The file went out in January, the aircraft has been sitting on the ramp since March, the crews are qualified, and the authority still hasn’t set a date for the demonstration flight. In an initial certification, the delay is almost never where you look for it: it seldom lies with the fleet, almost always with the gap between what the manual describes and what the organisation actually does. What remains is to know which documents drive the schedule, and which merely follow it.
What the authority reviews is not a project, it is proof
An application for an air operator certificate is not presented like an ordinary application for authorisation. The applicant is not requesting a right to operate: it is asserting that it already meets a body of requirements, and it supplies the documents that establish this. The authority does not build the file, does not complete it, and never drafts it in the applicant’s place; it examines it, samples it, and returns it if it does not hold up. The burden of proof remains entirely on the applicant’s side, from the first meeting through to the signature.
This asymmetry explains most scheduling misunderstandings. An operator who treats the review as a negotiation files early, waits for feedback, incorporates it, refiles, waits again: it stacks up reviews, each of which consumes a slot. An operator who treats it as a demonstration files later, only once, with a self-contained file, and often comes out first. The filing date is an indicator of internal communication; it has no predictive value for the issue date.
It is also worth seeing what the certificate does not attest. It says nothing about the airworthiness of the aircraft, which falls under a different chain, different bodies and different documents. It says nothing about the soundness of the company, which falls under the economic side. It attests to one thing only: that the organisation has the means, the methods, the people and the documentation needed for the type of operation requested, and that it has shown this in ways other than on paper.
Five stages, one clock that does not belong to the applicant
Authorities on both sides of the Atlantic break the review down the same way, under different names: a pre-application contact, a formal application, a document review, on-site inspections combined with demonstrations, then issuance. This breakdown is not administrative, it is causal: each stage can only open once the previous one has closed, and overlaps remain marginal — they are not negotiable.
The pre-application contact is the stage most readily skipped, and that is a scheduling error before it is a methodological one. This is where the scope is declared — types and registrations envisaged, bases, network, specific approvals, target date — and where the authority sizes the team it will devote to it. A scope declared late cannot be made up for: the people who were meant to review it have been assigned elsewhere.
The formal application, for its part, is not assessed on substance. It is either accepted or returned: accepted, it opens the review and takes a slot; returned, it leaves nothing behind. Acceptance for processing is not an approval, it is an acknowledgement of completeness — and it is the only stage of the process where one can be beyond reproach without having yet demonstrated anything, since the list of required documents is known in advance.
Across the whole process, the share of time the authority truly controls is shorter than one might think: the document review, and the formalisation of issuance. Everything else belongs to the applicant, including the silences during which it waits for an answer to a question it could have dealt with before filing.
The authority doesn’t build your file – they check what you submit. If it’s incomplete, they send it back, and you have to wait for a new review slot. The faster way is to submit a complete file the first time.
A new cargo operator in Europe submits their AOC application in January. The authority finds gaps in the compliance matrix and returns the file in March. The operator corrects the issues in two weeks, but must wait until June for the next available review slot, delaying their first flight by 3 months.
The compliance matrix decides the schedule more surely than the fleet
The document that governs the review is neither the manual nor the organisation chart: it is the table that maps each applicable requirement against the passage of documentation that satisfies it. It is the key to reading the whole file, and the inspector does not read it line by line: they sample it. They pull a few references, open the manual at the indicated place, and check whether they find there what was announced.
Two flaws recur, with the same effect. The first is the hollow reference: the paragraph cited rephrases the requirement instead of describing how it is met — who does what, at what point, with which document, and what is done when it is not possible. The second is the chapter-level reference: twelve pages are pointed to instead of a paragraph, which amounts to handing the inspector the demonstration work the table was supposed to carry.
When the sample reveals several hollow references, it is not a few isolated lines that fall: it is the entire table that loses its evidentiary value, since nothing any longer distinguishes the solid references from the rest. Reworking it is then not a correction, it is a fresh review — and it is handled as such, with a slot to find again.
The trap. Writing the compliance matrix before the manual has been frozen. Each revision shifts the numbering, the references slip by a paragraph or a section, and the table ends up pointing somewhere else without anyone noticing: whoever wrote it rereads what they meant to write, never what is actually written.
What shows on paper, what only shows on site
The review separates two regimes of proof. On paper, it is established that an intention is written down, consistent and complete. On site, it is established that it is upheld by real people, with the real means at their disposal, within the time available to them. Almost everything that matters is checked twice, and it is the second check that sorts the files.

Only one document is never observed on site: the compliance matrix, which exists solely for the review. Conversely, only one piece of evidence is never obtained on paper: the behaviour of the complete system during a representative rotation. Between the two, every element read will be reviewed in the field, and that is where a manual borrowed from elsewhere gives itself away — a procedure for equipment that is not installed, a reference to a department that does not appear on the organisation chart, a cited form that was never printed.
| File element | What the document review establishes | What the on-site check adds |
|---|---|---|
| Operations manual, general part | Internal consistency, coverage of requirements, existence of a procedure for each | That the crew opens the right chapter and finds there what it actually does |
| Type-specific procedures | Compliance with the flight manual and the manufacturer’s data | That the checklists in service match the approved version |
| Training and checking programmes | Content, duration, qualification of instructors and examiners | Individual records kept, and traceability of checks already carried out |
| Subcontracted activities | Existence of a contract, scope, access rights for the operator and the authority | The service provider observed on the ramp, and what it knows of the operator’s procedures |
| Continuing airworthiness and minimum equipment list | Derivation from the reference list, rectification intervals, maintenance programme | A deferral exercised on board, with its entry in the technical log and its follow-up |
| Emergency procedures and cabin crew | Roles described, equipment provided, associated training and checks | The actions performed during the demonstration, within the time and with the equipment on board |
Nominated persons: an approval granted at interview
The accountable manager and the nominated persons — flight operations, crew training, ground operations, continuing airworthiness — are not boxes on an organisation chart. Each is accepted individually, on three criteria that do not carry equal weight: experience in the field, knowledge of the applicable rules and of the operator’s documentation, and genuine availability.
It is the third that most often stalls things. A nominated person who holds the same role at another operator, or who is otherwise a commander on a full flying programme, presents an overlap that the interview exposes within a few questions. The reservation then concerns one person, but it freezes the entire file: the organisation chart supports everything else, and a management system with one pillar left hanging is not reviewed.
The interview itself is not a test of regulatory knowledge. It checks that the person knows their own manual: where the procedure is, who wrote it, what it provides for when the situation falls outside the framework, how it is revised and who must be informed. A nominated person who quotes the regulation from memory but opens the manual at random tells the inspector exactly what they were trying to find out.
The right habit. Request the nominated persons’ interviews as early as possible, even before the documentation is closed out. A reservation lifted while the manual is still being written costs nothing; the same reservation discovered at the end of the document review costs a whole slot, and sometimes a hire.
The on-site phase is not negotiated, it is prepared
On-site checks are not just another inspection: they constitute the demonstration. The demonstration flight or flights must be representative of the operation applied for — a station on the intended network, a rotation with its preparation, its weight and balance, its stopover, its fuel management, its technical follow-up, and at night if the operation is. It is not the aircraft that is being watched, it is the chain: the person who prepares the flight, the person who follows it, the person who decides when something unplanned occurs.
Depending on the aircraft type and the content of the application, specific demonstrations are added — emergency evacuation, ditching procedures, cabin crew exercises — whose extent depends on what has already been demonstrated for the same configuration. All of them require an available aircraft, in an approved and registered airworthy condition, and qualified crews: they set the end of the schedule, never its start.
The other trap. Booking the demonstration slot at the same time as the aircraft delivery. As long as the documentation phase is not closed, that slot does not exist: it can neither be held nor slightly rescheduled, and the grounded aircraft then serves only to make the delay visible from senior management.
A return does not cost the time of the correction
This is the point that project schedules model worst. Review capacity is not a continuous queue: it is a team assigned to a file for a given window. Losing the window does not cost the length of the correction, but the wait for the next window, which depends on the department’s workload and not on the applicant’s energy.
The consequence is counter-intuitive for a management team in a hurry: it is almost always less costly to delay a filing by a few weeks than to file an incomplete dossier on the promised date. The first delay is visible and controlled; the second is invisible at the moment it is created and only reveals itself upon return, once the margin has already been spent elsewhere. It is also why the applications that succeed fastest are rarely the ones filed earliest.
The right habit. Have the compliance matrix reread by someone who wrote none of the manual’s chapters, with a single instruction: open each randomly drawn reference and state, without interpreting, whether it describes a way of doing something. This cross-check costs a few days and replaces the only other way of discovering a hollow reference, which is having the file returned.
Specific approvals requested from day one
Each specific approval — performance-based navigation, low-visibility operations, extended diversion time operations, dangerous goods transport — carries its own chain of evidence: dedicated documentation, training, sometimes a demonstration in line operation or in a simulator. When requested in the initial application, they are reviewed in parallel with the rest, but the slowest of them then dictates, in practice, the date of the whole certificate.
| Criterion | Requested in the initial application | Requested after issuance, by amendment |
|---|---|---|
| Effect on the certificate date | The slowest approval dictates the date of the whole | None: the certificate is issued without it |
| Demonstration burden | Combined with that of certification, on the same people | Spread out, on an organisation already run in and already under oversight |
| Review cost | Shared with the main review | Taken on separately, with its own slot and its own file |
| Main risk | A secondary approval delays the base operation | A signed contract requires a capability not yet granted |
| Case where there is no trade-off | The intended operation is impossible without it from the first flight | The capability belongs to a dated development plan, not to start-up |
The trade-off is not decided in the technical room, it is decided against the contract. An approval without which the first contract cannot be performed is not deferrable; an approval that serves the three-year plan can wait for an amendment to the operations specifications. Confusing the two amounts to making the first commercial flight pay for a capability that will not be needed for several seasons.
The economic side is reviewed in parallel, not afterwards
In Europe, commercial air transport requires two credentials, not one: the certificate, which attests technical capability, and the operating licence, which attests financial capability, insurance cover, ownership and control of the company, and its principal place of business. The two are linked: the licence is not issued without a valid certificate, and the certificate alone does not confer the right to carry traffic for remuneration.
The two files nevertheless follow neither the same pace nor always the same department, and the second requires documents — a business plan, forecast accounts, insurance certificates, ownership structure — whose acquisition depends on third parties with no reason to adopt the project’s schedule. An applicant who treats this strand as a closing formality ends up technically certified and commercially grounded.
What to go and check at home
Five internal checks, which require nothing from the authority, shift the schedule more than a follow-up meeting does. That the compliance matrix refers to numbered paragraphs, never to chapters. That each nominated person has a named deputy and a genuine workload compatible with their role. That every subcontracted-activity agreement explicitly carries the operator’s and the authority’s right of access. That the aircraft definition and the operational suitability data are frozen before the training programmes are written. That the list of approvals requested from the outset is that of the first contract, not that of the development plan.
One grey area remains that the texts do not settle: how far an authority accepts a deliberately narrowed scope at issuance, with a view to extending it later. A narrow scope speeds up certification and shifts the burden onto later amendments; too narrow in light of the model announced at the pre-application contact, it raises questions about the sincerity of the application. The answer depends on the authority, the market targeted and the track record of the nominated persons: it is raised at the first meeting, it is not discovered at filing.
For anyone preparing a type rating, an initial certification reads differently: a new entrant is recruiting before it exists in regulatory terms, its first crews are hired to carry out the demonstration, and the employment contract sometimes precedes the certificate by several months. The exposure bears no resemblance to that of an established airline, and the openings at start-up stage say so clearly enough for anyone who knows how to read them.
In all cases, the applicable regulation, the decisions of the competent authority and the operator’s approved documentation prevail over what has just been described: this article sets out a review mechanism, it does not substitute for any binding procedure.
Questions fréquentes
Can a certificate application be filed without yet having the aircraft?
Yes, to enter the review: the file is built on an aircraft definition, not on a physical example, and the manual as well as the training programmes are written from the data of the type certificate holder. On-site checks, however, require a registered aircraft, in an approved and operable airworthy condition. The aircraft therefore does not condition the start of the process, it conditions its end.
Should the compliance matrix refer to the chapter or to the paragraph?
To the paragraph, and to its version. A chapter-level reference transfers to the inspector the burden of finding the procedure, which is exactly the work the table is supposed to have done. A reference that, once opened, does not describe a way of doing something but rephrases the requirement causes the whole table to lose its evidentiary value, since nothing then distinguishes the reliable references from the rest.
Can another operator's operations manual serve as a starting point?
As a structural basis, yes, and it is common practice. But it must be reworked line by line, because traces of its origin show through immediately: procedures written for equipment that is not installed, references to a department that does not appear on the organisation chart, cited forms that do not exist. These clues are found through sampling and cost more, in schedule terms, than the time they had saved.
How long does an initial certification take?
The question has no fixed-duration answer: the process adds together the applicant’s time, which dominates by far, and the review slots the authority can grant. The usual order of magnitude is counted in quarters rather than weeks, and the variable that dominates the gap between two comparable projects is neither the size of the fleet nor the complexity of the network, but the number of returns endured.
Can specific approvals be added after issuance?
Yes, by amendment to the operations specifications, each with its own chain of evidence and its own review slot. Requesting them in the initial application has them reviewed in parallel, but the slowest then fixes the date of the whole certificate. The trade-off is decided in light of the first contract to be honoured, not the development plan.
Who sets the date of the demonstration flight?
The authority, and in practice only once the documentation phase is closed or treated as such. Booking that slot earlier has no effect, since it cannot open while documentary reservations remain outstanding. The only useful action on the operator’s side is to guarantee the simultaneous availability of the aircraft, qualified crews and handling agents over the announced period.

Leave a Reply
You must be logged in to post a comment.