Custom ERP or off the shelf, for a small manufacturer.

Most small manufacturers looking at ERP get pushed toward the same handful of packaged systems, and for plenty of companies that is the right answer. The pitch is that you get a proven product for a monthly fee instead of paying to build something. What the pitch leaves out is that a packaged ERP asks your business to work the way the software works, and the gap between those two is where your team quietly loses its week.

Here is how we think about that decision, including the cases where we tell people not to build anything.

Where the packaged systems actually win

If your operation is close to standard, a packaged ERP is cheaper and faster, and you should buy one. Standard means your quoting, production, inventory and invoicing look roughly like everyone else's in your industry. Accounting rules are the same for everybody. Payroll is payroll.

You also get things you would otherwise pay to build: an update every few months, a support line, a network of consultants who already know the system, and a product that keeps working if any one person leaves. That is real value, and anyone who tells you custom always wins is selling something.

Where it stops working

The trouble starts with the part of your business that is not standard, which is usually the part you are good at. The odd way you price a rush order. The step your best customer requires. The scheduling logic that lives in one person's head because no product has a field for it.

When the software cannot do that, one of three things happens. You change how you work to match the software, which is fine for paperwork and expensive when it touches the thing that makes you money. You pay for customization, which works until an upgrade lands on top of it. Or somebody builds a spreadsheet beside the ERP, which is the most common outcome by far, and the point at which you are paying for a system that no longer holds the truth.

That last one is worth sitting with. If your people run the real schedule in a spreadsheet and enter it into the ERP afterward, the ERP is not running your operation. It is recording it, late.

The seat math changes as you grow

Packaged systems usually charge per user per month. That is friendly when you are ten people and less friendly at fifty, and it is a bill that grows exactly as you do. It also creates a quiet tax on transparency: when every additional login costs money, companies stop giving access to the people on the floor who would benefit from seeing the data, and the information stays with whoever holds a license.

Custom is the opposite shape. You pay to build it, then hosting and maintenance, and adding the eleventh or the fiftieth user costs nothing. Over a long enough horizon the two lines cross. Where they cross depends on headcount and on how much of the packaged system you would have had to bend anyway.

What you own at the end

This is the part that gets overlooked until it matters. With a subscription, you are renting, and the vendor decides the roadmap. They can raise the price, retire the module you depend on, get acquired, or change direction. Your data is yours in principle, and getting it out in a usable shape is a project.

With custom software you own the source code, the database, and the accounts. Nobody can change your terms. We have also watched that ownership show up somewhere owners do not expect it, which is at sale. When we replaced the production system for a plastics manufacturer across three lines, downtime dropped by more than 30%, and the clean records that came out of it helped the owners get a better valuation when they sold the company. That story is here. A buyer doing diligence is buying a business they can understand, and a tangle of spreadsheets is a discount.

The middle path most people miss

This is not actually a binary, and treating it as one is the most common mistake we see. Plenty of companies should keep their accounting package and build the piece that is genuinely theirs: the scheduling, the quoting, the shop floor capture, the customer portal. Then connect the two so the data moves once.

That gets you the standard parts from a product that does them well, and the specific parts working the way your business actually works, without paying to rebuild general ledger from scratch. It is most of the integration work we do, and when the old system genuinely has to go, we replace it in stages rather than all at once.

How to decide, in four questions

• What percentage of your operation is genuinely unusual? If it is small, buy. If the unusual part is how you win business, that part deserves to be built.
• How many spreadsheets live beside the current system? Each one is a place the product did not fit. Count them before anyone demos anything.
• What does the seat count look like in five years? Run the subscription math forward, not at today's headcount.
• Are you likely to sell the business? If yes, clean operational records and systems a buyer can understand are worth more than most owners expect.

When we tell people to buy instead

We are a custom software firm, so read this with that in mind, but we turn down this work regularly. If your processes are standard and your team is small, a packaged system is the right call and we will say so. If what is broken is that nobody has agreed on how the work should be done, software of any kind will just make the disagreement more expensive. And if the real problem is that your existing system is fine but nothing talks to anything, that is an integration project, not a replacement, and it costs far less.

The version of this that goes wrong is a company that replaces a system it understood with a system it does not, because the old one felt dated. Dated is not the same as broken. We wrote about the actual warning signs in five signs your legacy system is costing you money, and about how a replacement can happen without stopping production in how to modernize your ERP without shutting down the business.