Leaving OpenAir: an honest checklist

Who this is for

This is written by someone who spent fourteen years implementing that system, four of them inside Oracle NetSuite's own Professional Services team. Most of that work was putting firms into systems like this rather than taking them out, which is worth saying plainly: I know what a migration costs because I was the one doing it, not because I have sat where you are sitting.

I now sell an alternative, so read this with that in mind. I have tried to write the document I would have wanted handed to me at the start of one of these projects, which means it includes the cases where you should stay.

---

Part 1 — Before you decide anything, check these

Seat count against seats actually used

Firms carry leavers for months. The Users screen lists every user on the account. Compare that list against the seat count on your contract.

Which modules you are paying for and no longer use

The Resources module is the one I most often see paid for and then left alone — the skills matrix especially. It only works if people keep the underlying data current, and in most firms they stop. That is not really a criticism of the module: any tool with a skills matrix carries the same dependency, mine included. But check whether anyone has opened it this year before you pay for it again.

What the renewal uplift actually is

In writing, before you negotiate.

How much of your process lives in customization rather than configuration

This is the single biggest predictor of migration difficulty, and most firms underestimate it.

In the implementations I worked on it came down to two things: custom fields, and the scripting tool. There was no workflow builder, so anything that resembled a process rule ended up written as a script — and the scripts are the part that breaks.

---

Part 2 — What moves across

In my experience the data divides into three groups, and firms consistently misjudge the third.

Moves cleanly: clients, projects, people, rate cards, open timesheets.

Moves with work: historical timesheets and expenses, invoice history, anything with an approval trail attached.

Does not move, and has to be rebuilt:

Scripts. Whatever process logic lives in them is rebuilt from scratch in whatever you move to. This is the expensive one, and it is why the script inventory above matters more than it looks.

Approval workflows. They do not come across either. Rebuilding them means somebody has to be able to describe what the current ones actually do, and the person who configured them has usually left the firm.

---

Part 3 — The question most firms ask too late

What happens to your reporting history?

If your board pack compares this quarter against last year, and last year's data is in a system you have turned off, you have a problem nobody raises until the first month-end after go-live.

Three options, in descending order of cost: migrate the history, keep the old system read-only for a year, or accept a reporting discontinuity and document the date.

In the migrations I have run, firms typically carry about two years of transactional history into the new system and leave the rest behind. That is a sensible default to plan around: far enough back for year-on-year comparison, not so far that the project turns into archaeology.

Whichever option you pick, pick it before you sign anything — and ask your current vendor, in writing, what a read-only or archive arrangement would cost. Do not assume one is on offer.

---

Part 4 — If you are integrated with NetSuite

Most OpenAir installations post to NetSuite. That integration is usually the most fragile part of the migration, because it was configured once, years ago, by somebody who has left.

Before you move, find out:

• Which records flow, in which direction, and on what schedule
• What happens to in-flight records on cutover day
• Who at your firm can actually explain the mapping

The standard integration covers the main flows. What breaks it is usually not the integration itself but the data on either side of it: the same records have to exist and match in both systems, and once they drift apart the flow between them stops working. If that has been happening quietly for a year, a migration is when you find out.

---

Part 5 — Time and effort, honestly

For a 50-person firm, eight to sixteen weeks end to end. Two to four months is the honest planning range.

The costs firms forget: their own staff's time, running two systems in parallel, the first month-end being slower, and retraining everyone who approves anything.

---

Part 6 — When you should stay on OpenAir

The honest section, and the one worth reading twice.

Stay if the NetSuite integration is working. This is the one I would weigh most heavily. A running OpenAir-to-NetSuite integration is configuration that nobody at your firm can now fully explain, and moving means rebuilding it against something new. If the flows are posting and reconciling, that is worth more than a license saving.

Stay if your customization is deep. If significant process logic lives in scripts, rebuilding it elsewhere costs more than the license saving for years.

Stay if you are mid-year on a multi-year contract. Check your exit terms before anything else.

Stay if nobody internally owns the project. A PSA migration with no internal owner fails, regardless of the tool. This is the most common reason these go badly, and it has nothing to do with software.

Stay if the complaint is configuration, not the product. A great many firms want to leave a system that was set up badly four years ago. Reconfiguring what you have is cheaper and faster than replacing it — and sometimes that is the right advice even though I sell the alternative.

---

Part 7 — If you do move

1. Pick the internal owner before you pick the vendor
2. Agree the reporting-history decision up front — it is the one that surprises
3. Export everything while you still have access, whatever you think you need
4. Run parallel for one full month-end, not one week
5. Cut over at a period boundary, never mid-month

---

I build Leprechaun PSA, which is one of the alternatives you would be weighing. If the checklist above points you toward staying where you are, that is a perfectly good outcome and costs me nothing to say.