Bankinter: fall of app and web affects users for hours on the 27th of October
On October 27, 2026, Bankinter, one of the main financial institutions of Spain, he experienced a complete crash of the mobile application and website that affected thousands of customers for several hours. Bankinter confirmed the incident publicly through his mind of X, apologizing to the customers and ensuring that their team was working to restore service. The service was gradually recovered after 3 PM, with operations fully normalized later. For banks and fintech in Latin america, this incident illustrates a principle critical: when your digital infrastructure fails, thousands of customers lost access to their money, and instantly lose trust in your institution.
What happened exactly on the 27th of October?
Bankinter experienced a technical breakdown that affected ‘some communications systems,’ resulting in:
- Mobile App inoperative: Users were unable to access accounts, transfer balances,
- Site web fallen: Access blocked web completely
- Duration: Problems reported during the morning, recovery began at around 3 PM
What was the cause?
Bankinter described the problem as ‘technical breakdown that affected some communications systems,’ but did not specify technical details are accurate on what component has failed, if it was software, hardware, infrastructure, network, or database.
What was the impact?
For clients:
- Lack of access to accounts
- Locked transfers
- Consultation of balances impossible
- Card payments potentially rejected
- Hours without being able to manage digital money
For Bankinter:
- Loss of trust and confidence of customers
- Reputational risk in social networks
- Potential customers migrand competitors
- Impact on regulatory obligations (report downtime to the authorities)
For the financial sector:
- Reminder of the vulnerability of operations, 100% digital
- Importance of redundancy and fast recovery
Why does this matter?
Unlike a retailer where downtime causes loss of sales, in a financial institution for the downtime it causes panic. Customers unable to access their money. Even if the incident only lasts for hours, the psychological damage is real: ‘how safe Is my money in this bank?’
How does one solve this?
Technically (for Bankinter and fintech similar):
- Geographic redundancy: Multiple datacenters in geographies different
- Automatic Failover: Switch to datacenter backup in seconds
- 24/7 monitoring: to Detect degradation before the fall full
- Transparent communication: Update customers every 15 minutes during the incident
Operationally:
- Rapid detection: Know WHAT failed in <1 minute
- Scaling automatic: right Team notified immediately
- Coordinated response: infrastructure Equipment, frontend, database, working in parallel
- External communication: Customers informed, not guessing
How can you help 24Cevent in this scenario?
In financial institutions, detect, and respond to falls quickly the difference between ‘incident of 30 minutes’ and ‘incident of 3 hours’:
1. Automatic detection of degradation (‘t wait to fail completely)
24Cevent continuously monitors:
- Latency of API (how to respond more slowly?)
- Error rate of requests (can reject more requests?)
- Database availability
- Connectivity for data centers
Bankinter could have detected degradation in minute 2, not 1 hour.
2. Automatic correlation: identify single point of failure
When the app falls And web falls And clients report access blocked simultaneously:
No correlation = ‘3 problems’ = confusion in answer machine
With automatic correlation = ‘communication System central is down’ = clear focus
3. Scaling auto-immediately
In the time that it detects a fall:
- Automatic notification to VPs of infrastructure (no email, SMS + Slack + phone call)
- Team database alerted
- Computer network alerted
- Managers reported
- Crisis Plan activated
Without this, the first engineering manager learns when you see tweets of customers 30 minutes after.
4. Automated communication to customers
- Update status page: ‘to Investigate issues of access’
- Notify via SMS to customers (if integrated): ‘Your bank is experiencing technical problems. ETA: 15 minutes’
- Scale communication every 10 minutes
- Publish ‘Resolution is completed’ when it is normalized
Clients know what happens, do not speculate.
5. Traceability for post-mortem and regulators
For analysis, post-incident and regulatory reporting:
- 09:15 UTC – API latency increases 200%
- 09:16 UTC – Database connection pool exhaustion detected
- 09:17 UTC – Application server responses failing
- 09:18 UTC – Frontend unable to reach the backend
- 09:18 UTC – Incident escalated
- 09:25 UTC – Root cause identified: database connection leak
- 09:35 UTC – Database restarted
- 09:40 UTC – Normal operations resumed
- 10:00 UTC – Post-incident investigation begins
For regulators: ‘we Identified in minute 3, we responded in minute 9, we met in 25 minutes, with full traceability.’
Frequently asked questions
Was it a hacker attack?
Bankinter not indicated. Reported ‘incident technical’, suggesting failure of infrastructure, does not compromise security.
What is lost customer data?
There is No indication. The incident was an accident of availability (access), non-security (theft of data).
What is the fault of Bankinter?
Partially. Falls can happen (hardware fails, software has bugs). The question is: how fast is detected and responded to? I used to have redundancy? How quickly found an alternative?
How do you prevent this 24Cevent?
Does not prevent falls of hardware (out of control). But detected in minutes, not hours, allowing rapid response and transparent communication.
How should I be worried if I use Bankinter?
An incident casual is normal. Worry if: (1) are frequent, (2) last for hours, (3) there is loss of data.