Air traffic control selection, explained

NATS runs the UK’s main recruitment route for trainee air traffic controllers. Selection is four stages, sat in order, and each one screens out candidates before the next. This page covers all four — what happens, what NATS publishes about it, and what it does not.

Before you apply: are you eligible?

NATS publishes its entry requirements, and they are firm. You need all of them, not most of them.

• Aged 18 or over.
• An indefinite, unrestricted right to work in the UK.
• Five GCSEs at grade 4 or above (or equivalent), which must include English and Maths.
• The ability to pass SC-level security clearance.
• The ability to pass a Class 3 medical assessment.
• Willingness to adhere to the drug and alcohol policy.

You do not need a degree, a pilot’s licence, or any aviation background. NATS recruits ab-initio and trains you from scratch. What it will not do is train around a missing GCSE.

Stage 1 — the online ability tests

Stage 1 is a battery of eleven online ability tests, sat remotely. It is the biggest sift in the process: most candidates who are rejected are rejected here.

NATS names the areas the tests cover — spatial awareness, logical thinking, numerical ability, sense of direction and reaction speed — but it does not publish which individual tests it uses, how long each one runs, or the score you need. That is the single most important thing to understand about preparing for Stage 1: you can train the abilities, and you cannot rehearse the exact paper.

• Spatial awareness — holding an orientation in your head and transforming it.
• Logical thinking — deducing a rule from examples and applying it.
• Numerical ability — quick, accurate arithmetic and reading data under time pressure.
• Sense of direction — tracking a heading change and knowing which way you now face.
• Reaction speed — responding fast without letting accuracy collapse.

The practical consequence: speed matters as much as accuracy. These are speeded tests, and candidates who are careful but slow score badly. Practise against a clock from the start rather than working through questions untimed.

Stage 2 — situational judgement and listening

Pass Stage 1 and you sit two further online tests, virtually proctored by Aon: a situational judgement test and a listening comprehension test.

NATS describes the SJT as assessing how you evaluate possible responses to workplace situations. That one sentence is the entire published specification — the format, the number of items and the scoring are not released. The scenarios do not require air traffic knowledge, because you have not been trained yet; they are about judgement.

The listening test assesses, in NATS’s words, how accurately you understand, retain and use spoken information. In an operational room you take instructions and readbacks by ear, at pace, in a range of accents, and you cannot ask for a third repeat. The test is a proxy for that.

Stage 3 — the assessment centre

Stage 3 is a virtual assessment centre held over Microsoft Teams. It contains three things, and candidates routinely prepare for only the first.

• An interview — competency-based, with a single assessor.
• An air traffic control knowledge test.
• A retest of your earlier ability tests, while being observed.

The interview is evidence-based. Expect questions in the form ‘describe a time when…’, and expect follow-ups: what were the other options, what did you actually say, what would you do differently. Candidates consistently describe NATS interviewers as friendly and professional — it is not a stress interview, but a vague answer will be probed until it produces something concrete or falls over.

Answer in STAR: the situation, the task you held, the action you personally took, and the result. Most candidates deliver situation and action and then trail off, which loses the part that actually shows judgement. Say ‘I’, not ‘we’ — the assessor is marking you, not your team.

The knowledge test is drawn from documents NATS sends you before the centre; the invitation carries the link and the password. That pack is the only source that counts. Crowdsourced flashcard decks go stale between campaigns and carry other people’s errors.

The observed retest exists to confirm the Stage 1 and Stage 2 results were genuinely yours. There is no technique to it — if you sat the originals honestly, keep practising so you arrive in the same shape.

Stage 4 — the practical exercise

Stage 4 is a practical exercise, delivered over Teams or face to face at NATS’s Whiteley office. You are briefed, then work a simulated section of airspace across a series of short runs, explaining your reasoning and taking assessor feedback between them.

The structure is the point: it is not one attempt scored once. It samples whether you take correction and apply it under pressure — which is exactly what the training pipeline will ask of you for the next year and a half. Being coachable is being assessed.

The competencies the interview is built around

The established guidance consolidates to twelve areas. Prepare a real example for each; several can share a story as long as you can pull out the different part each time.

• Teamwork, communication and accountability.
• Adaptability and resilience.
• Decision-making under time pressure and with incomplete information.
• Learning from feedback, and performing while observed.
• Managing pressure and competing priorities.
• Following procedures — including when a shortcut would be faster.
• Responding to your own mistakes, with ownership rather than blame.
• Motivation for the career, and a realistic understanding of the training and mobility.

If you pass: training and the job

Training is roughly 12 to 18 months of college-based work depending on specialism, at the NATS training centre at Whiteley in Hampshire or at Global ATS in Gloucestershire. It combines theory with simulator practice and is assessed continuously. After college you are posted to an operational unit for hands-on training on live traffic until you validate — qualify to work the position unsupervised.

You do not choose your discipline. Aerodrome control (the tower), approach control (arrivals and departures) and area control (en-route, worked by radar from a centre) are assigned during training based on need and performance. Nor do you choose your unit. Both surprise candidates who have assumed otherwise, and both come up at interview.

What NATS does not publish

It is worth stating plainly, because most pages competing for this search will imply otherwise. NATS does not release its question bank, the timings of individual Stage 1 tests, the number of items in any test, the pass mark for any stage, the SJT scoring model, or the Stage 3 knowledge-test syllabus.

So no preparation resource — this one included — can honestly claim to reproduce the real tests. What it can do is train the abilities NATS names, in the same formats and under the same kind of time pressure, so that nothing on the day is unfamiliar. Treat any site that advertises the real questions with suspicion: it is either guessing or repeating someone’s recall of a different campaign.

A sensible preparation order

• Start with Stage 1, because that is where most candidates fail and where practice pays most. Work to a clock.
• Find your weakest ability area and give it disproportionate time — a single weak area is what fails a battery.
• Move to Stage 2 once Stage 1 feels comfortable: judgement scenarios, and listening at operational pace.
• Write your competency examples out before you need them. Six solid stories, each with a result, covers most of the interview.
• Read the Stage 3 pack properly when it arrives, more than once.
• Keep practising the ability tests right up to the assessment centre — the observed retest is the same tests again.

In the UK, most trainee air traffic controllers are recruited by NATS, the country's main air navigation service provider. You do not need an aviation background, a degree, or a pilot's licence. NATS recruits ab-initio — it trains people from scratch — which is why selection leans so heavily on aptitude rather than experience.

What the job actually is

Controllers keep aircraft safely separated and moving efficiently. The work splits broadly into three disciplines: aerodrome control (the tower — traffic on and immediately around the airfield), approach control (sequencing aircraft arriving at and departing from an airport), and area or en-route control (traffic in the airspace between airports, worked from a control centre rather than a tower).

You do not choose your discipline. It is assigned during training, based on where the need is and how you perform. That surprises a lot of candidates, and it comes up in interviews.

The four selection stages

• Stage 1 — a battery of eleven online ability tests. This is the big sift and most candidates do not get past it.
• Stage 2 — an online situational judgement test and an online listening comprehension test, both virtually proctored by Aon.
• Stage 3 — a virtual assessment centre over Microsoft Teams: an interview, an air traffic control knowledge test, and observed retesting of the earlier ability tests.
• Stage 4 — a practical exercise, delivered over Teams or face to face at NATS's Whiteley office, in which you work simulated airspace across a series of short runs.

Passing Stage 4 may lead to a conditional offer. That is still conditional on medical clearance, security screening and other pre-employment checks, and it does not guarantee an immediate course date.

What to be realistic about

• Shift work, including nights, weekends and public holidays.
• Geographic mobility — you go where you are posted, not where you would prefer.
• A long training pipeline with genuine failure rates. People do wash out, including after passing selection.
• Your licence is validated at a specific unit rather than being universal.

Where to start preparing

Stage 1 is where most people fail, so that is where preparation pays. The five ability areas NATS names publicly are spatial awareness, logical thinking, numerical ability, sense of direction and reaction speed. Practising those under time pressure is the single highest-value thing you can do — and unlike the later stages, it is entirely trainable.

NATS describes Stage 1 as a battery of eleven online ability tests. It is the first real filter in the process and it removes the majority of applicants. If you only prepare for one stage, prepare for this one.

The five ability areas

NATS names five broad areas publicly rather than listing eleven individual tests:

• Spatial awareness — understanding how objects sit and move in relation to each other.
• Logical thinking — finding a rule from examples and applying it consistently.
• Numerical ability — working accurately with numbers and data at speed.
• Sense of direction — orientation, bearings and tracking which way something is facing.
• Reaction speed — responding promptly and accurately to a change on screen.

What that means for preparation

You cannot rehearse the exact items, so rehearsing the underlying abilities is the whole game. All five areas respond well to practice — not because you memorise anything, but because familiarity removes the cost of working out what the task is while the clock runs.

That cost is larger than people expect. A candidate meeting a bearings question for the first time under time pressure spends their first several items simply understanding the format. A candidate who has seen it fifty times spends none.

Speed and accuracy together

Speeded ability tests almost always score both. Rushing produces errors that cost more than the time saved; working carefully produces too few answers. Finding your own sustainable pace is a skill in itself, and the only way to find it is to sit timed runs rather than untimed practice.

Common mistakes

• Practising untimed, then meeting the clock for the first time on the day.
• Only practising the areas you are already good at, because they feel better.
• Treating reaction speed as untrainable — consistency under fatigue absolutely is.
• Assuming a strong maths background carries numerical ability. It is speed under pressure, not depth.

Stage 2 follows Stage 1 and contains two assessments: an online situational judgement test and an online listening comprehension test. Both are virtually proctored by Aon, which means webcam and microphone consent for identity validation. Candidates who decline consent cannot proceed.

The situational judgement test

NATS says the SJT assesses how you evaluate possible responses to workplace situations. In practice that means realistic scenarios from a safety-critical operation, each followed by possible responses you judge between.

There is no ATC knowledge required — you have not been trained yet, and the test knows that. What is being assessed is judgement: when to speak up, who to tell, how to balance procedure against pace, and whether you report your own mistakes.

The listening comprehension test

NATS describes this as assessing how accurately you understand, retain and use spoken information. Listening is the core mechanical skill of the job — a controller works almost entirely by voice — so it is a sensible thing to test early.

The most transferable practice is holding several values from one spoken message: a callsign plus a level plus a runway, delivered once, at pace, in an accent you may not be used to. Sub-vocalising numbers as you hear them is the cheapest technique that works.

What NATS does not publish

• The number of questions in either assessment.
• The time limit, or whether questions are individually timed.
• Whether audio can be replayed, or whether notes are allowed.
• The scoring formula, the pass mark, or how the two assessments are weighted.

Treat any source that states those confidently with suspicion. They are not published.

Stage 3 is a virtual assessment centre conducted over Microsoft Teams. It has three components, and candidates routinely prepare for only one of them.

1. The interview

A competency-based conversation exploring motivation, your understanding of the role, and evidence of how you have behaved before. Expect themes like teamwork, communication, accountability, resilience and commitment to safety-critical work — and expect follow-up questions when an answer is thin.

Structure your answers with STAR — Situation, Task, Action, Result — and make sure the Action is what you personally did rather than what the team did. Interviewers probe for the missing part, and "we" answers are the most common weakness.

Your examples can come from anywhere: work of any kind, study, sport, volunteering, family responsibility. No aviation background is expected and its absence is not a weakness.

2. The air traffic control knowledge test

This tests whether you have genuinely researched the career — the role of NATS, the controller disciplines, the training pathway, the student licence, and the mobility the job requires. It is not testing technical material you have not been taught.

3. Observed retesting

Some of the earlier ability tests are repeated while an assessor watches. The stated purpose is confirming that your Stage 1 and Stage 2 performance was genuinely yours, and observing your concentration and how you follow instructions.

NATS does not specify which tests are repeated, how many, or whether every candidate gets the same ones. The practical implication is simple: your Stage 1 practice is not finished when you pass Stage 1. Keep those abilities warm.

The setup matters

It is a video assessment, so a quiet private room, decent lighting, a plain background, a stable connection and a working webcam and microphone are part of the exercise rather than incidental to it.

Stage 4 is the final assessment stage: a practical exercise delivered virtually through Microsoft Teams or face to face at NATS's Whiteley office. It runs on a secure digital-twin platform, and it is the closest thing in selection to the actual job.

What happens

• You are briefed on the task, the rules and what you are being asked to achieve.
• You manage a simulated section of airspace in real time.
• You are asked to verbalise your reasoning as you work.
• You complete a series of short runs rather than one long one.
• You get assessor feedback between runs, and are expected to act on it.

The thing most candidates miss

NATS lists what it assesses: safety, efficiency, task completion, situational awareness, communication, problem-solving — and learning potential.

That last one changes how you should approach it. The exercise is structured as multiple short runs with feedback in between precisely so that assessors can see whether you improve when corrected. A candidate who performs moderately on run one and visibly better on run three is demonstrating exactly what is being looked for. A candidate who performs identically each time, however competently, is not.

Verbalising your reasoning

Being asked to think aloud while concentrating is genuinely awkward the first time. Practise narrating your decisions during any timed task — what you are prioritising, what you are watching, why you chose this option over that one. Assessors cannot score reasoning they cannot hear.

Safety and efficiency together

Both are assessed. A run that is perfectly safe but achieves nothing is not a good run, and neither is a fast one that loses separation. The balance between them is the judgement being tested.

What is not published

NATS does not disclose task rules, scenarios, performance thresholds, scoring formulas or disqualification criteria. Passing Stage 4 may lead to a conditional offer, which remains subject to medical clearance and security screening and does not guarantee an immediate course date.

FEAST — the First European Air Traffic Controller Selection Test — is a tool designed by EUROCONTROL and used by air navigation service providers to identify suitable controller candidates. It has grown to around 50 providers across 40 countries, including NATS in the UK, DFS in Germany, skyguide in Switzerland, ENAV in Italy and the IAA in Ireland, plus non-European users including NAV CANADA, Dubai Air Navigation Services and CAAS in Singapore.

EUROCONTROL announced its 100,000th FEAST candidate in 2019, having run the test since 2006.

FEAST I — basic skills and abilities

• A cognitive battery covering decision-making, planning, memory, logical reasoning, visual perception, attention, multi-tasking and spatial orientation.
• An English language test assessing listening and comprehension.

EUROCONTROL gives the duration as roughly two and a half to four hours.

FEAST II — multi-tasking work samples

Two work-sample tests that simulate controller tasks requiring several activities at once. These are not question banks — they are performance tasks, and they are the part of FEAST that most preparation material does not touch, because a multiple-choice tool structurally cannot reproduce them.

FEAST III — personality questionnaire

A self-rating questionnaire on characteristics relevant to the role. There are no right or wrong answers, and attempting to game a personality inventory is both ineffective and detectable — most include consistency and social-desirability checks.

The parts by name

EUROCONTROL names the components on its own FEAST information site, and knowing the names is genuinely useful — it is what lets you tell whether a provider is running the full package or a subset.

• CLASSIC PLUS KNOWLEDGE — the FEAST I cognitive battery. EUROCONTROL lists the domains it covers: decision-making, planning ability, memory, logical reasoning, visual perception, attention, multi-tasking and spatial orientation.
• ELICIT — the English Language Listening and Comprehension test, sat as part of FEAST I. Native English speakers routinely treat this as free marks and lose some.
• RADAR — a FEAST II work sample: you work simulated traffic on a scope rather than answer questions about it.
• MULTI-PASS — the other FEAST II work sample, built around holding several demands at once.
• FPQ — the FEAST III personality questionnaire, rated on a 1–5 scale. There are no right answers and nothing to revise.

FEAST I runs roughly 2.5 to 4 hours; each FEAST II test runs about 1.5 to 2 hours; the questionnaire takes around 35 to 45 minutes. What EUROCONTROL does not publish is the individual subtests inside CLASSIC PLUS KNOWLEDGE, their timings, their item counts or any pass mark — so treat any site quoting those precisely as guessing.

How FEAST relates to NATS

NATS is one of the providers that uses FEAST, so a UK candidate preparing for NATS Stage 1 and a candidate preparing for FEAST elsewhere are training overlapping ground. The abilities are the same; the packaging and the stage names differ.

The Air Traffic Skills Assessment, usually shortened to ATSA, is the FAA's entry-level aptitude assessment for air traffic control specialists in the United States. It is taken by candidates referred from an FAA hiring bid.

Unlike NATS's staged process, the ATSA is a single sitting — commonly reported as around two hours and forty-nine minutes, with seven subtests and more than 200 questions.

What it assesses

Broadly: decision-making, spatial awareness, multitasking and working memory. Publicly documented subtests include:

• Arithmetic / memory — a digit appears, then another; you type the positive difference, at speed.
• Spatial recognition — two sections testing spatial reasoning.
• Reading comprehension — passages with questions against a tight clock.
• Air traffic scenarios — a radar-style display with converging traffic, where you predict conflicts and make routing decisions under load.
• A personality or behavioural component.

The bid-window pattern

FAA hiring runs in bid windows rather than continuously. When one opens, applications arrive in enormous volume — the FAA reported 6,000 applications within the first twelve hours of one recent window, and referred more than 8,320 candidates to the ATSA from another.

The practical consequence is that preparation time is short and unpredictable. Candidates who wait for a bid to open before starting are already behind. The FAA has published hiring targets of 2,200, 2,300 and 2,400 new controllers for FY2026, FY2027 and FY2028.

NAV CANADA is Canada’s air navigation service provider and recruits trainee air traffic controllers directly. Its selection runs as a sequence of screens: eligibility checks and an application, an online assessment, in-person testing, an assessment centre, language testing, and then pre-employment conditions. Passing one step does not guarantee an invitation to the next.

The online assessment

The first real hurdle is a short online assessment — around 30 minutes, sat remotely, made up of four timed sections. Each targets a different cognitive ability rather than any aviation knowledge, because at this point you have had no training.

• Inductive reasoning — two worked grids share a hidden rule; you recover it and pick which other grids follow the same one.
• Concentration — sustained, accurate attention on a simple task repeated at speed.
• Deductive reasoning — working out where a shape belongs within a matrix from the rules the grid implies.
• Spatial orientation — directing an aircraft using visual information, and holding an orientation in your head.

Thirty minutes across four sections means very little time per item. As with every ability battery, the failure mode is not misunderstanding the task — it is being careful but slow. Practise against a clock.

The in-person day is the FEAST exam

This is the part most candidates do not realise until they get there: NAV CANADA’s in-person stage uses EUROCONTROL’s FEAST — the same test package used across Europe, including by NATS. It runs around six hours at an air traffic control centre.

FEAST is built in modules. FEAST I is a battery of cognitive tests covering decision-making, planning, memory, logical reasoning, visual perception, attention, multi-tasking and spatial orientation, alongside ELICIT, an English listening and comprehension test. FEAST II is the work-sample half: RADAR and MULTI-PASS, which simulate controlling traffic and put multitasking under real load. There is also a personality questionnaire, FEAST III.

Note what EUROCONTROL publishes and what it doesn’t. It names the domains FEAST I assesses; it does not publish the individual subtests, their timings, item counts or pass marks. Not every organisation using FEAST uses every module, either — so the exact composition of your day is set by NAV CANADA, not by EUROCONTROL.

Language testing

Canada’s bilingual requirements mean language testing forms part of the process. Which language standard applies depends on the unit and the role, so treat this as a step to check against the specific posting you are applying for rather than assuming a single national rule.

What NAV CANADA does not publish

As with every provider in this space: no question bank, no per-section time limits, no pass marks, and no scoring model. The abilities it names publicly — memory, working speed, spatial visualisation, reasoning, attention, information processing and simple arithmetic — are the honest basis for preparation. Anyone offering the real NAV CANADA questions is guessing.

How to prepare

• Train the four online-assessment abilities first — that screen is what stands between you and the in-person day.
• Work every ability test against a clock. Untimed practice builds the wrong habit for a 30-minute, four-section screen.
• Then move to the FEAST ground: multi-tasking, memory, spatial orientation and listening at operational pace.
• Practise the radar work sample specifically. Directing traffic while a second task runs is unlike anything else on the battery.
• Do not neglect English listening — ELICIT is part of FEAST, and it is a common blind spot for first-language English speakers who assume it is free marks.

FEAST — the First European Air Traffic Controller Selection Test — is built in modules, and FEAST I is the one nearly every candidate sits. It is the longest single sitting in most European selection processes, and it is where the majority of candidates are screened out.

What FEAST I contains

Two things, sat in one session:

1. 1CLASSIC PLUS — the cognitive batteryA set of computer-based tests covering the abilities EUROCONTROL names for the controller job. A mix of multiple-choice items and speed-based reaction tests.
2. 2ELICIT — English Language Listening and ComprehensionA listening test sat alongside the cognitive battery. It is part of FEAST I, not a separate day, and it is the module first-language English speakers most often underestimate.

How long FEAST I takes

EUROCONTROL gives a range rather than a fixed figure, because the organisations using FEAST do not all use every element.

Plan for the top of that range rather than the bottom. A candidate who has rehearsed in twenty-minute bursts and then meets a four-hour sitting is not being tested on ability by hour three — they are being tested on stamina they did not train.

The abilities EUROCONTROL names

EUROCONTROL publishes the ability areas FEAST I assesses. It does not publish which individual test measures which, so treat these as the ground to cover, not a test list.

• Decision-making — choosing between options under time pressure, with incomplete information.
• Planning — sequencing actions ahead of time rather than reacting to each event as it lands.
• Memory — holding values, callsigns and instructions while something else demands attention.
• Logical reasoning — inferring a rule from examples and applying it to a new case.
• Visual perception — reading a picture quickly and accurately, and spotting what changed.
• Attention — sustaining accuracy on a simple task for far longer than it stays interesting.
• Multi-tasking — running two or more tasks at once without either collapsing.
• Spatial orientation — holding an orientation in your head and transforming it.

The instruction that catches people out

EUROCONTROL tells candidates to work as quickly and precisely as possible, and adds something most candidates miss: you are not required to complete every item to pass.

That single sentence changes how you should sit a speeded test. If finishing is not the standard, then grinding at one hard item while the clock eats three easy ones is a losing trade. Move on. The item you skipped costs you one mark; the three you never reached cost you three.

ELICIT — the module people give away marks on

ELICIT tests English listening and comprehension. In an operational room you take instructions and readbacks by ear, at pace, from people with a wide range of accents, and you cannot ask for a third repeat. ELICIT is a proxy for that.

First-language English speakers routinely treat it as free marks and drop some. The difficulty is not vocabulary — it is holding several values from one transmission while the next one is already arriving, and coping with an accent you have not heard before.

• Listen to the whole message before you start answering. The detail you need often comes last.
• Hold the callsign first — everything else attaches to it.
• Where a message addresses two aircraft, keep them separate in your head from the first word.
• Practise the accents you find hardest rather than the ones you find comfortable.

What EUROCONTROL does not publish

No item bank, no scoring model, and no pass mark. Each organisation using FEAST sets its own cut score, and EUROCONTROL neither owns nor distributes the results — your feedback comes from the organisation you applied to, and its policy on how much feedback to give is its own.

So there is no target number to aim at. Aim to score as highly as you can, and treat any page that quotes you a FEAST pass mark as guessing.

Where FEAST I measures abilities in isolation, FEAST II puts them together. It is the work-sample half of the battery: tasks built to resemble the shape of controlling traffic, sat by candidates who have never controlled anything.

What FEAST II contains

Not every organisation uses FEAST II, and of those that do, not all use both tests — MULTI-PASS may appear without RADAR. What your day contains is set by the organisation you applied to, not by EUROCONTROL.

What it is really testing

The skill under examination is not aviation. It is whether you can hold several jobs at once and keep all of them at an acceptable standard, rather than doing one perfectly while the others fail.

That distinction matters for how you practise. Most people, given a multi-tasking exercise, unconsciously pick a favourite task and protect it. Scoring rewards the opposite: an even hand across everything, with nothing abandoned.

• Scan on a rhythm rather than reacting to whatever moved last — the loudest event is rarely the most urgent.
• Accept being slightly late on everything rather than perfect on one thing and absent on the rest.
• When a task goes wrong, deal with it and let it go. Replaying a mistake costs you the next three events.
• Keep the secondary task alive. Candidates lose more on the task they quietly dropped than on the one they got wrong.

How FEAST II relates to the rest of the process

FEAST II is not the last word. Providers commonly follow it with an assessment centre, interviews, medical and security screening and, where a role needs it, language testing. Passing the work samples means you have cleared the aptitude bar, not that you have the job.

FEAST III is the FEAST Personality Questionnaire, the FPQ. It sits apart from the rest of the battery because it is not an ability test — there is nothing to be good at.

Why “answer naturally” is real advice, not a formality

Personality questionnaires of this kind carry a large number of items and revisit the same underlying traits from different angles. A candidate constructing an ideal controller answer-by-answer has to keep that construction consistent across the whole instrument, at speed, while the questions deliberately approach the same trait from opposite directions.

Most people cannot. What tends to emerge instead is an inconsistent profile — and inconsistency is more visible than any individual answer.

How to sit it

• Answer at pace. Your first instinct is the answer the instrument is designed to capture.
• Avoid parking everything in the middle of the scale. A flat profile says little and reads as evasive.
• Do not agonise over a single item — no one answer decides an outcome.
• Answer as you actually are at work, not as you are at your best on your best day.

Not every provider uses it

As with FEAST II, whether FEAST III forms part of your day is the applying organisation’s decision. Nor is a personality questionnaire unique to FEAST — the FAA’s ATSA carries one as a scored component, and it works the same way.

Most FEAST preparation advice skips the first and most obvious step: EUROCONTROL runs a free official preparation platform, and publishes ten named practice exercises. Start there, then use everything else to build speed and stamina on top.

Step one: the official platform

EUROCONTROL hosts a free FEAST training platform with demo tests. It is worth saying what it is and what it is not: EUROCONTROL states openly that its practice items are not identical to the real tests. They show you the shape of the task, the instructions and the interaction — not the item bank.

That is still the single most valuable hour of preparation available, because the biggest avoidable loss in a speeded test is time spent working out what the screen wants. Arrive already knowing.

Step two: know which module you are sitting

Do not prepare for the whole battery if your day is one module of it. Providers pick and choose.

Your invitation is the authority on which of these you will sit. Around 50 organisations across 40 countries use FEAST, and the composition of the day is set by each of them.

Step three: train speed, not just correctness

FEAST I is speeded. Working carefully and slowly produces a low score even when almost every answer is right — and EUROCONTROL’s own instruction, that completing all items is not required, tells you the test is built expecting you to leave some behind.

• Practise against a clock from the first session, not once you feel ready. Untimed practice trains the wrong habit.
• Set a per-item budget and abandon anything that overruns it. The skill is letting go, and it needs rehearsing.
• Track accuracy and attempts together. A rising accuracy with falling attempts is going backwards.
• Build up to a full-length sitting at least once. Hour three of a four-hour test is a different exercise from hour one.

Step four: treat ELICIT as a real module

If your day includes FEAST I, it includes an English listening test. Candidates whose first language is English routinely skip preparing for it and drop marks that were genuinely available. Practise holding several values from one spoken message, and practise on accents you find difficult.

EUROCONTROL's position on retaking FEAST

This is worth knowing before you sit it, not after. EUROCONTROL advises taking FEAST only once, and states that a candidate who re-sits after an initial failure may have a higher probability of failing during subsequent training.

Retesting policy itself is set by the organisation you applied to, not by EUROCONTROL, and policies vary — some allow a re-sit after a defined period, some do not. Ask the organisation, and treat your first sitting as the one that counts.

What preparation cannot do

It cannot give you the questions — the item bank is not public. It cannot give you a target score — each provider sets its own cut score and none publishes it. What it can do is remove every avoidable loss: unfamiliar formats, untrained speed, an untested attention span, and a listening module you did not take seriously.

FAA controller eligibility is unusually strict, and unusually clear. Unlike the test content, the rules are published — and they are absolute. There is no route around the age limit and no discretionary exception to citizenship.

The age limit, and the trap inside it

Applicants must be under 31. The detail that catches people out is how that is measured: the cutoff is generally applied to the point at which you would begin training at the FAA Academy, not the date you clicked apply.

The practical consequence is that turning 31 during a hiring cycle can end an application that was live when you submitted it. If you are close to the limit, the cost of waiting for a more convenient bid window is your eligibility.

The core requirements

How the bids work

The FAA does not accept applications continuously. It opens vacancy announcements — bids — on USAJOBS, sometimes for only a few days, and you can only apply while one is open. Missing a window means waiting for the next one, which is where the age limit does its damage.

Different bids target different pools: some are open to the general public, some to graduates of collegiate training programmes, some to candidates with prior military or federal controller experience. Read which pool a bid is for before spending time on it.

Where the ATSA sits

The ATSA is not the application. You apply to a bid, the FAA screens for eligibility, and eligible candidates are referred to the assessment. Only after that do you sit the Air Traffic Skills Assessment — and only candidates whose result falls in the higher score bands go forward.

What the FAA does not publish

The full ATSA subtest specification is not public. Neither is the scoring model, nor the score needed to reach a given band. Widely-circulated breakdowns of the subtests come from preparation vendors and candidate recollection rather than from the FAA, and should be read that way.

FAA controller hiring runs as a sequence of gates. Each one removes candidates, and clearing one does not carry any credit into the next. This is the order, and what each step is actually for.

The steps

1. 1A vacancy announcement opensThe FAA posts a bid on USAJOBS. Windows can be short — days, not months — and you cannot apply outside one. Different bids target different pools: the general public, collegiate training graduates, or candidates with prior experience.
2. 2You apply and are screened for eligibilityAge, citizenship, and the education-or-experience requirement are checked. This screen is mechanical: if you do not meet a rule, the application ends here.
3. 3You are referred to the ATSAEligible candidates are invited to sit the Air Traffic Skills Assessment. Referral is not selection — it is permission to take the test.
4. 4You sit the ATSAA computer-based assessment of roughly three and a half hours, combining aptitude subtests with a scored biographical questionnaire.
5. 5Your result is bandedResults are reported in bands rather than as a raw mark. Only the higher bands are considered for selection.
6. 6A tentative offer letter (TOL)Candidates in the higher bands may receive a tentative offer. It is conditional on everything that follows.
7. 7Medical, security and drug screeningA medical examination, a background and security investigation, and drug testing. Any of these can end the process after a tentative offer.
8. 8A final offer and the AcademyA firm offer, then training at the FAA Academy in Oklahoma City. The Academy is itself a gate — passing it is not automatic.

The ATSA is the sift, not the finish

The FAA states that it is looking for the candidates who demonstrate the highest aptitude as measured by the ATSA, and that only candidates with a well-qualified result are considered for selection. A merely qualified result is a pass in the ordinary sense and still does not move you forward.

That is the single most important thing to understand about preparing: the bar is not passing, it is placing. You are not trying to clear a line, you are trying to finish above other candidates who also cleared it.

The biographical questionnaire

The ATSA is not purely cognitive. It carries a biographical questionnaire that is scored independently and contributes to the composite result. It asks about your history, work style and preferences.

It behaves like a personality instrument: it revisits the same underlying ground from different angles across many items, so an invented profile tends to come out inconsistent. Answer it honestly and at pace.

How long the whole thing takes

The FAA does not publish a guaranteed end-to-end timeline, and it varies with the bid, the volume of applicants and how quickly medical and security clear. Candidates routinely report waits of months between steps, particularly between the ATSA result and a tentative offer.

What the FAA does not publish

The full subtest specification, the scoring model, the score boundaries between bands, and the item bank. Hiring targets and the broad structure are published; the internals are not.

The ATSA runs about three and a half hours and covers several distinct tasks. The FAA names the broad areas — decision-making, spatial awareness, multitasking, working memory — but does not publish a subtest-by-subtest specification.

Numbers and differences

A number appears. Then another. You report the difference between the one you are looking at and the one before it — while the next number is already on screen. The task is simple arithmetic wrapped around a memory load: the sum is easy, holding the previous value while the current one competes for attention is not.

What it measures: working memory under interference, and arithmetic that stays accurate when it is not the only thing you are doing.

Variables

You are given values for a few variables — A, B, C — and asked to hold them while questions build on them. Later items combine or transform the values rather than simply reading them back.

What it measures: holding several values at once and operating on them, which is the everyday shape of controlling traffic — a callsign, a level, a heading and a squawk, all live at the same time.

Spatial and visual relationships

An aircraft, a viewpoint, and a question about direction. The difficulty is the frame of reference: left from the cockpit is not left from the tower, and switching between the two under time pressure is where candidates lose marks.

What it measures: spatial orientation and perspective-taking — mentally standing somewhere other than where you are.

The ATC simulation

The subtest that most resembles the job. Aircraft move across a display and you keep them apart, while a stream of arithmetic runs alongside and also has to be answered.

What it measures: multitasking with a genuine cost to dropping either task. Candidates who protect the traffic and abandon the arithmetic score badly, and so do candidates who do the reverse. The scoring rewards an even hand.

Logical reasoning

Two parts in practice. The first gives you a statement and asks which assumption it rests on, or what follows from it. The second is an arrangement puzzle — people or objects placed around a table or in a row, with a set of constraints, and a question about who sits where.

What it measures: deduction from stated rules, and the discipline of using only what you are told rather than what seems likely.

Reading comprehension

Passages followed by questions on what the passage actually says. The trap is consistent: the wrong answers are plausible, true-sounding statements that the passage does not support.

What it measures: reading precisely and at pace, and resisting the pull of an answer that sounds right.

The personality questionnaire

A forced-choice or rating instrument covering work style, temperament and how you operate under pressure. It is scored and it contributes to the composite result.

It cannot be usefully gamed. The same traits are approached repeatedly from different directions, and a constructed profile comes out inconsistent — which is more visible than any single answer.

How to prepare across all of it

• Train against a clock from the first session. Every cognitive subtest here is speeded.
• On the dual-task subtests, practise keeping the secondary task alive. That is where the marks are lost.
• On reading and logical reasoning, practise answering only from what is stated. Both punish inference.
• Sit at least one full-length session. Three and a half hours is a stamina test as well as an ability test.
• Answer the personality questionnaire honestly and quickly.

NAV CANADA’s first real gate is an online assessment you sit at home. It is short — around 30 minutes in total — and it decides whether you are invited to the in-person day.

The four sections

The assessment is built on Aon’s cut-e instruments, which is why the same section names appear in European selection processes.

The deductive section is the one Aon brands gapChallenge, and it is adaptive — the difficulty moves with your performance rather than following a fixed script. The inductive section gets harder as you progress through it.

Each section carries its own bar

This is the detail that changes how you prepare. A strong overall score does not rescue a weak section: candidates are commonly reported to need a minimum mark in each section to advance. One collapse ends the application even if everything else was excellent.

So preparation should be levelling rather than optimising. Find your weakest of the four and spend your time there, however unsatisfying that is compared with getting better at the one you enjoy.

Section by section

Inductive reasoning

Two worked grids are linked by a rule. The rule is usually a transformation — a rotation, with symbols swapped for other symbols on a fixed mapping. Your job is to recover it from the pair, then find which two of the four candidate grids are linked by the same rule.

• Recover the rotation first, then the symbol mapping. Doing both at once is where people stall.
• Early items tend to be pure rotation with the symbols untouched. Later ones move every symbol.
• Check a candidate pair against the rule rather than looking for one that ‘feels’ similar.

Ability to concentrate

Two minutes, a very simple judgement, repeated at speed. There is nothing to work out — the difficulty is entirely in not making careless errors while going fast, and in not slowing down as the seconds run.

• Settle on a rhythm in the first ten seconds and hold it.
• Do not re-check. A second look costs more than the error it prevents at this speed.
• Watch for the near-miss variants — the almost-right item is the whole test.

Spatial orientation

Two instruments tell you where you are. A direction indicator gives your heading; a relative bearing indicator points to a beacon relative to your nose. From those you work out which aircraft is pointing your way and where you sit relative to the beacon.

The step that catches people is the reciprocal: the beacon’s direction from you is not your direction from the beacon. Practise that flip until it is automatic rather than calculated.

Deductive reasoning (gapChallenge)

A grid of symbols where each symbol appears exactly once in every row and once in every column — Sudoku with shapes. One cell holds a question mark, and you deduce what must go there by elimination across its row and column.

• Work the row and the column of the target cell only. Solving the whole grid wastes time you do not have.
• Eliminate rather than search. Ask what cannot go there, not what might.
• Because it is adaptive, a run of hard items means you are doing well. Do not read it as failure and panic.

Practical points

• Sit it on a desktop or laptop, not a phone or tablet.
• It is timed and it runs remotely — sort your connection, browser and a quiet hour first.
• If you are unsuccessful, NAV CANADA publishes a one-year wait before you can reapply.

The second gate is a full day at an area control centre. NAV CANADA publishes the shape of it: approximately six hours, held at its area control centres, assessing memory, working speed, spatial visualisation, thinking and reasoning, attention, information processing and simple mathematics.

What the day contains

NAV CANADA names the abilities; EUROCONTROL names the modules. Putting the two together:

Note the two-year wait. It is double the online assessment’s, and it is the reason this day deserves more preparation than the 30-minute screen that got you here.

What FEAST means in practice

FEAST is modular. FEAST I is the cognitive half — a computer-based battery EUROCONTROL calls CLASSIC PLUS — sat alongside ELICIT, its English listening and comprehension test. FEAST II is the work-sample half, RADAR and MULTI-PASS, which simulate controlling traffic. FEAST III is a personality questionnaire.

Which modules appear on your day is NAV CANADA’s decision, not EUROCONTROL’s — not every organisation using FEAST uses every module. A six-hour day is consistent with FEAST I plus work samples.

EUROCONTROL publishes two things about FEAST I that are worth carrying into the room: it asks candidates to work as quickly and precisely as possible, and it states that completing every item is not required to pass. On a speeded test, that makes abandoning a hard item the correct move rather than a failure.

The same-day interview

Candidates who do well on the testing are commonly invited to a short pre-screening interview on the same day — around fifteen minutes. It is brief, but it is not a formality, and turning up with nothing prepared because you were focused on the tests is an avoidable loss.

Have ready: why air traffic control specifically, why NAV CANADA, what you understand the job to actually involve, and your position on relocating and on shift work. Short, concrete answers.

How to prepare

• Start with EUROCONTROL’s own free FEAST training platform. It shows you the formats, and it is official.
• Train stamina, not just ability. Six hours is a different exercise from a 30-minute screen, and hour five is where prepared candidates separate.
• Do not skip English listening. ELICIT is part of FEAST I, and first-language English speakers routinely drop marks there.
• Practise multi-tasking with a genuine secondary task you are not allowed to abandon. The work samples score you on the task you dropped.
• Eat and sleep like it is a working day, because it is one.

Clear the six-hour testing day and three gates remain. They test different things from everything before them — the aptitude question is settled by this point, and what follows is about judgement, communication and clearance.

The assessment centre

NAV CANADA publishes that this is a half-day session containing group and individual exercises as well as an interview, designed to assess skills, abilities and attitudes. Bilingual candidates in some locations complete additional assessments.

The three-year reapplication wait attached to this stage is the longest in the process — longer than the online assessment’s one year and the in-person test’s two.

Group exercises: what is actually marked

Group exercises are widely misread as competitions. They are not — assessors are marking how you operate with other people, and the candidate who dominates a group discussion usually scores worse than the one who moves it forward.

• Bring other people in by name. Noticing the person who has not spoken is marked; talking over them is also marked.
• Make your reasoning audible. An assessor cannot credit a conclusion you reached silently.
• Change your mind out loud when someone makes a better point. Updating on evidence is a controller trait, not a weakness.
• Watch the clock for the group. Someone has to, and it is rarely the loudest person.
• Disagree with the argument, never with the person.

The interview

Expect competency questions — evidence from your own history rather than opinions about the job. Answer in STAR: the situation, the task you held, the action you personally took, and the result.

The most common failure is the missing result. Candidates give a situation and an action and then stop, which loses the part that actually shows judgement. Say ‘I’ rather than ‘we’: the assessor is marking you, not the team you were in.

Have concrete examples ready for: working under pressure, catching your own mistake, handling a difficult colleague, making a decision without complete information, and holding to a procedure when it was inconvenient.

Language testing

Where a role requires it, NAV CANADA runs language testing by phone with a third party. English is required for all regions; English and French are required for Quebec and Ottawa positions.

It is a real assessment of operational communication, not a formality. Speak clearly and at a measured pace — a phone line strips out everything you would normally get from a face-to-face conversation, which is exactly the point.

Selection and pre-employment conditions

An offer comes with conditions attached, and NAV CANADA publishes them: medical and licensing requirements, security clearance, and drug testing — which includes cannabis. Legal status in Canada does not exempt it from the testing regime for a safety-critical role.

How long the whole process takes

Commonly reported as up to around four months end to end, and longer where the next training intake is further out. NAV CANADA’s general recruitment timelines for other roles do not apply to controller hiring — do not plan against them.

NAV CANADA publishes its training durations and pay in more detail than most air navigation service providers, which makes this one of the few parts of ATC selection where you can plan against real numbers.

Two streams, two lengths

You do not simply choose your stream. Placement follows operational need and how you perform in training, and NAV CANADA requires willingness to relocate as an eligibility condition — which is a real commitment, not a box to tick.

Training pay

Training is paid and full-time. NAV CANADA publishes a starting rate of about $58,606, rising to about $61,337 after one year of training, before the substantial step up on licensing.

The gap between training pay and licensed pay is the shape of the whole career: a long, demanding, well-supported apprenticeship followed by a salary that reflects how few people complete it.

Training is a gate, not a formality

This is the part candidates most often skip past. Selection gets you into training; it does not get you a licence. NAV CANADA states plainly that meeting the pre-employment conditions does not guarantee training completion, licensing or employment.

Controller training washes people out everywhere in the world, and it does so late — after months of paid, full-time work. That is not a reason to avoid it. It is a reason to arrive with the underlying abilities genuinely trained rather than just sufficient to pass a screen.

What comes with the job

• Shift work, including nights and weekends. Traffic does not keep office hours.
• Ongoing medical certification. It is not a one-off at hiring.
• Relocation, potentially more than once.
• Continuous assessment. Validation on a position is not the end of being checked.

If you are applying in more than one country — and many candidates do — the first useful question is how much of your preparation carries across. The answer is: more than you would expect, because two of these four are the same test.

The processes at a glance

The big overlap: NATS and NAV CANADA both use FEAST

NAV CANADA’s in-person day is EUROCONTROL’s FEAST exam. NATS uses FEAST as part of its own process too. Around 50 organisations across 40 countries use it, including DFS in Germany, skyguide in Switzerland, ENAV in Italy and the IAA in Ireland.

The practical consequence is large: preparing properly for FEAST prepares you for a Canadian application, a UK application and most of continental Europe at the same time. If you are applying broadly, FEAST is where preparation time compounds.

The odd one out: the FAA

The ATSA is its own instrument and does not share content with FEAST. The abilities beneath it are the same — working memory, spatial orientation, multitasking, deduction — so ability training transfers even though the formats do not.

What does not transfer is the eligibility. The FAA’s under-31 age limit has no equivalent in the UK or Canada, and it is measured against entry to the Academy rather than the application date. If the US is on your list, its timing constrains everything else.

How they differ in what they publish

EUROCONTROL is the most open about test structure, and is the only one of the four running a free official practice platform. NAV CANADA is the most open about the process and what the job pays. NATS and the FAA publish the least about content.

Cost of failure: the reapplication rules

Worth knowing before you sit anything. NAV CANADA publishes waits of one year after the online assessment, two after the in-person test and three after the assessment centre. EUROCONTROL advises sitting FEAST only once, and notes that re-sitting after a failure may carry a higher probability of failing later training.

For the FAA, the binding constraint is not a waiting period but the age limit — every cycle you spend not being selected is a cycle closer to ineligibility.

What to train if you are applying to more than one

• Working memory under interference — holding values while something else demands attention. Every one of the four tests it.
• Spatial orientation — reading a heading and transforming a viewpoint. Tested by all four.
• Multi-tasking with a scored secondary task — the trait every work sample is built around.
• Deduction from stated rules — grid puzzles in Europe and Canada, verbal and arrangement reasoning in the US.
• Speeded accuracy — all four are speeded, and careful-but-slow scores badly everywhere.
• English listening — explicitly assessed by NATS and by FEAST’s ELICIT, and useful everywhere else.

Search for air traffic control test preparation and you will find pages stating question counts, time limits and pass marks with total confidence. Most of those numbers are not published by anyone. This page sets out what is actually public, so you can tell the difference.

NATS

Public: that Stage 1 is a battery of eleven online ability tests; the five ability areas it covers; that Stage 2 is a situational judgement test and a listening comprehension test proctored by Aon; that Stage 3 is a virtual assessment centre with an interview, a knowledge test and observed retesting; and that Stage 4 is a practical exercise over Teams or at Whiteley.

Not public: individual test names, question counts, time limits, scoring formulas, progression thresholds, the Stage 2 question bank, the Stage 3 knowledge-test syllabus, which tests are repeated at Stage 3, the Stage 4 task rules and scenarios, and any pass mark at any stage.

EUROCONTROL

Public: the FEAST module structure, roughly how long each takes, and which broad abilities each measures. EUROCONTROL also runs a free official preparation platform — and states openly that its practice items are not identical to the real tests.

Not public: the item bank, the scoring, and each individual provider's cut scores, which vary between the roughly 50 organisations using FEAST.

The FAA

Public: that the ATSA exists, is taken after referral from a hiring bid, and covers decision-making, spatial awareness, multitasking and working memory. Hiring targets are published.

Not public: the full subtest specification. Two of the seven subtests are not publicly identified, and the widely-circulated breakdown comes from preparation vendors rather than the FAA.

How to read a preparation source

• Does it distinguish between what the provider published and what it inferred? If not, it does not know the difference either.
• Does it claim to have the real questions? Nobody outside the provider does.
• Does it state a pass mark? None is published for any of these tests.
• Does it explain its answers, or just mark them? On judgement tests especially, an unexplained answer key is an assertion, not teaching.