When a company asks for a quote on an ERP project, it almost always compares the numbers. What it costs, how many hours it includes, which one is cheaper.
The number is the least important thing in that document. What determines how the project will go is the contracting model, because that decides what happens when something does not go as planned. And in a software project, something never goes as planned.
The essentials, in four points
- Hour blocks are not a quote, they are a balance. They tell you how much you will spend, not what you will receive.
- Meetings consume that balance. Scoping, questions and small changes come out of the same pot as the work.
- Each model creates a different incentive. Not out of bad faith, but out of arithmetic.
- No model is always better. Some work can only be contracted by the hour.
Where your purchased hours actually go
An hour block works like a balance. You buy forty hours, the provider deducts what it works and tells you when it is running out. Put that way it sounds like the most transparent arrangement there is.
Picture a development estimated at eight hours. A new report, a simple automation.
There is a first meeting to properly understand what is needed, because the initial request never carries the full detail. One hour. Two questions come up that only someone in accounting can answer and you have to wait until that person has a moment. Another half hour. During development an ambiguity appears about an edge case, an email goes out, the reply arrives the next day. Twenty minutes. A work in progress is shown, the user sees the screen and says a field was missing. Another hour between the meeting and the adjustment.
None of those things is wrong. That is how useful software gets built. But all four came out of your balance. Of the eight hours you bought, a little over five were left for the work, and the work still needed eight.
From there you have two paths and both are uncomfortable. Either you buy more hours, with the feeling that you are paying twice for the same thing, or the consultant finishes with what is left, which is the rushed version.
The incentive nobody says out loud
This is not a question of good or bad faith, it is arithmetic. When you charge by the hour, billing grows with the length of the project. When you charge by delivery, margin grows with efficiency.
That does not mean whoever charges by the hour is stretching the work. The vast majority of consultants who bill that way are serious professionals. It means something harder to detect. In an hourly model, nobody on the provider's side has an economic reason to fight the project's inefficiencies. The forty minute meeting that could have been an email costs nobody anything except you, and you find out at the end.
The most expensive effect shows up later. A developer who knows three hours of balance are left and that five are needed makes different decisions. The edge case does not get tested, the code is not left ready for the change coming in six months, nothing is documented. That comes back as an error during month end close, or as a customization nobody understands two years later, something we see often and write about in What Happens After Implementing NetSuite.
Hour pressure does not produce rushed work because people are bad. It produces it because the time to do it properly was never bought.
The three models you will come across
| Model | What you buy | When it fits | Where it hurts |
|---|---|---|---|
| Hour blocks | A team's availability | Work that cannot be estimated, incident support | Meetings and changes drain the balance without warning |
| Fixed price by scope | A defined deliverable | Very clear, stable scope from the start | Every change gets renegotiated, and the unforeseen lands on whoever signed |
| Time blocks with scope | A delivery with a date | Bounded projects where details get discovered while working | Scope has to be defined before starting |
When hour blocks are the right call
This is worth stating, because an article that presents one model as the answer to everything is a brochure.
Hour blocks are the right fit when nobody knows yet what needs doing, such as an initial review of a messy account, because asking for a fixed price there forces the provider to guess. Also for support, where nobody can commit to a date for a problem that has not happened. And when years of relationship already exist and you know the team's pace.
Outside those cases, when you need something specific to exist by a date, the clock and the result start competing.
Three practices we do not follow
We are talking about practices, not companies. They are widespread, they are legal, and in our view they are wrong.
Charging for scoping meetings. That is billing the client for explaining their own business to us. Definitions are part of our work, not theirs. When they deduct from the balance, the client starts holding back questions to save hours, and the question left unasked is exactly the one that creates the rework billed separately later.
Estimating without having seen the process. A figure given over the phone on the first call is not an estimate, it is a figure. It always gets adjusted upwards afterwards, and the client who decided based on that number has already reorganized a budget around something that did not exist.
Charging to fix something that did not work as requested. If we deliver something that does not do what was agreed, fixing it is our obligation and not a new sale. This is the most uncomfortable question you can put to any quote, which is why it tells you the most.
How we work, and what we give up
Since June 2025 we have not sold hours. We sell time blocks with an agreed scope, usually one or two weeks, with a delivery date.
Inside that block, meetings deduct nothing. If three calls are needed to properly define an invoice document, all three happen. The client does not watch the clock while asking.
And we size the block with room to spare. If the work is four days, the block is one week. The price does not change because of that. If the task is worth a hundred dollars, it is worth a hundred dollars whether it ships in four days or in one week. What we widen is the timeline, not the amount. That margin is what allows testing the edge case, leaving the development ready for what comes next, and handing over something finished, instead of closing in a rush because the balance is running out. The practical result is that we usually deliver ahead of the date, and that since we started working this way we have not had to go back over a delivery to correct it.
What this costs.
Scope has to be defined before starting. That is a longer first conversation than asking for a number of hours, and some clients find it uncomfortable until they see what it is for.
A scope change moves the date. We do not absorb it silently. If a new requirement shows up mid block, we decide together whether it goes in by moving the delivery or waits for the next block. That is less comfortable than an automatic yes, and it is the only way for the date to mean anything.
It does not fit everything. For support we use a different arrangement, and the same goes for an open ended assessment.
The diagnostic, and why we do not charge for it
A paid diagnostic has its logic, it takes real time from someone senior and it filters out unserious enquiries.
We choose not to charge for it because the diagnostic may conclude that you need an area we do not cover, or that the problem can be solved without hiring anyone. Those two outcomes are valuable to you and they are among the most likely. Charging you to reach them would put us in the position of having our income depend on the answer always being the same.
Five questions for any quote
Take these with you even if you hire someone else.
- Do scoping meetings deduct from the total? If the answer is yes, add fifteen to twenty percent to whatever you are comparing.
- Is there a delivery date or only a number of hours? Those are different commitments and only one of them can be missed.
- What happens if the hours run out before the work does? The answer has to be in writing.
- Are fixes for something that does not work as requested billed separately? The most uncomfortable one and the most revealing.
- Who does the work? An estimate in hours says nothing if you do not know whether a senior or a beginner is executing them.
If you are at the earlier stage, choosing who to ask for a quote, we covered that separately in How to Choose the Right Partner to Implement NetSuite.
In short
The hour blocks versus fixed price debate looks administrative and it is not. It is the decision that determines whether, when something unexpected shows up, you and the provider will be on the same side or on opposite ones.
The most reliable sign that you are talking to someone serious is that they explain which model they use, why, and what that model does not solve.