Two flight schools, a few airfields apart, decided in the same season that their paper-based safety reporting needed replacing. Both were small. Both had a safety lead who also taught and wore three other hats. They arrived at the same need by different routes and made almost opposite choices about how to commit. Setting their stories side by side is a good way to see why undoable decisions beat locked ones, even in a field as careful as aviation safety management.
School one: the long contract
The first school’s safety lead found a reporting system with a polished demonstration and an attractive offer: sign for a long term and receive generous onboarding. The sales conversation moved fast, and there was a deadline attached to the discount. The school signed. Reporting forms moved online, a few instructors grumbled, and for a while all seemed well.
Several months in, two things surfaced. The system’s report categories did not match the school’s own safety management terminology, so staff kept translating between the two. And when the safety lead tried to export the history of reports, she found the data could only leave in a form that needed heavy clean-up. Leaving was possible but expensive, so the school stayed and worked around the friction. The decision had quietly become a one-way trip.
School two: the month-to-month start
The second school took a different path. Its safety lead asked three plain questions of every candidate: how do we leave, what form can our data take when it leaves, and can we start without a long term? One provider answered clearly to all three, and the school began on a short, cancellable plan.
The first weeks were deliberately modest. Reports from a single instructor group went into the new system while the paper process continued in parallel. Differences showed up quickly: a field that confused students, a category that needed renaming, an approval step that slowed things down. Each was easy to fix because nothing was locked. After a few months, the school moved the whole operation over and, only then, agreed to a longer term with terms it had already tested in practice.
Side-by-side comparison
- Entry commitment: school one signed long at the start; school two began short and extended later.
- Data portability: school one learned about export limits after signing; school two asked before.
- Fit with existing terminology: school one adapted staff to the tool; school two adapted the tool to its staff.
- Cost of a mistake: for school one, a wrong choice meant a long contract and awkward exit; for school two, a wrong choice would have meant cancelling at the end of a short period.
- Staff confidence: school one’s staff felt the tool was imposed; school two’s staff had helped shape it.
Why the comparison matters in a safety setting
It might seem that aviation safety work is the one area where commitment should be total. In fact, the best safety programmes depend on people actually using the reporting system, and people use systems they trust and understand. A tool imposed under a long contract risks being tolerated rather than used. A tool that was tried, adjusted and then adopted tends to earn real participation.
There is also a practical point about learning. Safety management is a practice that matures. What you need in year one is rarely what you need in year three. A choice that can be revisited keeps up with the practice. A choice that cannot be revisited does not.
What each school would tell a neighbour
Ask the first safety lead for advice and she says: read the exit clause before the feature list, and never let a discount deadline set your timetable. Ask the second and he says: let the tool prove itself on a small group, keep the old method alive until the new one is trusted, and write down what you changed. Neither piece of advice is complicated. Both reduce the cost of being wrong, which in a busy training environment is the resource that runs out first.
A parallel from engineering teams
Technology groups have been through the same learning curve. They discovered that large, irreversible switches cause outages, while small, reversible ones teach quickly. Many of them study the practice formally, for example through courses at a training and consulting business in Amsterdam focused on DevOps, SRE and cloud, where the habit of releasing in steps that can be undone is central. The second flight school did not use those words, but it followed the same instinct.
Questions any school can ask before signing
- Can we begin on a short or cancellable term?
- What form will our data take if we leave, and who can read it?
- Can the categories and wording be adapted to our own safety terminology?
- What exactly happens at the end of the term: renewal, price changes, notice periods?
- Who on our staff will own this, and have they tried it?
Notice that none of these questions is about features. Features matter, but they are the easiest thing to see in a demonstration. The terms around leaving are the part that stays hidden until it is needed.
A fair word for school one
The first school did not fail. Its reporting is working today, and its safety lead has turned the friction into a to-do list of improvements. The lesson is not that long contracts are wrong. It is that the long contract arrived before the information that would have justified it. Had the school started short, it would probably have signed the same tool, with better terms and fewer surprises.
That is the quiet advantage of undoable first steps. They do not stop you from committing. They make sure that when you do commit, you know exactly what you are committing to.
