Every ERP evaluation reaches the same moment: someone asks what it will cost, and the answer that comes back is a license figure plus a services estimate. Both numbers are accurate. Together they describe maybe half of what the project will actually consume.
This is not vendors being dishonest. It is that a quote prices what the vendor delivers, and a budget has to price what the project consumes. The gap between those two things is where most ERP budget overruns live.
Licensing is the easy part
Subscription pricing has made this line predictable. Most modern platforms price on some combination of users, consumption, and module selection, and you can model it three years out without much guesswork.
Two details change the math more than people expect:
- Named versus concurrent users. A warehouse running three shifts pays very differently under each model.
- Consumption-based tiers. Growth is good until it moves you into the next tier mid-contract.
Implementation services carry the variance
Services are quoted as a range because they genuinely are one. The spread comes down to how much of your process is standard and how much is specific to how you actually operate.
Integrations deserve their own line
An ERP that does not talk to your shipping platform, your CRM, or your e-commerce storefront is a reporting tool with extra steps. Scope integrations during selection, not after configuration starts.
The budget conversation is really a scope conversation wearing a different hat.— A sentence you will hear in most discovery workshops
Data is where timelines go to die
Migration is a technical exercise. Cleanup is an organizational one, and it is the part that needs people who know why a customer record has three addresses and which one gets invoiced.
- Inventory what you have, including the spreadsheets nobody mentions in meetings.
- Decide what actually migrates. Historical data does not all need to come along.
- Assign owners per data domain, then give them time that is protected from their day job.
