Implementation is the variable line
Licence varies by perhaps 20% between comparable quotes after negotiation. Implementation for the same scope varies by more than 2×, and that is where the money and the risk are.
Research · updated August 2026
Almost nobody in the mid-market publishes pricing, which means buyers compare quotes they cannot benchmark against anything. This is our attempt to fix that using real quotes customers have shown us, with the pattern rather than any individual number.
Send it. We will tell you where it sits against comparable deals and which lines are soft.
Model the three-year total yourself. Assumptions are derived from the quote sample and stated on the calculator rather than hidden behind it.
What we found
Patterns that hold across vendors and across company sizes, and that are difficult to see when you are only looking at your own two quotes.
Licence varies by perhaps 20% between comparable quotes after negotiation. Implementation for the same scope varies by more than 2×, and that is where the money and the risk are.
A mid-market implementation consumes 400 to 1,200 hours of your own people. At loaded cost that is $40,000 to $150,000, and it appears in no vendor proposal we have seen.
Quotes are built around year-one cost. Seat growth, module additions, and annual uplift mean year three commonly runs 30 to 50% above year one for the same business.
The same configuration quoted in month one of a quarter and week twelve differs materially. This is well known and still under-exploited by mid-market buyers.
Integration appears in about half the quotes we see, and where it appears it is usually scoped for two systems when the customer runs eight. That gap surfaces in month four.
Median variance to original implementation quote across the sample is +27%. Partners who will state their own historical variance are rare and worth preferring.
Unpublished pricing lets a vendor charge different customers different amounts for the same configuration, which is rational for the vendor and expensive for the buyer who does not know where they sit in that distribution.
The 2.1× spread we see between the lowest and highest quote for comparable scope is not a reflection of different requirements. It is a reflection of different negotiating positions, different timing, and different information.
Your own team’s time is the single largest cost in most mid-market implementations and it appears in no vendor proposal, because a vendor cannot charge for it and including it makes their number look worse.
Four hundred hours is a light implementation with clean data. Twelve hundred is a multi-entity deployment with data problems. At a loaded cost of $100 an hour that is $40,000 to $120,000 of real cost, and it is drawn from the same people who still have to close the books each month.
Buyers spend most of their negotiating effort on licence, which is the line vendors are least willing to move on beyond a standard discount band. Implementation is quoted by a partner with more discretion and considerably more variance.
The most effective questions we have seen buyers ask are: what was your median variance to original quote last year, what happens if we exceed the estimated hours, and will you name the specific people assigned. Partners who answer all three are usually the ones worth working with, and the answers change the price.
We publish ours, which is unusual in this category and is partly a positioning decision. Our implementation multiple is roughly 0.25× year-one subscription against a market median of 1.4×, and the honest reason is that we do less work — no partner margin, a narrower product, and an integration-first approach that defers the migration.
That is a real difference and it is not entirely a fair comparison. A NetSuite implementation covers more ground than ours does. The number to compare is total three-year cost against what you actually need, which is what the calculator above is for.
Sixty-three quotes shown to us by prospects and customers between 2024 and 2026, covering NetSuite, Sage Intacct, Acumatica, Dynamics 365 Business Central, Odoo, SAP Business One, and our own. All are for US mid-market companies between roughly $10M and $150M in revenue.
No quote is reproduced or identifiable. Figures are medians and ranges across the sample, and any vendor with fewer than five quotes is excluded from vendor-level comparison and included only in the aggregate.
Internal time estimates come from customers who tracked it, which is a minority — eleven of the sixty-three. That figure is therefore the weakest number on this page and we would treat it as an order of magnitude rather than a precise range.
The sample is self-selecting: these are quotes shown to a competitor, usually by buyers who felt the number was high. That biases the sample upward and we state it rather than correcting for it, because we cannot correct for it honestly.
Questions
We will tell you where it sits against comparable deals and which lines have room in them.