, engineering change order workflows with five-stage approval, BOM management as a discipline owned by a dedicated engineering team. Those are real things at large manufacturers. At SMB scale, BOMs that are two or three levels deep with clear cost and version data are sufficient.

A production work order is the instruction to the floor. "Build 50 units of Widget Model A starting Monday, due Friday. Use BOM version 1.4. Run on Line 2. Lot number PWO-2026-0421."

SMB needs: a production work order record per batch with quantity, due date, assigned line/cell, BOM version, lot number, materials reservation against inventory, labor time tracking, status (planned → released → in progress → completed → closed). Materials consumed get logged against the work order. Labor hours get logged against the work order. When the work order closes, finished goods inventory increments.

What you do not need: MRP-driven automatic work order generation with multi-plant constrained scheduling and capacity leveling across 15 work centers. At SMB scale, scheduling fits on a weekly whiteboard meeting and a simple capacity view per line.

Tracking what you buy, from whom, when it is expected, and what it costs.

SMB needs: PO records with vendor, items, quantities, expected receipt date, unit costs, payment terms. Status tracking (ordered → confirmed → shipped → received → invoiced → paid). Receiving — when goods arrive, the PO line items get marked received and inventory increments. Three-way match between PO, receipt, and vendor invoice before payment.

Lead time data is critical. Each part should track its typical lead time from each vendor. When you set reorder points, lead time is what determines them. "Reorder when we have 14 days of usage left" only works if you know lead time is 10 days.

What you do not need at SMB scale: automated MRP-driven PO generation with vendor performance scoring, EDI-based PO transmission with 25 trading partners, blanket-PO release management. Useful at scale. Overhead at $5M.

Customer orders coming in, getting acknowledged, getting promised delivery dates, and feeding into production scheduling.

SMB needs: sales order records with customer, items, quantities, pricing, promised delivery date, status (quoted → ordered → acknowledged → in production → shipped → invoiced → paid). Critical: the link from sales order to production. When a sales order comes in, you need to know whether the finished goods exist in inventory (ship today) or need to be produced (schedule a production work order, communicate realistic ship date). Without this link, sales over-promises and production gets blamed.

This is also where CRM matters. The sales order does not start in production — it starts as a quote in your CRM, gets won, becomes a sales order. The handoff between sales and production is where most small manufacturers drop the ball.

What was made, with what materials, on what date, by whom, with what defects.

SMB needs: lot tracking on finished goods that links back to the production work order, which links back to the BOM version and the source materials lots. Defect logging at QC — what failed, what root cause, what disposition (rework, scrap, deviation). Customer complaint tracking that links back to the lot it came from.

The practical test: a customer calls and says "the units we received on March 12 are failing test X." You need to know in under 10 minutes which production work order shipped those units, what materials lot was used, what BOM version was in effect, and whether any other shipments could have the same issue. That is what traceability buys you.

What you do not need at SMB scale (unless regulated): full electronic batch records with 21 CFR Part 11 signatures, statistical process control charts on every parameter, formal CAPA workflows. Useful in regulated industries. Overkill for a sheet-metal job shop.

Each of the six capabilities is a feature, not a product. The mistake small manufacturers make is buying a product per capability — one inventory tool, one BOM tool, one production tool, one accounting tool — and then watching them fail to integrate. The data does not flow between them, so the staff manually re-enters everything, and within six months the system breaks.

The alternative pattern is to run all six capabilities inside one operations platform, on shared data. Inventory levels are visible to the production module because they are the same data. The BOM lives in the same system as the production work order, which lives in the same system as the customer order. No integrations to maintain. No data sync issues. No "the inventory tool says 40, the production tool says 22" mismatches.

Here is how to map the six capabilities onto a consolidated platform like Deelo:

There is a real cutoff above which a consolidated SMB platform stops being enough. Crossing one or more of these lines means you have outgrown the blueprint and need to evaluate a real ERP:

If you check three or more of those boxes, a real ERP is probably the right call. If you check zero or one, you almost certainly do not need one. At that scale, an integrated operations platform like [Deelo Manufacturing](/apps/manufacturing) covers the working capabilities without the six-figure deployment.

A small manufacturer running on a consolidated operations platform typically pays $300 to $1,500 a month for software depending on team size. The implementation is weeks, not months. There is no dedicated admin. Updates happen automatically.

The same manufacturer running NetSuite or Dynamics 365 Business Central pays $1,000 to $4,000 a month in subscription, $150,000 to $400,000 in implementation costs in year one, and either 0.5 to 1 FTE of internal admin time or an external consultant on retainer for $3,000 to $6,000 a month.

The difference over five years is between $150,000 and $1.5 million. That is real money for an SMB. It is also the gap between hiring a production supervisor and not.

If you are running on spreadsheets and QuickBooks today and feel the pain of broken inventory, missed POs, and orders shipping late, the next step is not ERP. The next step is a consolidated operations platform that gives you the six capabilities above in one place, without the six-figure deployment.

Look at [Deelo Manufacturing](/apps/manufacturing) for the production and BOM side, [Deelo Inventory](/apps/inventory) for stock and PO tracking, [Deelo CRM](/apps/crm) for sales orders, and [Deelo Invoicing](/apps/invoicing) to close the financial loop. Run a 30-day pilot on a single product line. Measure the change in inventory accuracy, on-time shipment rate, and time spent on Sunday-night reconciliation. The numbers will make the decision for you. See [Deelo pricing](/pricing) for what an SMB-scale deployment actually costs.