The number on the ERP quote is the smallest number in the decision. It is the licence, sometimes with implementation hours attached, and it is the number the board approves because it is the number that was written down. The five-year cost is made of lines the vendor does not sell, and the largest of them is one most buyers never write down until year three, when it has already been paid.
This post lays out the five-year total cost of ownership for a mid-market ERP as five lines, says how to size each for a company of one to three hundred people, and names the line that decides the purchase. It is written for the operator who is comparing a suite against continuing to run integrated best-of-breed systems, because that comparison is only honest when both sides carry all five lines.
The subscription is the easy line and it is still misquoted, because it is quoted for today's users and modules. Size it for year five: the headcount you expect, the modules you will add once the first one is in (a company that buys finance and inventory adds purchasing, then projects, then a warehouse module), and the renewal uplift, which in the suite market runs well above inflation. A quote that is flat for five years is a quote for a company that is not growing. We wrote about what a NetSuite renewal actually buys in the context of alternatives; the mechanism applies to every suite.
The implementation estimate is the second number on the quote and the first one that is wrong, because the vendor sizes it from a demo of your business rather than from your business. The real cost is set by three things the vendor cannot see from a demo: how many of your processes are mapped, how clean the data is that has to move, and how much of your current estate has to be re-integrated. We have argued that ERP implementation cost is decided before you choose a vendor, and the five-year view is where that shows: a company that walks in with maps and clean customer and product records pays a fraction of what a company that walks in with a whiteboard pays, from the same vendor, for the same product.
Size it honestly by adding the remediation you know you need to the vendor's number, and add the internal time. Implementation consumes your best operations people for six to twelve months. Their salary is on line two whether or not anyone writes it there.
Moving the data is not the cost. Fixing it so it can move is. Every mid-market company has customers created three times, products with four codes, open orders that reference items that no longer exist, and a ledger whose chart was designed by whoever was there in the first year. The ERP migration cost is dominated by the remediation, and the honest way to size it is to run the extract once, count the exceptions, and multiply by the time a person takes to resolve one. Most companies are surprised by the count and not by the rate.
This line has one property the others lack: it can be paid before the purchase, at leisure, and it is worth the same whichever ERP is chosen, or none. Remediating the customer and product masters now is a cost you carry into the decision rather than out of it.
The promise of a suite is that integration goes away. It does not. The CRM the sales team will not give up, the e-commerce platform, the shipping carriers, the bank, the payroll provider, the customer portal, the industry-specific tool the suite does not replace: each one still has to be joined to the ERP, and each join is built, monitored and rebuilt at every upgrade for five years. Best-of-breed estates carry more joins and know it; suites carry fewer and pretend otherwise. Size line four by listing every system that survives the purchase and pricing a join per system per year, including the person who owns the join when it fails.
The largest line, and the one nobody writes down, is the cost of the business changing to fit the system, and of the system being changed to fit the business, for five years. Customization that has to be re-applied at each upgrade. Reports rebuilt because the standard ones ask the wrong question. Training every time a module is added or a process is redesigned. The consultant on retainer because nobody in the building can change a workflow. The month-end that takes longer for a year because the new process has exceptions the old one absorbed silently.
This line is where the suite versus best-of-breed comparison is actually decided, and it points in different directions for different companies. A company whose processes are standard and whose estate is small pays little here on a suite and a lot on a best-of-breed estate, because the joins are the change. A company whose processes are specific, whose vertical the suite half-fits, and whose estate already works pays a great deal here on a suite, because the change is the business bending to the software's idea of it. Neither answer is general. The line has to be sized for the company in front of you, and it usually cannot be sized from a demo.
For a mid-market operator over five years, a rough shape from the estates we have costed: the licence is a fifth to a third of the total; implementation plus migration is another third, most of it remediation and internal time; integration and change share the remainder, with change the largest single line for any company with specific processes. The vendor quote covers the first line and part of the second. The board approves a number that is, on this shape, a third of what will be spent.
None of this is an argument against buying an ERP. It is an argument for buying it with all five lines on the page, sized for your company, and for noticing that lines three and four can be paid down before the purchase and carried into it. A company that remediates its masters and draws its integration map first buys the ERP cheaper, later, or discovers it does not need one this cycle, and any of those three is a better outcome than the one the quote was leading to.
Discovery, paid for only if you proceed, produces the five lines for your company from your own data: the extract and exception count for line three, the surviving-system list for line four, the process maps that size lines two and five. Where the answer is to build the foundation first, integration and remediation on the systems you own, we give a guaranteed estimate and absorb the difference if we are low. Where the answer is a suite, the five-line view is the document you take into the negotiation, and it changes the negotiation.
The ERP Readiness Review is a 90-minute working session plus a written scorecard across master data quality, documented exceptions, integration scope and data ownership. Fixed scope, no obligation. Or see how we approach it.
More on the same problem:
Mid-Market ERP Readiness: What to Fix Before You Evaluate a Single Vendor (article)
Why ERP Implementations Fail in the Mid-Market, and the Three Decisions Made Before the Kickoff. (article)
Five lines over five years. Licence, sized for year-five users and modules with the renewal uplift. Implementation, including the vendor's hours, the remediation you already know about and the internal time. Migration, sized by running an extract and counting the exceptions. Integration, one join per surviving system per year with an owner. Change, meaning customization re-applied at upgrades, reports rebuilt, training and the consultant on retainer. Put both options you are comparing on the same five lines.
For a mid-market company over five years, a fifth to a third, based on the estates described on this page. Implementation and migration are another third, most of it remediation and internal time, and integration and change share the rest, with change the largest single line for a company with specific processes. The vendor quote covers the licence and part of implementation, so the approved number is often a third of what is spent.
A suite subscription is quoted for today's users and modules, and renewals in the suite market typically rise faster than inflation, with added users and modules compounding on top. A flat five-year projection describes a company that is not growing. Size year five from the headcount you expect and the modules you will add, and ask the contract what the renewal cap is. The renewal letter in year three is where an unsized licence line is discovered.
Because the suite never replaces everything. The CRM the sales team keeps, the e-commerce platform, the carriers, the bank, payroll, the customer portal and the industry-specific tool all survive, and each still needs a join that is built, monitored and rebuilt at upgrades. Suites carry fewer joins than best-of-breed estates but not zero. Price a join per surviving system per year, including the person who owns it when it fails.
Remediating the customer, item and vendor masters, and drawing the integration map with a named owner for every connection. Both are worth the same whichever ERP is chosen or none, both are cheaper done at leisure than inside a signed project, and both reduce the implementation and migration lines directly. A company that does them first buys the ERP cheaper, later, or finds it does not need one this cycle, and any of those beats the quote.
"Something isn't working." "We're outgrowing our tools." "We're afraid to make the wrong move." No-Risk Discovery is a short, practical conversation that gets you clarity before you commit to anything big. We'll tell you if we're a fit. If we're not, we'll tell you that too.
Say who you are and ask. A person replies within one business day.
