Crisis Signal Intelligence
How the Iran–Israel–US conflict
disrupted global air travel
Conversation intelligence from 19,843 real conversations across 19 airlines and 138 countries, February 26 to March 12, 2026
Conversations analyzed
19,843
Airlines tracked
19 carriers
Data window
Feb 26 – Mar 12 '26
Countries covered
138 markets
01

How We Built This Report

We built this report using DeepDive, analyzing 19,843 real public conversations from X (Twitter) across 138 countries between February 26 and March 12, 2026. Every data point is drawn from what people actually said publicly, with no surveys or estimated figures.

The full collection parameters are below.

Data Source
Public Social
Posts, replies, and threads from X (Twitter). Public conversations only, no private data, no DMs.
Carriers Monitored
19 Airlines
Emirates, Qatar Airways, Etihad, Air India, British Airways, Air France, American Airlines, United Airlines, Lufthansa, Turkish Airlines, Singapore Airlines, KLM, Cathay Pacific, Saudia, Japan Airlines, Delta, Virgin, Swiss, Biman Bangladesh.
Volume
19,843
Total conversations processed: parent posts, replies, and threaded comments in the 14-day window.
Time Window
14 Days
February 26 – March 12, 2026. Captures pre-crisis baseline, detonation, surge, sustained impact, and early recovery phases.
Sentiment
AI-Scored
Each post scored 0–100 with confidence weighting. Topics, intents, and trend keywords extracted via DeepDive's NLP classification pipeline.
Geography
138 Countries
Geolocated by profile metadata, post location tags, and contextual inference. US, India, UK, France, UAE, Qatar, Pakistan, Japan, Australia, and 129 additional markets.

Operational figures cited in this report (total flights cancelled, airport capacity percentages, cargo rate changes) are sourced from aviation industry trackers including Flightradar24, Cirium, and Rotate, as referenced within the social conversations we analysed. These figures are attributed to their originating sources throughout the report, not presented as DeepDive calculations.

This report is part of DeepDive's Pro Bono Intelligence Series, periodic public releases that demonstrate the conversation intelligence available to brands, agencies, airports, and operators on the DeepDive platform. The full dataset and real-time monitoring dashboard are available to DeepDive subscribers.


02

The Largest Aviation Crisis Since COVID-19

On February 28, 2026, Iranian ballistic missiles struck infrastructure near Dubai International Airport. DeepDive captured what followed in real time: airspace closures cascading across the Gulf and the aviation industry entering its most acute disruption since COVID-19.

6,908
conversations about flight cancellations in 14 days
34.8% of all brand posts
3,154
flight safety concern posts logged across 19 carriers
15.9% of all posts
1.4M
total social engagements tracked across all conversations
reactions + retweets
2,105
stranded passenger conversations logged in the dataset
10.6% of all posts

DeepDive processed 19,843 conversations spanning 19 major carriers across 138 countries between February 26 and March 12, 2026, capturing official airline statements, passenger escalations, real-time aviation tracking posts, and repatriation coverage simultaneously.

The numbers above are DeepDive signal metrics derived directly from public conversation data. Operational figures cited elsewhere in this report (total flights cancelled, airport capacity levels) are sourced from aviation tracking services (Flightradar24, Cirium) as referenced in the social conversations we tracked, and are attributed accordingly.

🚨
Highest-Engagement Signal · Feb 28 · 57,348 engagements
"Dubai just shut down. The busiest international airport on earth. Closed. Indefinitely. Over 280 flights canceled. 250 more delayed. The airspace that handles more international passengers than any hub on the planet went dark this morning."
✈️

34.8% of All Airline Conversation Was About Cancellations

More than one in three posts mentioning any tracked airline referenced flight cancellations, a signal density that dwarfs normal baseline levels and confirms the scale of the operational collapse.

🗺️

138 Countries. One Crisis.

Social signal originated from 138 countries simultaneously, confirming this was never a regional story. It hit every major travel market: US, India, UK, France, UAE, Qatar, Pakistan, Japan, Australia and 129 more.

📡

1.4M Engagements = Unprecedented Reach

Total reactions and retweets across the dataset reached 1,427,991, indicating this was not just high in volume but high in amplification. Every post was being seen and shared far beyond its original audience.


03

A Timeline of the Iran–Israel Aviation Disruption

Our analysis maps the crisis across four distinct phases, from near-silence before February 28, through a sustained anger peak on March 9, to the first fragile signs of recovery that followed.

Daily Conversation Volume & Negative Sentiment: Feb 28 to Mar 12

Volume bars (left axis) overlaid with negative sentiment % line (right axis)
0 1000 2000 3000 0% 25% 50% ANGER PEAK 44% · Mar 9 DUB CLOSES 28 Feb 2 Mar 4 Mar 6 Mar 8 Mar 10 Mar 12 Mar
Gray bars = daily conversation volume (left axis)Red line = % of posts with negative sentiment (right axis)
Source: DeepDive · 19,843 conversations

The anger curve peaks 9 days after the operational shock, driven by refund failures, not the original closures.

The negative sentiment curve reveals something counterintuitive: the initial shock on Feb 28 produced only 28% negative posts, passengers were confused and alarmed, but not yet angry. Anger peaked at 44% on March 9, eleven days into the crisis, driven primarily by refund denials, unreachable customer service lines, and the news of British Airways extending cancellations through year-end. The original crisis grounded flights; the aftermath damaged brands.


04

How Three Gulf Hubs Shut Down in One Night

We tracked all three shutdowns as they unfolded. Three airports, three home carriers, near-simultaneous closures. The conflict struck the connective tissue of global aviation, and our data captured every signal in real time.

The sequence mattered. Dubai International and Al Maktoum suspended operations simultaneously: two airports, zero warning. Qatar Airways' official airspace closure statement followed 2.5 hours later. By the time Etihad began its Abu Dhabi shutdown, the story had already been amplified millions of times. Per Flightradar24 data cited widely across the conversation, over 21,000 flights were cancelled across seven Gulf airports in the 14 days following February 28, with Dubai International operating at approximately 85% below normal departure capacity at the worst point.

† Operational figures (21,000+ flights, 85% capacity reduction) cited from Flightradar24 and Cirium data as referenced within tracked social conversations. These are industry tracker figures, not DeepDive calculations.

05

What 19,843 Crisis Conversations Reveal

Topic and sentiment distribution across 19,843 posts gives us the clearest view of where the crisis actually lived, and where brand damage was most concentrated.

Crisis Conversation Topic Distribution

Relative share of posts per topic category · posts can carry multiple topics
34.8%
Flight Cancellations
34.1%
Airline Comms & Updates
17.4%
Customer Service
15.9%
Safety Concerns
15.6%
Rebooking & Refunds
10.6%
Stranded Passengers
9.1%
Travel Disruptions
5.4%
Delays
Source: DeepDive topic classification · 19,843 posts · note: posts carry multiple topics, totals exceed 100%

Brand Exposure vs. Perception: The Crisis Scatter

X axis: total brand mentions (√ scaled) · Y axis: avg sentiment score · size = relative exposure
Low exposure · resilient High exposure · managed Low exposure · damaged High exposure · damaged 0 25 40 55 Sentiment Score Brand Mention Volume (√ scaled) Low Medium High Qatar Emirates Etihad Air India British Air France American United Lufthansa Turkish Delta Singapore JAL Cathay KLM
Critical sentiment (<35) Stressed (35–48) Resilient (>48) Dot size reflects relative exposure magnitude
Source: DeepDive · sentiment scoring × mention volume · 19 carriers · Feb 26–Mar 12, 2026

The 42-point gap between Qatar Airways (12.8) and Japan Airlines (54.9) is the clearest measure of how differently this crisis was experienced by carriers depending on geography, communication, and crisis response quality. The 35-point threshold separates carriers that sustained severe brand damage from those that managed relatively intact. Only three carriers sit below it, all Gulf-based, all hub-dependent.

DeepDive tracks this live. Sentiment scores like these are available in real time on the platform, not 14 days later in a report. Your team can watch brand perception shift as a crisis unfolds and respond accordingly.
See the platform →

06

Brand Exposure and Sentiment Damage Across 19 Airlines

All metrics in this section come directly from DeepDive's social signal dataset. Note that Cancellation Conversation Share is not a flight cancellation rate; it measures the percentage of each carrier's total social mentions tagged as flight cancellation discussions.

Cancellation Conversation Share by Carrier

% of each carrier's total brand mentions that were tagged as flight cancellation discussions · sorted highest to lowest
Etihad Airways
63.9%
Emirates
59.2%
Lufthansa
54.7%
British Airways
53.7%
Saudia Airlines
52.8%
Qatar Airways
45.2%
Air France
36.0%
Air India
34.0%
Turkish Airlines
23.1%
Singapore Airlines
18.2%
Japan Airlines
12.5%
Gulf hub carriersEuropean carriersResilient / Asia-Pacific
Source: DeepDive · cancellation conversation share = (cancellation-tagged posts / total brand mentions) · not an operational flight cancellation rate
AirlineTotal MentionsCancel Conv. Share ①Safety PostsStranded PostsRelief PostsAvg Sentiment /100
Qatar Airways4,71145.2%82893346312.8
Etihad Airways1,19163.9%2311859531.3
Emirates2,41059.2%66037716034.3
Air India2,55734.0%40533417841.9
Lufthansa40654.7%136322342.8
Air France2,04336.0%4582614844.6
British Airways2,04453.7%6061266844.7
Delta Air Lines63916.9%5035545.4
Saudia Airlines42852.8%9728546.4
American Airlines2,2395.5%131402546.6
KLM Airlines51542.7%9430946.6
United Airlines2,0514.7%93451047.4
Turkish Airlines88223.1%13733749.4
Singapore Airlines59218.2%9217651.8
Japan Airlines40912.5%279354.9

① Cancellation Conversation Share = (posts mentioning airline tagged as 'flight cancellations') ÷ (all posts mentioning that airline). This is a share-of-voice metric, not an operational flight cancellation rate.


07

Carrier-by-Carrier Brand Performance Analysis

Our team analyzed the communication and operational decisions of each major carrier throughout the crisis, identifying exactly what drove brand outcomes and what the data says should have been done differently.

Qatar Airways

12.8/100
CRITICAL · Worst in Dataset
What Went Wrong
  • Full airspace closure with no pre-emptive passenger advisory; news broke on social before official channels
  • 933 stranded passenger posts (highest of any carrier), indicating mass communication failure to in-transit travellers
  • No flexibility policy announced within first 24 hours, forcing passengers to escalate publicly
  • Relief-to-cancellation ratio of only 1:4.6; far too little positive narrative to offset damage
What Went Right
  • Official statement published at 15:47 on Feb 28, among the first formal airline communications
  • Operated relief flights from neutral hubs (Muscat, Riyadh) before Doha reopened
  • Ran repatriation routes on March 7 to London, Paris, Madrid, Rome and Frankfurt, demonstrating operational flexibility
  • 463 relief posts tracked; significant recovery narrative despite the headline damage
What Should Have Been Different
  • A passenger-first communication blitz within the first 2 hours (SMS, app push, email) before social media broke the story
  • Proactive rebooking portal announced same day to prevent phone line overload
  • Framing every repatriation flight as a named, counted operation with public announcements to build relief narrative
  • Earlier deployment of neutral-hub operations, Muscat and Riyadh options could have been announced Feb 28 night
Total mentions: 4,711Cancellation conv. share: 45.2%Stranded posts: 933Safety concern posts: 828Relief posts: 463

Emirates

34.3/100
SEVERELY STRESSED
What Went Wrong
  • Aborted approach footage and smoke near Terminal 3 went viral (2,766 engagements) before any official Emirates communication
  • Highest raw cancellation volume in dataset driven by Dubai's complete shutdown
  • Repatriation capacity was far smaller than demand; passengers who couldn't access relief flights became loudest critics
  • Second negative wave when limited flight resumptions were overbooked and passengers were again left behind
What Went Right
  • Ran repatriation and freighter ops from March 3–4, earlier than most Western carriers
  • Public announcement of 60% network recovery target by March 7 (106 daily round trips to 83 destinations) generated strong positive signal
  • Boeing 777 deployment for evacuation communicated publicly; the capacity announcement read as confidence signal
What Should Have Been Different
  • Social monitoring response: a real-time ops post acknowledging the aborted approaches while official investigation was underway would have closed the information vacuum
  • Publish daily operational capacity numbers (even if minimal) to anchor public expectations and prevent rumour amplification
  • Prioritise widebody repatriation with named aircraft and passenger counts (the Air India model) to generate positive coverage
Total mentions: 2,410Cancellation conv. share: 59.2%Stranded posts: 377Safety concern posts: 660

British Airways

44.7/100
STRESSED · Late Damage
What Went Wrong
  • Year-long Abu Dhabi cancellation announcement became the single most shareable negative airline story in the late-phase data
  • Announcement framing ("not next week, not next month, the rest of the year") read as a business decision, not passenger-care, triggering viral backlash
  • EU261 compensation claim conversation exploded and sustained: passengers found the refund process difficult, escalating publicly
  • No visible repatriation operation in data despite having passengers stranded at Gulf connections; missed brand moment
What Went Right
  • Early disruption notifications for affected passengers on initial cancellation days
  • Route-specific cancellation updates were clear and timely
  • Relatively lower stranded passenger count (126) vs. total mentions suggests connections management was adequate in early phase
What Should Have Been Different
  • Abu Dhabi cancellation framing: lead with passenger flexibility and rebooking options, not the duration of the suspension
  • Proactive EU261 guidance published on BA's channels before passengers had to ask; this would have dramatically reduced public escalation
  • Even a small, named repatriation operation for passengers stranded at Gulf transfer points would have created a positive counter-narrative
Total mentions: 2,044Cancellation conv. share: 53.7%Stranded posts: 126Safety concern posts: 606

Air India

41.9/100
TEMPLATE FOR CRISIS RESPONSE
What Went Wrong
  • Sentiment still sub-50 because initial cancellations generated frustration before repatriation narrative built momentum
  • App and booking system complaints surfaced under the crisis posts; technology failure compounded disruption experience
  • Some passengers reported difficulty reaching customer service during peak disruption period
What Went Right
  • Highest relief post count of any carrier: 178, building sustained positive narrative throughout crisis period
  • Flight AI916D homecoming: named aircraft (VT-EDC), passenger count (149), official timestamp (1058 hrs IST): textbook crisis PR
  • Boeing 777 widebody deployment publicly announced; a capacity signal, not just logistics
  • Multilingual media coverage: Malayalam, Hindi, English simultaneously; reached diaspora communities directly
  • Official Air India Newsroom account drove the repatriation story actively, not reactively
What Should Have Been Different
  • Stronger proactive communication in hours 0–8 before repatriation began; the initial cancellation coverage drove the sentiment down
  • Tech resilience: customer service channel overload during peak was predictable and should have been pre-empted with chatbot capacity or callback systems
  • More repatriation frequency earlier; demand significantly exceeded supply in the first 72 hours
Total mentions: 2,557Cancellation conv. share: 34.0%Stranded posts: 334Relief posts: 178 (highest in dataset)

Singapore Airlines & Japan Airlines: The Insulated Carriers

51.8 / 54.9/100
RESILIENT
Where Exposure Existed
  • Both carriers still had some Gulf-routed flights that required cancellation or rerouting
  • Passengers booked on codeshare connections via Gulf hubs experienced disruption
  • Singapore Airlines' cancellation conversation share of 18.2%: not zero, but manageable
What Protected Them
  • Pacific and Indian Ocean routing provided structural insulation; primary trunk routes bypass Gulf airspace entirely
  • Clear communication about any rerouted or cancelled services maintained trust
  • The crisis served as implicit advertising: passengers publicly asking "why aren't SIA/JAL flights affected?" created unprompted competitive differentiation
  • Maintained operational continuity when competitors were paralysed, reinforcing "premium reliability" brand positioning
The Strategic Opportunity
  • Neither carrier visibly capitalised on the competitive narrative forming in public conversation: a missed brand moment
  • Proactive messaging about route security and alternative connectivity could have captured demand from travellers permanently reassessing Gulf hub reliance
  • Data shows travellers comparing routing options in real time; a targeted paid media push during the surge phase would have had exceptional timing
SIA mentions: 592SIA cancel conv. share: 18.2%JAL mentions: 409JAL cancel conv. share: 12.5%

08

Stranded Passengers and Public Anger in Their Own Words

We tagged 2,105 conversations as stranded passenger accounts. These are unfiltered, real-time records from people abandoned mid-journey, and some of the most direct brand-reputation signals in the entire dataset.

Qatar Airways accounted for 933 stranded passenger posts, 44% of the entire dataset's stranded volume, reflecting the combination of being the world's highest single-hub transit carrier and having its airspace closed without warning. The dominant emotional arc across the data ran: confusion → anger → relief. Each phase had a clear trigger: confusion from the communication vacuum, anger from refund failures, relief when repatriation flights finally appeared.

📢

Communication Failure Was the Crisis Within the Crisis

A significant share of negative posts wasn't about the conflict itself, it was about airlines' failure to communicate proactively. Passengers reported learning about cancellations from social media before their airline had reached them. DeepDive's data shows complaint posts spiking 8+ hours before most airlines issued formal statements.

💳

Refund & Rebooking Demand: 3,089 Posts (15.6% of All Content)

Passengers weren't just asking for refunds; they were publicly escalating after phone lines failed. "Full refund" appeared 161 times in trend keywords. EU261 compensation claims were referenced repeatedly in European-origin posts targeting BA, Lufthansa, and Air France, creating a long-tail brand problem that persists well after airports reopen.

Top Trend Keywords
airspace closurestranded passengersflight cancelledmiddle eastdubai airportrelief flightsrepatriation flightsairspace instabilityfull refundflights suspendedrebookingrebookdohaabu dhabifuel surchargewar riskcustomer servicelimited flights

09

Cargo Routes, Insurance Markets, and Long-Term Brand Damage

While the passenger story dominated headlines, our data surfaces two signals that received almost no coverage: a global air cargo crisis and a war-risk insurance withdrawal that will reshape Gulf aviation economics for years to come.

−18%
Global air cargo capacity decline in the week following Feb 28, per Rotate data cited in conversation
† Per Rotate / supply chain signal tracked in dataset
+6%
Rate spike for SE Asia→Europe air freight, jumping to $3.82/kg within days of the closure
† Per supply chain tracking accounts in dataset
29
Qatar Airways freighters parked as Doha airspace closed, removing a major freight artery overnight
† Per FlightGlobal data cited in conversation
$1,500
War risk surcharge per TEU that began appearing in air freight quotes within 72 hours of the crisis
† Per supply chain tracking accounts in dataset

The data also surfaces an underreported signal: war-risk insurance withdrawal from Gulf airspace. Posts from aviation and insurance accounts note that insurers officially pulled war-risk coverage for Gulf routes, a structural shift that means British Airways' year-long Abu Dhabi cancellation is not just a safety decision, it's an insurance and economics decision. The conversation frames this as "Icarus economics": the Gulf hub model's efficiency depended on stability that no longer exists at prior insurance rates.


10

Which Airlines Showed Up During Repatriation

Repatriation became the defining brand-building moment of the entire crisis. DeepDive's signal data is unambiguous: carriers that operated named, publicly announced evacuation flights built lasting positive equity. Those that stayed silent are still paying for it.

🇮🇳

Air India: The Repatriation Template

178 relief-tagged posts (highest in the dataset). Named aircraft, passenger count, official timestamp, Boeing 777 widebody deployment announcement. Multilingual coverage across Malayalam, Hindi and English simultaneously. The homecoming narrative generated sustained positive signal for 8+ days, the longest positive arc of any carrier in the dataset.

🇶🇦

Qatar Airways: Flexibility Despite Grounding

Operated relief flights from Muscat and Riyadh on March 5 while Doha remained closed. Ran repatriation routes on March 7 to London, Paris, Madrid, Rome and Frankfurt. 463 relief posts tracked (the second-highest in the dataset), though overshadowed by 2,131 cancellation mentions. The ratio, not the volume, is where Qatar Airways needs to improve.

🌍

Western Carriers: The Absent Rescuers

British Airways, Lufthansa, Air France and KLM are almost entirely absent from the repatriation data. Their role in this crisis was defined by cancellations and refund demands. Passengers stranded at Gulf transit points on European carrier tickets had no visible rescue operation from their airline: a missed brand and goodwill opportunity that the data makes starkly clear.

🇦🇪

Emirates Recovery: 60% Network in 7 Days

Emirates' announcement that it would reach 60% network capacity (106 daily round trips, 83 destinations) by March 7 was the single most positive Emirates signal in the entire dataset. The specificity (named routes, named capacity targets) is what made it credible. Airlines that publish specific recovery milestones generate materially better coverage than those that say "operations are resuming."

DeepDive Platform
Track any brand's crisis as it happens.
DeepDive monitors airlines, airports, and travel brands across social, news, and conversational data in real time. See sentiment shifts the moment they emerge, not 14 days later in a report.

11

Five Crisis Intelligence Signals for Airline Brands

This data doesn't just tell you what happened. It tells you what's coming. The patterns we identified across 19,843 conversations will shape competitive positioning for carriers and airports in the months and years ahead.

  • 01
    The communication window in a geopolitical crisis is measured in hours, not days

    The 8-hour gap between airspace closures and airline communications on Feb 28 was the primary driver of passenger frustration, not the cancellations themselves. Carriers that issued proactive advisories within the first two hours generated neutral-to-positive responses. Those that waited for official confirmation before communicating were buried in complaint posts. In any future disruption scenario, the threshold for proactive passenger communication should be "information sufficient to act on," not "complete operational certainty."

  • 02
    Gulf hub dependency is now a publicly understood and actively discussed strategic risk

    The crisis permanently elevated "Gulf airspace" and "hub routing" as concepts in global consumer consciousness. Travellers, travel managers, and corporate buyers are now publicly discussing routing alternatives on social media. Singapore Airlines and Japan Airlines' operational insulation received explicit, unprompted comparative commentary, their Pacific routing is being framed as a competitive advantage. This is new. Carriers with alternative routing should be activating this narrative in their communications now.

  • 03
    Repatriation is a brand-building event. Treat it like one.

    Air India's evacuation flights generated the highest positive sentiment density of any carrier during the crisis period. The specific elements that drove this, named aircraft, exact passenger counts, official timestamps, multilingual coverage, Newsroom account as narrator, are all replicable. Airlines that treat repatriation as a logistics obligation are leaving significant brand equity on the table. Every evacuation flight is a story. The carriers that tell it will come out of this crisis with stronger brands than they entered with.

  • 04
    The refund experience determines long-term brand trust more than the cancellation itself

    3,089 rebooking and refund posts, 15.6% of all content, signal that the financial resolution of this crisis will outlast the operational one by months. Carriers whose refund processes force social media escalation face a second wave of brand damage long after airports reopen. The data already shows EU261 compensation claims running as an ongoing conversation for British Airways, Lufthansa, and Air France into mid-March. A proactive, transparent, frictionless refund process is not just a customer service issue, it is a brand management imperative.

  • 05
    Social media is now a frontline safety intelligence layer, not a PR channel

    Posts about aborted approaches, smoke near terminals, and planes circling in holding patterns appeared on social media hours before official communications. The Emirates Terminal 3 smoke post accumulated 2,766 engagements before any official confirmation. Airlines and airport operators that monitor social signal as part of their crisis intelligence infrastructure, not just their marketing function, will consistently have earlier warning, better situational awareness, and faster response capability than those who don't.