Customer Service During Strikes: Continuity with AI in 2026

The customer service during strikes has become a hot search topic in mid-2026 — and for good reason. While Spain faces the fifth week of doctors' strike against the Estatuto Marco, along with teacher strikes and railway transport strikes, thousands of companies suddenly discover their operations grinding to a halt. In this guide, I show, with real cases, how to keep operations running when a strike interrupts teams, channels, and logistics.

TL;DR

• Strikes give no warning: the doctors' strike in Spain (15-19 June) has already lasted five weeks, and the FGV strike in Valencia transport runs from 22 June to 2 October 2026.
• What breaks first in a strike is not delivery — it's customer service: no one answers clients, orders, or queries.
• WhatsApp automation + AI agents maintain 70-100% of first-level service without relying on those who have downed tools.
• Continuity is not about breaking a strike: it's about shielding the end customer from operational collapse.
• Build the plan before the strike — improvising during it costs dearly.

Why "huelga" and "greve" exploded in searches in 2026

The wave started in Spain and spread to Brazil via social media. The State Confederation of Medical Unions (CESM) called strikes against the reform of the Estatuto Marco, and the fifth week of strike took place between 15 and 19 June 2026, with a central rally in front of the Ministry of Health in Madrid. The agreed minimum services maintain 100% in emergencies but reduce primary care to between 25% and 75%, depending on the autonomous community.

This was not an isolated case. On 9 June, five teachers' unions suspended classes in Catalonia, with a demonstration in Barcelona on the 14th. And the drivers of Ferrocarrils de la Generalitat Valenciana (FGV) went on strike from 22 June, with a calendar extending until 2 October. Health, education, and transport blocked simultaneously — the cascade effect on any business is immediate.

For entrepreneurs, the lesson is borderless. A strike by bank workers, postal workers, public transport, or even an internal strike by your own team produces the same symptom: the customer knocks on the door and finds no one. And those searching for "huelga" or "greve" worried are, in practice, asking the same thing: how not to stop along with them.

Most exposed sectors when people are missing

Not every business feels a strike with the same intensity. The risk of customer service during strikes collapsing is higher in sectors with high contact volume and sensitive deadlines. From my experience serving companies in Brazil and abroad, the most vulnerable are:

• Healthcare and clinics: they depend on schedules. A strike forces mass rescheduling, exactly the bottleneck seen in the Spanish medical strike.
• Education and EAD: suspended classes generate a flood of queries from students and parents.
• E-commerce and logistics: delivery delays trigger a barrage of "where's my order?" messages.
• Scheduled services: salons, workshops, offices — any appointment-based operation.

The pattern is clear: the more a business relies on conversation and deadlines, the sooner it feels the lack of staff. And that is precisely where automation makes the biggest difference.

What really stops when a strike begins

There is a myth that strikes only affect "production." In practice, the first point to collapse is the communication layer. I have seen this for over 15 years serving clients at Agathas Web: when people are missing, the phone rings into the void, WhatsApp accumulates hundreds of unread messages, and the customer, without a reply, goes to the competition.

Typical effects of a strike on operations:

• First-level service disappears: simple queries (hours, order status, duplicate invoices) go unanswered.
• Logistics delays: deliveries dependent on public transport or striking teams grind to a halt.
• Schedules collapse: appointments, meetings, and scheduled services need mass rescheduling.
• Reputation bleeds: negative reviews and complaints on social media spike when silence sets in.

The key point: much of this damage is avoidable, because a large part of the service that stops during a strike is repetitive and automatable. Think about how many of the messages arriving at your company today actually require human judgment. In most businesses I analyse, 60% to 80% of the volume is repeated questions — and it is precisely this portion that sustains customer service during strikes when the team is unavailable.

Customer service during strikes: what can be kept running

Here comes the practical part. Customer service during strikes does not have to be all-or-nothing. The right strategy is to separate what requires a human from what an automation layer can handle alone — and protect the end customer from the impact.

Layer 1 — Automatic replies and triage

An AI agent on WhatsApp resolves, without human intervention, the questions that represent the majority of volume: order status, opening hours, policies, location, duplicate invoices. During a strike, this layer alone already absorbs the peak of messages that would normally overwhelm a reduced team.

Layer 2 — Scheduling and rescheduling

When a strike forces cancellations (think of the medical strike and suspended appointments), an automated flow can offer new slots, confirm rescheduling, and update the customer in real time — without anyone on the phone.

Layer 3 — Intelligent escalation

What the AI cannot resolve, it forwards. Sensitive cases go to the reduced number of available humans, already with the conversation context summarised. Thus, the lean team focuses on what matters instead of drowning in "what are your hours?"

It is exactly this layered logic that I apply in my clients' WhatsApp automation projects. To understand why the right channel matters, it is worth reading WhatsApp Business App vs Official API: which makes sense in 2026.

Manual vs automated: what changes in practice

The difference between facing a strike on the fly and having an automated layer ready appears in operational numbers. The table below summarises the contrast I observe in the field.

Dimension	100% manual operation	Operation with automation + AI
Capacity during strike	Drops with the team	Maintains 70-100% of 1st level
Response time	Hours or days	Seconds, 24/7
Marginal cost per service	High (overtime, on-call)	Near zero
Risk of losing customer	High	Low
Time to activate	Reactive, on the fly	Already live

The message from the table is simple: automation does not replace the team — it ensures that the temporary absence of the team does not bring down the business. A good customer service during strikes plan does not try to clone the human team; it isolates what can run on its own and keeps that viable minimum running while the situation normalises.

Playbook: preparing continuity before the strike

Strikes are rarely a total surprise: unions announce calendars in advance, as CESM did in Spain. Use that time. A customer service during strikes plan built with a buffer costs a fraction of what it costs to put out fires on the day of the strike. Here is the step-by-step I recommend:

1. Map repetitive service. List the 10 most common questions. They are obvious candidates for automation.
2. Choose the right channel. In Brazil, WhatsApp concentrates service. Use the Official API to scale without risk of blocking.
3. Build the AI flows. Configure replies, triage, and scheduling before the critical date.
4. Define escalation. Decide which cases go to a human and who covers the minimum on-call.
5. Test with real volume. Simulate a peak before the strike — discovering a failure during it is the worst scenario.
6. Communicate proactively. Inform customers that service will continue via the automated channel.

When NOT to automate everything

Automation has limits. Serious complaints, contract negotiation, emotional crisis situations, and high-value decisions require a human. Forcing AI in these cases worsens the experience and reputation. The rule: automate volume, preserve the human for what is sensitive. On the limits and the cost of charging per employee, I wrote in Unlimited Agents on WhatsApp: why paying per employee failed.

The role of AI agents in operational continuity

The difference in 2026 from previous years is the maturity of AI agents. We are no longer talking about rigid decision-tree chatbots, but agents that understand natural language, consult data, and make micro-decisions. This changes the continuity game: the agent withstands a strike, a Black Friday peak, or a night shift with the same consistency.

For companies, the gain is strategic. An agent that operates 24/7 does not take holidays, does not get sick, and does not go on strike. It does not eliminate human value — it frees the team for relationships and complex problems. I delved deeper into this transition in AI Agents: what Gemini Spark changes for businesses.

In practice, it is this consistency that makes customer service during strikes reliable. Unlike an improvised on-call, where two or three employees try to handle the volume of an entire team, the agent maintains the same response time in the first minute and the fifth hour of operation. The customer on the other end does not notice that there is a strike happening behind the scenes — and this invisibility is, in the end, the best indicator that the plan worked.

There is also the risk dimension. Concentrating all service on a personal WhatsApp number is fragile: a single block can stop the operation — a self-imposed "strike." Migrating to the official infrastructure is part of the protection, as I detail in How to avoid blocking your business WhatsApp in 2026.

Continuity is not "breaking a strike" — it's protecting the customer

An important ethical distinction is worth making. Automating service during a strike does not mean replacing workers in protest or weakening a legitimate cause. The Spanish doctors' strike, for example, discusses working hours, professional classification, and remuneration — human issues that no software can solve.

What technology does is different: the goal of customer service during strikes is to prevent the end customer — the patient who needs to reschedule, the consumer waiting for an order — from being left in limbo while the labour negotiation runs its course. Operational continuity protects those at the sharp end, without entering into the merits of the conflict. This is the benchmark I advise adopting.

Conclusion: prepare today the plan you will need tomorrow

The 2026 strikes in Spain are a cheap reminder of an expensive lesson: an operation that depends entirely on human presence is fragile. Customer service during strikes — or any unexpected stoppage — only stays up for those who built the automation layer before the crisis.

If your company still serves customers 100% manually, the next step is simple: map the repetitive questions and move them to an AI agent on WhatsApp. Not to fire anyone, but to ensure that no customer is left without a reply on the day the team cannot attend. Want help designing this flow? That is exactly the kind of project we handle at Agathas Web.