Start from what hurts.
Most ERP searches start with a problem, not a product. Pick yours — each page maps the way out.
Six tools that don't talk to each other — and your team is the integration.
Accounting in one program, stock in a spreadsheet, sales in a CRM, payroll and shipping somewhere else — each was a sensible choice when you made it. Now every order touches half of them, and someone re-types the same data at each hop. The tools work; the gaps between them are what costs you.
Running the whole company on Excel worked — until it didn't.
Somewhere along the way, the spreadsheet that tracked orders became the system the company depends on. Nobody chose that; it happened one added column at a time. If checking a number now means asking who has the latest file open, this page is for you.
The stock count never matches the system, and nobody is surprised anymore.
A stock count that disagrees with the system is one of the most common problems we are asked to fix, and one of the most fixable. It is rarely about careless people; it is about records kept in too many places. Here is how the gap forms, and how we close it with Odoo.
The ERP the whole company runs on — and nobody dares touch.
The system still runs, and that is exactly the problem: it runs on one aging server, a set of workarounds, and the memory of whoever set it up. Every change request gets the same answer — too risky. What you are looking for is not just new software, but a way off a system that has become untouchable, without breaking the business on the way.
Three companies, two currencies, one spreadsheet holding the group together.
Each of your companies keeps workable books in its own system. The trouble starts when the group needs one set of numbers: exports, rate conversions, intercompany matching, and a consolidation workbook that takes days and breaks when someone renames a column. That monthly rebuild is the problem this page is about.
Quoted at one rate, invoiced at another, paid at a third.
If your company sells in dollars, buys in yuan and reports in baht, you already know the pattern: every number is right in its own currency and wrong in every other. The margin you approved is not the margin that lands, and month-end reconciliation turns into detective work. None of this means your team is careless — it means the rate moved between systems that each froze it at a different moment.
"Which lot did that come from?" should not take days and a box of delivery notes to answer.
Your factory makes good product, but the paper trail behind each batch stops at the warehouse door. When a customer, an auditor or a supplier asks which lot went where, the answer lives in delivery notes, shift logs and someone's memory. That works until the day the question is urgent.
Every tax invoice gets made twice — once in your system, once for the Revenue Department.
The e-Tax Invoice & e-Receipt mandate arrived after most Thai companies had already chosen their accounting system, so the electronic side got bolted on around it: a provider portal here, an export file there, a spreadsheet tracking what was accepted. The invoices themselves are correct — getting them to the Revenue Department in the right shape is what eats the month. That gap is closable without changing how your accountants think; only the duplicate work has to go.
Every breakdown is a surprise, and every repair is an emergency.
The production plan assumes the machines will run, and the machines have other ideas. One stoppage pulls the supervisor off the line, sends the maintenance team into firefighting mode and puts delivery dates back on the phone for renegotiation — and by the time it is fixed, the next one is already brewing. If maintenance at your plant only ever happens after something stops, this page is for you.
Choosing MRP software is easy. Choosing one your floor still uses in six months is not.
Every MRP demo shows the same factory: clean BOMs, one smooth routing, no rush orders, no rework. Your factory has partial batches, substituted materials and a customer who called this morning. The system worth buying is the one that survives that gap — and this page is about how to tell before you sign.
You quote from gut feel, win the job — and find out at month-end whether it actually made money.
Somewhere in the office there is a costing sheet that was accurate on the day it was built. Since then material prices have moved, specs have changed, and quoting has quietly become experience plus a safety margin. Most months it works — but you cannot see which jobs are leaking until the accounts say so, weeks after the price was set.
The production plan lives on a whiteboard — and by 10am it's fiction.
The plan gets written before the first shift: run order, machine assignments, what ships today. Then a machine stops, a material delivery comes up short, a rush order lands — and by mid-morning the whiteboard describes a factory that no longer exists. The rest of the day runs on shouting, walking and guesswork, and tomorrow it starts again.
Every check gets done. Every audit still turns into a paper hunt.
The checks happen — incoming inspection, in-process measurements, a final sign-off before shipping. But each result lives on a form that goes into a binder, and the binder is where the information stops. When an auditor or a customer asks you to prove what happened to one specific lot, the plant drops real work to go digging.
One quote: three days, two engineers, and a spreadsheet nobody fully trusts.
When products are configured or engineered to order, quoting is real work: options to validate, costs to build up, drawings to check. But when that work runs through email threads and personal spreadsheets, every quote is slow and every quote is a chance to be wrong. Buyers give the order to whoever came back first with a number they could trust.