Everything the analysis covers

Four ways to load one flight. Loading another replaces it.

1. 1Flight lookup by number and dateFlight lookup retrieves the recorded ADS-B track for a flight number (e.g. BA123) and a UTC date.
2. 2Route searchEnter two airports as 3-letter IATA codes (for example LHR for London Heathrow) and a UTC date, then choose a flight from the results.
3. 3CSV uploadDrag and drop or pick a flight track CSV export with seven columns: Timestamp, UTC, Callsign, Position, Altitude, Speed, Direction. Parsed entirely in the browser and never sent to the server.
4. 4CSV validation and limitsExact seven-column header, with a message naming the expected columns; rows outside position, altitude, speed or direction bounds dropped and counted; duplicate timestamps removed; 10 MB and 100,000-row limits checked before parsing.
5. 5Live viewNeeds a flight lookupA flight still in the air shows where it is now: position, altitude, speed, progress along the route and estimated arrival. The full analysis opens once it has landed.
6. 6Sample flightA synthetic London–New York track generated to demonstrate every analysis. Not a real flight.
7. 7Multi-session detectionA file splits into sessions at gaps over 20 minutes, with a session selector that defaults to the longest flight. An airborne coverage gap within one continuous flight is not split.
8. 8UTC anchoringDates and times are UTC throughout; lookup dates are bounded by the server's UTC date, not the visitor's clock.

9. 9Nine-phase classificationIdle, taxi-out, takeoff roll, climb, cruise, descent, approach, landing roll and taxi-in, from height above the measured ground elevation, groundspeed and sustained movement. Approach begins below 5,000 ft above the arrival airport; the landing roll ends below 35 kt.
10. 10Nine event markersTaxi-out start, takeoff roll start, takeoff, top of climb, descent start, approach start, landing, taxi-in start and final stop.
11. 11Phase windowsEach phase as a bounded segment, driving the charts, the map and the tables.

12. 12Block timeFrom the start of sustained taxi to the final stop, or the last recorded position while still taxiing (then marked approximate).
13. 13Airborne timeTakeoff to landing.
14. 14Taxi-out durationStart of taxi to the start of the takeoff roll.
15. 15Taxi-in durationFrom below 35 kt after touchdown until groundspeed stays below 1 kt at the stand. When the recording ends while the aircraft is still taxiing, it runs to the last recorded position and is marked approximate.
16. 16Takeoff roll duration
17. 17Takeoff-to-cruise durationTakeoff to top of climb.
18. 18Cruise durationTop of climb to descent start.
19. 19Descent durationDescent start to approach start.
20. 20Approach durationApproach start to landing.
21. 21Total track distanceGreat-circle distance summed point to point, in nautical miles.
22. 22Ground distanceDistance covered on the ground.
23. 23Airborne distanceDistance flown between takeoff and landing.
24. 24Great-circle distanceThe shortest path between the resolved airports.

All speeds are groundspeed, not airspeed. Speed statistics use observed airborne samples only: interpolated points and position-derived estimates are excluded.

25. 25Maximum altitudeFrom the median-filtered altitude, so one corrupt sample cannot set it.
26. 26Cruise altitudeThe 70th percentile of level-flight altitude above the 18,000 ft cruise floor (lower for flights that never reach it), rounded to 100 ft.
27. 27Maximum speed at the 99th percentileDeliberately not the raw maximum, so one corrupted sample cannot define the headline figure.
28. 2895th-percentile speed
29. 2975th-percentile speed
30. 30Median airborne speed
31. 31Mean airborne speed
32. 32Average climb rateTakeoff to top of climb, in feet per minute.
33. 33Average descent rateDescent start to landing, in feet per minute.

34. 34Initial cruise altitude
35. 35Final cruise altitude
36. 36Step-climb detectionCount, and per step the altitude gain, duration and distance. Median filtering suppresses noise, short holds at the same level are merged, and changes under 800 ft are ignored. Step descents are recorded too.

Interpolated points are excluded, so a coverage gap cannot fabricate a circuit.

37. 37Holding pattern detectionA circuit is 270–450° of same-direction turning within 45–600 seconds that closes back near its start, above 200 ft and 60 kt, with no more than 2,500 ft of altitude variation.
38. 38Circuit countingPer hold and in total. Circuits less than five minutes apart form one hold.
39. 39Hold durationPer hold and in total.
40. 40Per-hold detailEach hold kept individually, with its altitude band, for display and mapping.

Analysed. The score starts at 100 and each finding deducts a capped penalty, so the grade always reconciles with the findings that produced it.

41. 41Efficiency score0–100.
42. 42Letter gradeA at 90 or above, B at 80, C at 70, D at 60, otherwise E.
43. 43Route extension percentageAirborne track distance above the great circle.
44. 44Excess track milesTrack beyond a routing allowance of 3% of the great circle plus 30 nm for departure and arrival procedures.
45. 45Severity gradingEvery finding rated good, minor, notable or significant.
46. 46Finding: track miles above great circle
47. 47Finding: extended taxi-outAgainst a general 15-minute reference, not the airport's own unimpeded taxi time.
48. 48Finding: airborne holding
49. 49Finding: no step climb on a long flightAssessed on flights airborne for more than three hours.
50. 50Finding: level-offs during descentLevel segments of a minute or more between top of descent and 5,000 ft above the arrival airport.

51. 51Day, civil twilight and night durationsAcross the airborne phase. Sunrise and sunset at −0.833° solar elevation; civil twilight down to −6°.
52. 52Day and night percentages
53. 53Daylight segmentsThe flight split into lit and dark stretches, with timestamps.
54. 54Sunrise and sunset crossingsDetected in flight, with position, altitude and UTC.
55. 55Solar azimuth at each crossing
56. 56Maximum solar elevation
57. 57Minimum solar elevation
58. 58Sun time on the left of the cabinFrom the bearing to the sun relative to the aircraft's track. Sun within 10° of the nose or tail counts for neither side.
59. 59Sun time on the right of the cabin
60. 60Best view sideThe side with more sun, or either when the two are within 10% of each other.
61. 61Departure and arrival local clocksLocal times with UTC offsets, from each airport's time zone.
62. 62Time-zone shift and red-eye detectionRed-eye: departs 21:00–02:00 local, arrives 04:00–10:00 local, at least three hours airborne and at least 40% of it in twilight or darkness.

Optional and started by you. Available for any loaded flight whose airports are resolved, including an upload.

63. 6330-day city-pair windowUp to 60 sectors sampled from the last 30 days on the same route. The loaded flight is never compared with itself.
64. 64Sample size reportingReturned rows, usable sectors, and live or incomplete rows excluded from every statistic.
65. 65Block time distributionp10, median and p90, from first to last ADS-B contact.
66. 66Airborne time distribution
67. 67Route extension distribution
68. 68Taxi-out distributionA proxy: first ADS-B contact to takeoff.
69. 69Taxi-in distributionA proxy: landing to last ADS-B contact.
70. 70Percentile placementWhere the loaded flight sits in each distribution, measured the same way as the sample. Withheld below five usable sectors.
71. 71Benchmark verdictA plain-language reading of each placement.
72. 72Operator shareWhich airlines fly the route, by frequency.
73. 73Operator countDistinct operators, plus sectors with no reported operator.
74. 74Aircraft type share and type count
75. 75Departure runway usageCommonest first.
76. 76Arrival runway usageCommonest first; diverted sectors excluded.
77. 77Diversion rate and distinct flight numbersDiversions are left out of the distributions but reported here.

Where the flight started and ended, and what was around it.

78. 78Gate to gateNeeds a flight lookupThe departure and arrival gates and the reported gate, takeoff and landing times, beside the times observed in the track.
79. 79Runway identificationThe runway used at each end: as reported for a lookup, otherwise matched from the track to the runway centreline, so uploads get it too.
80. 80Runway detailLength, width, surface, lighting, direction and threshold elevation, for about 10,000 runways worldwide.
81. 81Where on the runwayHow far along the runway the aircraft was airborne, or below 35 kt after landing, and how much runway was left. Read from the sampled track, so shown as upper bounds.
82. 82Airspaces crossedNeeds a flight lookupEach airspace the flight crossed, in order, with the time it entered and the time spent inside.
83. 83The aircraft's dayNeeds a flight lookupStarted by you. The other flights the same aircraft made around this one: where it came from, the time on the ground, and where it went next.
84. 84Departure trafficNeeds a flight lookupStarted by you. The other takeoffs from the same airport within half an hour of this one.

Optional and started by you. Needs a flight number — from the lookup, or derived from the callsign of an upload.

85. 85Weekly operating calendarThe last 28 complete UTC days, laid out in Monday-to-Sunday weeks.
86. 86Pattern descriptionA written summary of the days the flight was observed — from flight records, not a published timetable.
87. 87Per-day detailEach day marked operated, not observed, or outside the window.
88. 88Recent operationsEach recent operation's takeoff, landing, airborne time, aircraft and registration, with this flight against the median.

89. 89Route mapA dark, radar-style chart drawn on the page: coastlines, borders, country names and a 10° grid, with the runways at both airports.
90. 90Phase-coloured trackThe route drawn in flight-phase colours.
91. 91Altitude-weighted lineLine weight varies with altitude.
92. 92Great-circle overlayThe direct route drawn against the flown track.
93. 93Departure and arrival markers
94. 94Top-of-climb marker
95. 95Top-of-descent marker
96. 96Step-climb markers
97. 97Holding markers
98. 98Sunrise and sunset markers
99. 99Gate and airspace markersNeeds a flight lookupThe reported gates and each airspace boundary crossed.
100. 100Legend, hover inspector and track playbackPlayback animates progression along the route.

101. 101Vertical profileAltitude across the sector with phase bands.
102. 102Speed panelGroundspeed in its own panel, aligned beneath the profile on the same time axis rather than on a second y-axis.
103. 103Hover and keyboard inspectionAltitude, speed, direction and UTC at any point, by pointer or keyboard.
104. 104Phase time barsHow block time was distributed across the phases.
105. 105Score dial and sparklines

Type details come from the built-in catalogue once the aircraft type is known.

106. 106More than 80 type profilesTurboprops, regional jets, narrowbodies and widebodies.
107. 107RegistrationNeeds a flight lookup
108. 108Type codeReported with a lookup; for an upload, the type can be chosen.
109. 109Mode-S addressNeeds a flight lookup
110. 110OperatorReported with a lookup, including wet-lease livery versus operating airline; for an upload, derived from the callsign prefix.
111. 111Manufacturer
112. 112Engine count and engine type
113. 113Typical seats, MTOW and maximum range
114. 114Automatic type matchingFrom IATA or ICAO type codes, with a warning when the sector exceeds the type's published range.

115. 115About 8,800 airportsEvery airport with an IATA code: ICAO code, name, city, country, position, elevation and time zone.
116. 116Nearest-airport resolutionFrom the track ends, so uploads without airport fields still resolve. Low-confidence matches are flagged.
117. 117Lookup by IATA or ICAO code

118. 118Confidence score0–100 in high (80 and above), medium (55–79) and low bands, penalising gaps, interpolation, irregular or sparse sampling and inconsistent velocity data.
119. 119Gap detectionCount and longest gap. A gap is an interval longer than three times the track's normal sampling cadence, and at least two minutes.
120. 120Sampling intervalMedian seconds between observed points.
121. 121Gap interpolationSynthetic points on the great circle at 15–45 second spacing, at most 180 per gap — flagged, penalised, and excluded from speed statistics and holding detection.
122. 122Velocity consistency checkReported groundspeed or track that contradicts the movement between positions — a signature of GNSS interference — is replaced by a position-derived estimate for phases, holding and cabin-side sun, and excluded from speed statistics.
123. 123Position-spike removalA single point implying over 1,000 kt that returns to the track is removed.
124. 124Analysis notesSpecific issues in plain language — a gap during approach, and what it makes approximate.
125. 125Reported versus analysed valuesWhat came with a flight lookup, kept apart from what the tool computed.
126. 126Source attributionEvery analysis states whether it came from a lookup, an upload or the sample.

A full report of up to 18 pages or a four-page summary, and a square flight card. Pages without data are omitted rather than printed empty, and every page carries the informational-use footer.

127. 127Client-side generationBuilt in the browser, so the flight data never leaves the tab.
128. 128Full report or summaryThe A4 landscape report, or four pages: the cover, at a glance, the route map and the timeline.
129. 129CoverThe flight's own route on a map, its identity, the headline figures and a contents list.
130. 130At a glanceThe score, six headline figures, the shape of the flight and what stands out.
131. 131Route mapThe track coloured by phase over the world map, with numbered events, the distances, the airports and the airspaces crossed.
132. 132Gate-to-gate timelineEvery event in order, in UTC and local time, and the time spent in each phase.
133. 133Vertical profile
134. 134Climb, cruise and descentCruise levels, step climbs, the climb, and how much of the descent was continuous.
135. 135Efficiency
136. 136Where the time wentA waterfall from the time without the measured losses to the time as flown, then the route, the holding and the descent.
137. 137Ground operations
138. 138Airports and trafficGates, runways, and the departure and arrival traffic when looked up.
139. 139The aircraftIts type, its day and its last seven days, when looked up.
140. 140Light conditions
141. 141How it comparesEvery sampled sector as a dot, with this flight's place in plain words. Included when the benchmark has been run.
142. 142Route compositionOperators, types and runways from the benchmark.
143. 143Operating patternThe days it operated, with its usual takeoff, aircraft and time in the air over four weeks, when looked up.
144. 144AppendixThe method and provenance, data quality, and the specific ways figures can differ from an operator's own records.
145. 145Flight cardA square image of the flight, its route and four headline figures, with the logo, ready to post.

146. 146Analysis tabsOverview, Route & timing, Airports & aircraft, Efficiency, Daylight, and Assumptions & data.
147. 147KPI dashboardHeadline figures as dial, spark and stat tiles.
148. 148Table togglesEvery chart expands to its underlying numbers.
149. 149Aircraft panelIdentity and specification for the analysed aircraft.
150. 150Load-new-flight flowOne flight at a time: loading another replaces the current one. The route benchmark provides the comparison.

151. 151Light, dark and system themesRemembered per browser.
152. 152Accessibility and keyboard navigation
153. 153Responsive layoutsDown to phone width.
154. 154404 and error screensIn the product's own visual language.
155. 155About, Features and FAQ pages
156. 156Structured data and sitemap

157. 157Informational-use positioningStated in the workspace and on every slide of the report.
158. 158Observed, analysed and reported labellingWhat the track shows, what flightanalyser worked out and what the data source reported are labelled wherever they appear.
159. 159Named limitationsThe specific mechanisms by which the analysis can differ from an operator's records.
160. 160Local CSV processingUploaded files are parsed in the browser and never sent to the server.